{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.9999398785546805, "eval_steps": 500, "global_step": 8316, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "auxiliary_loss_clip": 0.05366347, "auxiliary_loss_mlp": 0.02535328, "balance_loss_clip": 2.34421349, "balance_loss_mlp": 2.15385866, "epoch": 0.00012024289063909097, "flos": 24932487802200.0, "grad_norm": 40.098859089680154, "language_loss": 2.58174515, "learning_rate": 0.0, "loss": 1.90765333, "num_input_tokens_seen": 20375, "step": 1, "time_per_iteration": 13.083091974258423 }, { "auxiliary_loss_clip": 0.03558458, "auxiliary_loss_mlp": 0.01712358, "balance_loss_clip": 1.56047201, "balance_loss_mlp": 1.4642117, "epoch": 0.00024048578127818193, "flos": 30664628792640.0, "grad_norm": 54.97890702529871, "language_loss": 1.88754892, "learning_rate": 5.021476677069823e-07, "loss": 1.94025707, "num_input_tokens_seen": 39035, "step": 2, "time_per_iteration": 2.5865931510925293 }, { "auxiliary_loss_clip": 0.03594151, "auxiliary_loss_mlp": 0.0167374, "balance_loss_clip": 1.57273066, "balance_loss_mlp": 1.43455887, "epoch": 0.0003607286719172729, "flos": 19026231349440.0, "grad_norm": 65.22346371164528, "language_loss": 1.61895967, "learning_rate": 7.958852231401551e-07, "loss": 1.67163861, "num_input_tokens_seen": 57600, "step": 3, "time_per_iteration": 2.4232583045959473 }, { "auxiliary_loss_clip": 0.03558509, "auxiliary_loss_mlp": 0.0175054, "balance_loss_clip": 1.56482279, "balance_loss_mlp": 1.49342895, "epoch": 0.00048097156255636386, "flos": 19316317445760.0, "grad_norm": 39.43348613526535, "language_loss": 1.64347243, "learning_rate": 1.0042953354139647e-06, "loss": 1.69656301, "num_input_tokens_seen": 76465, "step": 4, "time_per_iteration": 2.5108418464660645 }, { "auxiliary_loss_clip": 0.03569447, "auxiliary_loss_mlp": 0.01699445, "balance_loss_clip": 1.56282616, "balance_loss_mlp": 1.45225275, "epoch": 0.0006012144531954548, "flos": 13991267809920.0, "grad_norm": 54.96748318258028, "language_loss": 1.93611419, "learning_rate": 1.1659507774310057e-06, "loss": 1.98880303, "num_input_tokens_seen": 94350, "step": 5, "time_per_iteration": 2.6544857025146484 }, { "auxiliary_loss_clip": 0.03553147, "auxiliary_loss_mlp": 0.01755853, "balance_loss_clip": 1.56032586, "balance_loss_mlp": 1.49530911, "epoch": 0.0007214573438345458, "flos": 23148991979040.0, "grad_norm": 46.38096410444089, "language_loss": 1.61038816, "learning_rate": 1.2980328908471373e-06, "loss": 1.66347814, "num_input_tokens_seen": 114595, "step": 6, "time_per_iteration": 2.742940902709961 }, { "auxiliary_loss_clip": 0.03605901, "auxiliary_loss_mlp": 0.01414012, "balance_loss_clip": 1.69371986, "balance_loss_mlp": 1.29651952, "epoch": 0.0008417002344736367, "flos": 67663257644640.0, "grad_norm": 4.590296888918176, "language_loss": 0.81466365, "learning_rate": 1.4097067265369432e-06, "loss": 0.8648628, "num_input_tokens_seen": 179590, "step": 7, "time_per_iteration": 3.173299789428711 }, { "auxiliary_loss_clip": 0.03563982, "auxiliary_loss_mlp": 0.01701817, "balance_loss_clip": 1.55776894, "balance_loss_mlp": 1.4431808, "epoch": 0.0009619431251127277, "flos": 21281384444640.0, "grad_norm": 41.70079948587207, "language_loss": 1.5916394, "learning_rate": 1.506443003120947e-06, "loss": 1.64429748, "num_input_tokens_seen": 195090, "step": 8, "time_per_iteration": 2.7129979133605957 }, { "auxiliary_loss_clip": 0.03544358, "auxiliary_loss_mlp": 0.01708205, "balance_loss_clip": 1.55650651, "balance_loss_mlp": 1.46024966, "epoch": 0.0010821860157518186, "flos": 23331346360320.0, "grad_norm": 17.914949330413414, "language_loss": 1.47852027, "learning_rate": 1.5917704462803102e-06, "loss": 1.53104591, "num_input_tokens_seen": 211635, "step": 9, "time_per_iteration": 2.71372389793396 }, { "auxiliary_loss_clip": 0.03536068, "auxiliary_loss_mlp": 0.01764022, "balance_loss_clip": 1.57115078, "balance_loss_mlp": 1.51168013, "epoch": 0.0012024289063909096, "flos": 17010168655680.0, "grad_norm": 13.492733316314233, "language_loss": 1.5314889, "learning_rate": 1.6680984451379884e-06, "loss": 1.58448982, "num_input_tokens_seen": 224705, "step": 10, "time_per_iteration": 2.6606972217559814 }, { "auxiliary_loss_clip": 0.03536564, "auxiliary_loss_mlp": 0.01687678, "balance_loss_clip": 1.56465185, "balance_loss_mlp": 1.44105744, "epoch": 0.0013226717970300007, "flos": 21288136833120.0, "grad_norm": 13.49298192833375, "language_loss": 1.32783484, "learning_rate": 1.7371455188905097e-06, "loss": 1.38007712, "num_input_tokens_seen": 244635, "step": 11, "time_per_iteration": 2.7418112754821777 }, { "auxiliary_loss_clip": 0.03529644, "auxiliary_loss_mlp": 0.01697419, "balance_loss_clip": 1.56185186, "balance_loss_mlp": 1.45556688, "epoch": 0.0014429146876690916, "flos": 27237889303680.0, "grad_norm": 10.982228551810161, "language_loss": 1.25381148, "learning_rate": 1.8001805585541196e-06, "loss": 1.30608201, "num_input_tokens_seen": 265765, "step": 12, "time_per_iteration": 2.753885507583618 }, { "auxiliary_loss_clip": 0.03518455, "auxiliary_loss_mlp": 0.01694749, "balance_loss_clip": 1.55840015, "balance_loss_mlp": 1.44088054, "epoch": 0.0015631575783081825, "flos": 19062177838080.0, "grad_norm": 6.855177661308341, "language_loss": 1.29357553, "learning_rate": 1.8581671739548328e-06, "loss": 1.34570742, "num_input_tokens_seen": 283500, "step": 13, "time_per_iteration": 2.7152504920959473 }, { "auxiliary_loss_clip": 0.0353025, "auxiliary_loss_mlp": 0.01643428, "balance_loss_clip": 1.5619055, "balance_loss_mlp": 1.39356577, "epoch": 0.0016834004689472734, "flos": 48139479862560.0, "grad_norm": 6.601997685333307, "language_loss": 1.13540924, "learning_rate": 1.9118543942439254e-06, "loss": 1.18714607, "num_input_tokens_seen": 305685, "step": 14, "time_per_iteration": 4.512828588485718 }, { "auxiliary_loss_clip": 0.03487741, "auxiliary_loss_mlp": 0.01699377, "balance_loss_clip": 1.55513942, "balance_loss_mlp": 1.44398284, "epoch": 0.0018036433595863645, "flos": 34970031139200.0, "grad_norm": 5.712553335207134, "language_loss": 1.12838519, "learning_rate": 1.961836000571161e-06, "loss": 1.18025637, "num_input_tokens_seen": 327340, "step": 15, "time_per_iteration": 3.7123470306396484 }, { "auxiliary_loss_clip": 0.03462552, "auxiliary_loss_mlp": 0.01315466, "balance_loss_clip": 1.67473114, "balance_loss_mlp": 1.20102489, "epoch": 0.0019238862502254555, "flos": 59768293865760.0, "grad_norm": 3.809555946591953, "language_loss": 0.64729548, "learning_rate": 2.0085906708279293e-06, "loss": 0.69507575, "num_input_tokens_seen": 382710, "step": 16, "time_per_iteration": 3.1708130836486816 }, { "auxiliary_loss_clip": 0.0348228, "auxiliary_loss_mlp": 0.01637303, "balance_loss_clip": 1.55179203, "balance_loss_mlp": 1.39716792, "epoch": 0.0020441291408645466, "flos": 20814547012800.0, "grad_norm": 4.298050687990577, "language_loss": 1.16069603, "learning_rate": 2.0525099325728135e-06, "loss": 1.21189165, "num_input_tokens_seen": 400890, "step": 17, "time_per_iteration": 2.7016355991363525 }, { "auxiliary_loss_clip": 0.03411549, "auxiliary_loss_mlp": 0.0129247, "balance_loss_clip": 1.66752577, "balance_loss_mlp": 1.17955494, "epoch": 0.0021643720315036373, "flos": 63857011605600.0, "grad_norm": 3.5268757325230466, "language_loss": 0.72193289, "learning_rate": 2.0939181139872922e-06, "loss": 0.76897305, "num_input_tokens_seen": 462605, "step": 18, "time_per_iteration": 3.1188483238220215 }, { "auxiliary_loss_clip": 0.03484225, "auxiliary_loss_mlp": 0.01648748, "balance_loss_clip": 1.55365145, "balance_loss_mlp": 1.39735913, "epoch": 0.0022846149221427284, "flos": 31284985675200.0, "grad_norm": 5.420198377510579, "language_loss": 1.01521397, "learning_rate": 2.1330868934640175e-06, "loss": 1.06654358, "num_input_tokens_seen": 483280, "step": 19, "time_per_iteration": 2.7777931690216064 }, { "auxiliary_loss_clip": 0.03345712, "auxiliary_loss_mlp": 0.01274692, "balance_loss_clip": 1.65763736, "balance_loss_mlp": 1.1640656, "epoch": 0.002404857812781819, "flos": 51083656679520.0, "grad_norm": 3.5593271403317996, "language_loss": 0.76443136, "learning_rate": 2.170246112844971e-06, "loss": 0.81063545, "num_input_tokens_seen": 537620, "step": 20, "time_per_iteration": 2.9341611862182617 }, { "auxiliary_loss_clip": 0.03386784, "auxiliary_loss_mlp": 0.01607737, "balance_loss_clip": 1.5459491, "balance_loss_mlp": 1.36912823, "epoch": 0.0025251007034209102, "flos": 15815350377120.0, "grad_norm": 4.604985715998187, "language_loss": 1.01529694, "learning_rate": 2.2055919496770983e-06, "loss": 1.06524205, "num_input_tokens_seen": 555760, "step": 21, "time_per_iteration": 2.690427780151367 }, { "auxiliary_loss_clip": 0.03325353, "auxiliary_loss_mlp": 0.01587605, "balance_loss_clip": 1.53079224, "balance_loss_mlp": 1.3476603, "epoch": 0.0026453435940600014, "flos": 37851862501920.0, "grad_norm": 3.8797102234200356, "language_loss": 0.89392138, "learning_rate": 2.2392931865974923e-06, "loss": 0.94305098, "num_input_tokens_seen": 578450, "step": 22, "time_per_iteration": 2.819737434387207 }, { "auxiliary_loss_clip": 0.03314441, "auxiliary_loss_mlp": 0.01489456, "balance_loss_clip": 1.52735257, "balance_loss_mlp": 1.25866652, "epoch": 0.002765586484699092, "flos": 21141980359200.0, "grad_norm": 4.220349920742895, "language_loss": 1.01748157, "learning_rate": 2.271496085962064e-06, "loss": 1.06552052, "num_input_tokens_seen": 596145, "step": 23, "time_per_iteration": 2.678163528442383 }, { "auxiliary_loss_clip": 0.03253617, "auxiliary_loss_mlp": 0.01546516, "balance_loss_clip": 1.51158249, "balance_loss_mlp": 1.30619025, "epoch": 0.002885829375338183, "flos": 20667384864000.0, "grad_norm": 3.7903051570491852, "language_loss": 1.02460945, "learning_rate": 2.3023282262611022e-06, "loss": 1.07261074, "num_input_tokens_seen": 614920, "step": 24, "time_per_iteration": 2.7070088386535645 }, { "auxiliary_loss_clip": 0.03272185, "auxiliary_loss_mlp": 0.0154219, "balance_loss_clip": 1.51860547, "balance_loss_mlp": 1.31292713, "epoch": 0.003006072265977274, "flos": 34823874665280.0, "grad_norm": 3.010344437893407, "language_loss": 0.92515391, "learning_rate": 2.3319015548620114e-06, "loss": 0.97329772, "num_input_tokens_seen": 636060, "step": 25, "time_per_iteration": 2.795827865600586 }, { "auxiliary_loss_clip": 0.03242297, "auxiliary_loss_mlp": 0.01503173, "balance_loss_clip": 1.51051712, "balance_loss_mlp": 1.27200222, "epoch": 0.003126315156616365, "flos": 24422024619840.0, "grad_norm": 2.3345484260764704, "language_loss": 0.92867357, "learning_rate": 2.3603148416618152e-06, "loss": 0.97612822, "num_input_tokens_seen": 655575, "step": 26, "time_per_iteration": 2.7368767261505127 }, { "auxiliary_loss_clip": 0.03201596, "auxiliary_loss_mlp": 0.01444611, "balance_loss_clip": 1.5079143, "balance_loss_mlp": 1.22555208, "epoch": 0.003246558047255456, "flos": 23622330380640.0, "grad_norm": 3.5824496040839895, "language_loss": 1.0100925, "learning_rate": 2.3876556694204647e-06, "loss": 1.05655456, "num_input_tokens_seen": 675730, "step": 27, "time_per_iteration": 2.68611478805542 }, { "auxiliary_loss_clip": 0.03175291, "auxiliary_loss_mlp": 0.01521694, "balance_loss_clip": 1.50540185, "balance_loss_mlp": 1.29185796, "epoch": 0.003366800937894547, "flos": 17820278813280.0, "grad_norm": 4.34760700452351, "language_loss": 0.90589428, "learning_rate": 2.414002061950908e-06, "loss": 0.95286405, "num_input_tokens_seen": 694605, "step": 28, "time_per_iteration": 2.690016984939575 }, { "auxiliary_loss_clip": 0.03139451, "auxiliary_loss_mlp": 0.01461017, "balance_loss_clip": 1.49564314, "balance_loss_mlp": 1.23633099, "epoch": 0.003487043828533638, "flos": 24426119153280.0, "grad_norm": 2.232278441555184, "language_loss": 0.9985382, "learning_rate": 2.4394238264681557e-06, "loss": 1.04454291, "num_input_tokens_seen": 714340, "step": 29, "time_per_iteration": 2.710050582885742 }, { "auxiliary_loss_clip": 0.03114549, "auxiliary_loss_mlp": 0.01453349, "balance_loss_clip": 1.49432361, "balance_loss_mlp": 1.23705566, "epoch": 0.003607286719172729, "flos": 26140314987840.0, "grad_norm": 2.19296592682794, "language_loss": 0.9947325, "learning_rate": 2.4639836682781433e-06, "loss": 1.04041147, "num_input_tokens_seen": 734470, "step": 30, "time_per_iteration": 2.7162296772003174 }, { "auxiliary_loss_clip": 0.03107418, "auxiliary_loss_mlp": 0.01432281, "balance_loss_clip": 1.49296546, "balance_loss_mlp": 1.22609639, "epoch": 0.00372752960981182, "flos": 20593085450400.0, "grad_norm": 2.4569886854128185, "language_loss": 1.00138164, "learning_rate": 2.487738122623307e-06, "loss": 1.04677868, "num_input_tokens_seen": 753380, "step": 31, "time_per_iteration": 2.669455051422119 }, { "auxiliary_loss_clip": 0.03037367, "auxiliary_loss_mlp": 0.01438222, "balance_loss_clip": 1.47547078, "balance_loss_mlp": 1.23156118, "epoch": 0.003847772500450911, "flos": 22674611985600.0, "grad_norm": 3.1544509912421996, "language_loss": 0.99146742, "learning_rate": 2.510738338534912e-06, "loss": 1.03622317, "num_input_tokens_seen": 772105, "step": 32, "time_per_iteration": 2.710158586502075 }, { "auxiliary_loss_clip": 0.02942763, "auxiliary_loss_mlp": 0.0140643, "balance_loss_clip": 1.45475078, "balance_loss_mlp": 1.20196199, "epoch": 0.003968015391090002, "flos": 17967800131680.0, "grad_norm": 2.519556286866108, "language_loss": 1.02635252, "learning_rate": 2.5330307420306648e-06, "loss": 1.06984448, "num_input_tokens_seen": 788955, "step": 33, "time_per_iteration": 2.665346145629883 }, { "auxiliary_loss_clip": 0.02899459, "auxiliary_loss_mlp": 0.01398844, "balance_loss_clip": 1.4410882, "balance_loss_mlp": 1.1888454, "epoch": 0.004088258281729093, "flos": 27304107401280.0, "grad_norm": 2.6076670300138263, "language_loss": 0.88098711, "learning_rate": 2.554657600279796e-06, "loss": 0.9239701, "num_input_tokens_seen": 810230, "step": 34, "time_per_iteration": 2.72796893119812 }, { "auxiliary_loss_clip": 0.02819996, "auxiliary_loss_mlp": 0.01369296, "balance_loss_clip": 1.4372021, "balance_loss_mlp": 1.16931069, "epoch": 0.004208501172368184, "flos": 23258591376000.0, "grad_norm": 2.479863222682673, "language_loss": 1.0338285, "learning_rate": 2.5756575039679493e-06, "loss": 1.0757215, "num_input_tokens_seen": 829780, "step": 35, "time_per_iteration": 2.722639322280884 }, { "auxiliary_loss_clip": 0.02811778, "auxiliary_loss_mlp": 0.01354314, "balance_loss_clip": 1.41952848, "balance_loss_mlp": 1.15575957, "epoch": 0.0043287440630072746, "flos": 17312107348800.0, "grad_norm": 2.04669398841038, "language_loss": 0.95167983, "learning_rate": 2.5960657816942747e-06, "loss": 0.99334073, "num_input_tokens_seen": 848695, "step": 36, "time_per_iteration": 2.6462478637695312 }, { "auxiliary_loss_clip": 0.02478329, "auxiliary_loss_mlp": 0.01259316, "balance_loss_clip": 1.4756006, "balance_loss_mlp": 1.168908, "epoch": 0.004448986953646365, "flos": 53092500064320.0, "grad_norm": 1.3889127892782689, "language_loss": 0.60963476, "learning_rate": 2.6159148575788668e-06, "loss": 0.64701122, "num_input_tokens_seen": 906730, "step": 37, "time_per_iteration": 3.176968574523926 }, { "auxiliary_loss_clip": 0.02727113, "auxiliary_loss_mlp": 0.01343921, "balance_loss_clip": 1.41289365, "balance_loss_mlp": 1.15299499, "epoch": 0.004569229844285457, "flos": 13444168749120.0, "grad_norm": 2.3829251505271287, "language_loss": 0.9865604, "learning_rate": 2.635234561171e-06, "loss": 1.02727079, "num_input_tokens_seen": 925125, "step": 38, "time_per_iteration": 2.6606338024139404 }, { "auxiliary_loss_clip": 0.02692852, "auxiliary_loss_mlp": 0.01306705, "balance_loss_clip": 1.40040433, "balance_loss_mlp": 1.11749625, "epoch": 0.0046894727349245475, "flos": 16209612409440.0, "grad_norm": 2.1952569337402457, "language_loss": 0.94142985, "learning_rate": 2.6540523970949877e-06, "loss": 0.98142529, "num_input_tokens_seen": 939970, "step": 39, "time_per_iteration": 2.66924786567688 }, { "auxiliary_loss_clip": 0.02666623, "auxiliary_loss_mlp": 0.01318671, "balance_loss_clip": 1.39838469, "balance_loss_mlp": 1.13690019, "epoch": 0.004809715625563638, "flos": 23914248241920.0, "grad_norm": 3.835967081507455, "language_loss": 0.92552972, "learning_rate": 2.6723937805519533e-06, "loss": 0.9653827, "num_input_tokens_seen": 957470, "step": 40, "time_per_iteration": 3.609064817428589 }, { "auxiliary_loss_clip": 0.02634577, "auxiliary_loss_mlp": 0.01286682, "balance_loss_clip": 1.38894808, "balance_loss_mlp": 1.1038624, "epoch": 0.00492995851620273, "flos": 20773033395360.0, "grad_norm": 2.5776196027512825, "language_loss": 0.93008697, "learning_rate": 2.690282243737839e-06, "loss": 0.96929955, "num_input_tokens_seen": 976405, "step": 41, "time_per_iteration": 3.5503299236297607 }, { "auxiliary_loss_clip": 0.02590757, "auxiliary_loss_mlp": 0.01289986, "balance_loss_clip": 1.37719107, "balance_loss_mlp": 1.11088645, "epoch": 0.0050502014068418205, "flos": 20338658507040.0, "grad_norm": 3.9625066678787624, "language_loss": 0.993119, "learning_rate": 2.7077396173840807e-06, "loss": 1.03192651, "num_input_tokens_seen": 994690, "step": 42, "time_per_iteration": 2.699166774749756 }, { "auxiliary_loss_clip": 0.02570738, "auxiliary_loss_mlp": 0.01286706, "balance_loss_clip": 1.38066721, "balance_loss_mlp": 1.10789192, "epoch": 0.005170444297480911, "flos": 25994876853120.0, "grad_norm": 2.5228681956572143, "language_loss": 0.92702579, "learning_rate": 2.7247861909342594e-06, "loss": 0.96560019, "num_input_tokens_seen": 1015615, "step": 43, "time_per_iteration": 2.717562437057495 }, { "auxiliary_loss_clip": 0.02532331, "auxiliary_loss_mlp": 0.01268905, "balance_loss_clip": 1.37174964, "balance_loss_mlp": 1.09791136, "epoch": 0.005290687188120003, "flos": 20954058849120.0, "grad_norm": 4.1549540396179445, "language_loss": 0.83082914, "learning_rate": 2.7414408543044743e-06, "loss": 0.86884153, "num_input_tokens_seen": 1031255, "step": 44, "time_per_iteration": 2.6934540271759033 }, { "auxiliary_loss_clip": 0.02510993, "auxiliary_loss_mlp": 0.01248949, "balance_loss_clip": 1.36581898, "balance_loss_mlp": 1.08501196, "epoch": 0.005410930078759093, "flos": 15851404616640.0, "grad_norm": 5.619725833133144, "language_loss": 0.79310387, "learning_rate": 2.7577212237113157e-06, "loss": 0.83070332, "num_input_tokens_seen": 1048295, "step": 45, "time_per_iteration": 2.651435613632202 }, { "auxiliary_loss_clip": 0.0248566, "auxiliary_loss_mlp": 0.01272728, "balance_loss_clip": 1.36595154, "balance_loss_mlp": 1.10421383, "epoch": 0.005531172969398184, "flos": 21104992278720.0, "grad_norm": 6.442061881449115, "language_loss": 1.04297948, "learning_rate": 2.7736437536690466e-06, "loss": 1.08056331, "num_input_tokens_seen": 1067925, "step": 46, "time_per_iteration": 2.6891989707946777 }, { "auxiliary_loss_clip": 0.02466881, "auxiliary_loss_mlp": 0.0124033, "balance_loss_clip": 1.36438394, "balance_loss_mlp": 1.07534432, "epoch": 0.005651415860037276, "flos": 20844890455680.0, "grad_norm": 2.2526715773569177, "language_loss": 1.080037, "learning_rate": 2.789223836941131e-06, "loss": 1.11710906, "num_input_tokens_seen": 1088060, "step": 47, "time_per_iteration": 2.678799867630005 }, { "auxiliary_loss_clip": 0.02438118, "auxiliary_loss_mlp": 0.01286597, "balance_loss_clip": 1.35410988, "balance_loss_mlp": 1.11760592, "epoch": 0.005771658750676366, "flos": 13260198104640.0, "grad_norm": 2.252035183305262, "language_loss": 1.09003329, "learning_rate": 2.8044758939680847e-06, "loss": 1.12728035, "num_input_tokens_seen": 1104130, "step": 48, "time_per_iteration": 2.6509101390838623 }, { "auxiliary_loss_clip": 0.02402289, "auxiliary_loss_mlp": 0.01262858, "balance_loss_clip": 1.3456037, "balance_loss_mlp": 1.10473859, "epoch": 0.005891901641315457, "flos": 24425400814080.0, "grad_norm": 5.044497445159399, "language_loss": 1.01968551, "learning_rate": 2.8194134530738863e-06, "loss": 1.056337, "num_input_tokens_seen": 1122900, "step": 49, "time_per_iteration": 2.7348599433898926 }, { "auxiliary_loss_clip": 0.02391707, "auxiliary_loss_mlp": 0.01272555, "balance_loss_clip": 1.34587097, "balance_loss_mlp": 1.11281443, "epoch": 0.006012144531954548, "flos": 23076201077760.0, "grad_norm": 2.572127306849132, "language_loss": 0.90170217, "learning_rate": 2.834049222568994e-06, "loss": 0.93834472, "num_input_tokens_seen": 1140250, "step": 50, "time_per_iteration": 2.693673849105835 }, { "auxiliary_loss_clip": 0.02385404, "auxiliary_loss_mlp": 0.01222612, "balance_loss_clip": 1.34710908, "balance_loss_mlp": 1.06830776, "epoch": 0.006132387422593639, "flos": 22528778764320.0, "grad_norm": 1.9907008998522286, "language_loss": 0.92584467, "learning_rate": 2.848395155712969e-06, "loss": 0.96192485, "num_input_tokens_seen": 1160470, "step": 51, "time_per_iteration": 2.679353713989258 }, { "auxiliary_loss_clip": 0.02339271, "auxiliary_loss_mlp": 0.01229237, "balance_loss_clip": 1.33981216, "balance_loss_mlp": 1.08857012, "epoch": 0.00625263031323273, "flos": 27628344138240.0, "grad_norm": 2.166138868895258, "language_loss": 0.97727555, "learning_rate": 2.8624625093687977e-06, "loss": 1.01296055, "num_input_tokens_seen": 1177605, "step": 52, "time_per_iteration": 2.696199417114258 }, { "auxiliary_loss_clip": 0.02336223, "auxiliary_loss_mlp": 0.01230812, "balance_loss_clip": 1.34042835, "balance_loss_mlp": 1.08165765, "epoch": 0.006372873203871821, "flos": 23110674971040.0, "grad_norm": 1.9958918705664328, "language_loss": 0.88983476, "learning_rate": 2.876261897070029e-06, "loss": 0.9255051, "num_input_tokens_seen": 1197735, "step": 53, "time_per_iteration": 2.70637583732605 }, { "auxiliary_loss_clip": 0.0231898, "auxiliary_loss_mlp": 0.01215762, "balance_loss_clip": 1.33194315, "balance_loss_mlp": 1.07242501, "epoch": 0.006493116094510912, "flos": 22856032525920.0, "grad_norm": 2.299300123708419, "language_loss": 0.92515004, "learning_rate": 2.889803337127447e-06, "loss": 0.9604975, "num_input_tokens_seen": 1216335, "step": 54, "time_per_iteration": 2.6844124794006348 }, { "auxiliary_loss_clip": 0.02298053, "auxiliary_loss_mlp": 0.01230577, "balance_loss_clip": 1.32839811, "balance_loss_mlp": 1.090482, "epoch": 0.006613358985150003, "flos": 23071711457760.0, "grad_norm": 2.5616623194392263, "language_loss": 0.84698945, "learning_rate": 2.903096296321516e-06, "loss": 0.88227582, "num_input_tokens_seen": 1234480, "step": 55, "time_per_iteration": 2.731206178665161 }, { "auxiliary_loss_clip": 0.02275564, "auxiliary_loss_mlp": 0.01219784, "balance_loss_clip": 1.32631147, "balance_loss_mlp": 1.08045244, "epoch": 0.006733601875789094, "flos": 26537665878720.0, "grad_norm": 2.1066879057362216, "language_loss": 0.91594416, "learning_rate": 2.9161497296578907e-06, "loss": 0.95089763, "num_input_tokens_seen": 1253870, "step": 56, "time_per_iteration": 2.6958069801330566 }, { "auxiliary_loss_clip": 0.02278481, "auxiliary_loss_mlp": 0.01190699, "balance_loss_clip": 1.33325005, "balance_loss_mlp": 1.06319261, "epoch": 0.006853844766428185, "flos": 15523181097120.0, "grad_norm": 2.3742862840669225, "language_loss": 0.86123461, "learning_rate": 2.928972116604173e-06, "loss": 0.89592642, "num_input_tokens_seen": 1270145, "step": 57, "time_per_iteration": 2.683048725128174 }, { "auxiliary_loss_clip": 0.02228923, "auxiliary_loss_mlp": 0.0121768, "balance_loss_clip": 1.31953251, "balance_loss_mlp": 1.08140004, "epoch": 0.006974087657067276, "flos": 24243764772000.0, "grad_norm": 1.9400776511118876, "language_loss": 1.02201557, "learning_rate": 2.9415714941751377e-06, "loss": 1.0564816, "num_input_tokens_seen": 1291365, "step": 58, "time_per_iteration": 2.68156099319458 }, { "auxiliary_loss_clip": 0.02235438, "auxiliary_loss_mlp": 0.01231648, "balance_loss_clip": 1.32006037, "balance_loss_mlp": 1.09288883, "epoch": 0.007094330547706367, "flos": 25772517366720.0, "grad_norm": 2.722420998725607, "language_loss": 0.93581241, "learning_rate": 2.9539554871897396e-06, "loss": 0.9704833, "num_input_tokens_seen": 1311535, "step": 59, "time_per_iteration": 2.708895206451416 }, { "auxiliary_loss_clip": 0.02180154, "auxiliary_loss_mlp": 0.01202367, "balance_loss_clip": 1.31086612, "balance_loss_mlp": 1.07953358, "epoch": 0.007214573438345458, "flos": 21319019030400.0, "grad_norm": 1.9913102482844265, "language_loss": 0.97317666, "learning_rate": 2.9661313359851253e-06, "loss": 1.00700188, "num_input_tokens_seen": 1329420, "step": 60, "time_per_iteration": 2.6815218925476074 }, { "auxiliary_loss_clip": 0.02173411, "auxiliary_loss_mlp": 0.0119785, "balance_loss_clip": 1.30972242, "balance_loss_mlp": 1.07244158, "epoch": 0.007334816328984549, "flos": 24937092140640.0, "grad_norm": 2.652612925886084, "language_loss": 0.9391892, "learning_rate": 2.978105921839922e-06, "loss": 0.97290176, "num_input_tokens_seen": 1349965, "step": 61, "time_per_iteration": 2.7057220935821533 }, { "auxiliary_loss_clip": 0.02149485, "auxiliary_loss_mlp": 0.01199616, "balance_loss_clip": 1.30637562, "balance_loss_mlp": 1.0778321, "epoch": 0.00745505921962364, "flos": 18510589157280.0, "grad_norm": 2.4098231852882237, "language_loss": 0.72134167, "learning_rate": 2.9898857903302893e-06, "loss": 0.75483274, "num_input_tokens_seen": 1368915, "step": 62, "time_per_iteration": 2.6989035606384277 }, { "auxiliary_loss_clip": 0.02142628, "auxiliary_loss_mlp": 0.01203285, "balance_loss_clip": 1.30360615, "balance_loss_mlp": 1.08646011, "epoch": 0.007575302110262731, "flos": 18477659693280.0, "grad_norm": 3.6102950616141114, "language_loss": 0.87794864, "learning_rate": 3.001477172817253e-06, "loss": 0.91140783, "num_input_tokens_seen": 1386805, "step": 63, "time_per_iteration": 2.636389970779419 }, { "auxiliary_loss_clip": 0.02105454, "auxiliary_loss_mlp": 0.01198982, "balance_loss_clip": 1.29610646, "balance_loss_mlp": 1.08091736, "epoch": 0.007695545000901822, "flos": 24973182297120.0, "grad_norm": 3.051908810975808, "language_loss": 0.9646731, "learning_rate": 3.012886006241894e-06, "loss": 0.9977175, "num_input_tokens_seen": 1406190, "step": 64, "time_per_iteration": 2.7109620571136475 }, { "auxiliary_loss_clip": 0.02109242, "auxiliary_loss_mlp": 0.01205831, "balance_loss_clip": 1.29311061, "balance_loss_mlp": 1.09034073, "epoch": 0.007815787891540913, "flos": 21324226989600.0, "grad_norm": 4.437579562814786, "language_loss": 0.8820926, "learning_rate": 3.0241179513858383e-06, "loss": 0.91524339, "num_input_tokens_seen": 1425500, "step": 65, "time_per_iteration": 2.6591217517852783 }, { "auxiliary_loss_clip": 0.02080877, "auxiliary_loss_mlp": 0.01188431, "balance_loss_clip": 1.28526664, "balance_loss_mlp": 1.0764693, "epoch": 0.007936030782180003, "flos": 21575780576160.0, "grad_norm": 2.165781495442745, "language_loss": 0.87599897, "learning_rate": 3.035178409737647e-06, "loss": 0.908692, "num_input_tokens_seen": 1442950, "step": 66, "time_per_iteration": 4.34979772567749 }, { "auxiliary_loss_clip": 0.02061051, "auxiliary_loss_mlp": 0.0120102, "balance_loss_clip": 1.28071761, "balance_loss_mlp": 1.08734226, "epoch": 0.008056273672819095, "flos": 20120788640640.0, "grad_norm": 2.4405820584900284, "language_loss": 0.88599026, "learning_rate": 3.046072539090907e-06, "loss": 0.91861093, "num_input_tokens_seen": 1460915, "step": 67, "time_per_iteration": 3.551788330078125 }, { "auxiliary_loss_clip": 0.02036092, "auxiliary_loss_mlp": 0.01191328, "balance_loss_clip": 1.27873862, "balance_loss_mlp": 1.08127427, "epoch": 0.008176516563458186, "flos": 18333119482560.0, "grad_norm": 4.011463560840345, "language_loss": 1.04631495, "learning_rate": 3.056805267986779e-06, "loss": 1.0785892, "num_input_tokens_seen": 1478385, "step": 68, "time_per_iteration": 3.390784978866577 }, { "auxiliary_loss_clip": 0.02007736, "auxiliary_loss_mlp": 0.0115734, "balance_loss_clip": 1.26939297, "balance_loss_mlp": 1.05510604, "epoch": 0.008296759454097276, "flos": 21872080306560.0, "grad_norm": 2.567302469928703, "language_loss": 0.95492846, "learning_rate": 3.0673813091022194e-06, "loss": 0.98657924, "num_input_tokens_seen": 1497605, "step": 69, "time_per_iteration": 2.661527633666992 }, { "auxiliary_loss_clip": 0.01841704, "auxiliary_loss_mlp": 0.01182443, "balance_loss_clip": 1.34406972, "balance_loss_mlp": 1.14486873, "epoch": 0.008417002344736368, "flos": 63408240404640.0, "grad_norm": 1.3423526011696276, "language_loss": 0.62037665, "learning_rate": 3.0778051716749317e-06, "loss": 0.65061814, "num_input_tokens_seen": 1561150, "step": 70, "time_per_iteration": 3.27614688873291 }, { "auxiliary_loss_clip": 0.01990752, "auxiliary_loss_mlp": 0.01170017, "balance_loss_clip": 1.26066661, "balance_loss_mlp": 1.07317138, "epoch": 0.008537245235375458, "flos": 22966458012960.0, "grad_norm": 2.1504907527780577, "language_loss": 0.90387595, "learning_rate": 3.0880811730470094e-06, "loss": 0.93548369, "num_input_tokens_seen": 1580605, "step": 71, "time_per_iteration": 2.656669855117798 }, { "auxiliary_loss_clip": 0.01807216, "auxiliary_loss_mlp": 0.01119371, "balance_loss_clip": 1.32971454, "balance_loss_mlp": 1.08542001, "epoch": 0.008657488126014549, "flos": 61984058832480.0, "grad_norm": 1.1735202189707563, "language_loss": 0.58581877, "learning_rate": 3.098213449401257e-06, "loss": 0.61508459, "num_input_tokens_seen": 1647535, "step": 72, "time_per_iteration": 3.164236307144165 }, { "auxiliary_loss_clip": 0.01953832, "auxiliary_loss_mlp": 0.01162819, "balance_loss_clip": 1.25225449, "balance_loss_mlp": 1.06563985, "epoch": 0.00877773101665364, "flos": 30296795254560.0, "grad_norm": 2.2499362469361994, "language_loss": 0.9888671, "learning_rate": 3.1082059657570015e-06, "loss": 1.0200336, "num_input_tokens_seen": 1666770, "step": 73, "time_per_iteration": 2.7026522159576416 }, { "auxiliary_loss_clip": 0.01929949, "auxiliary_loss_mlp": 0.01166908, "balance_loss_clip": 1.25266719, "balance_loss_mlp": 1.06972814, "epoch": 0.00889797390729273, "flos": 23514059911200.0, "grad_norm": 2.476860791880842, "language_loss": 0.96730208, "learning_rate": 3.1180625252858496e-06, "loss": 0.99827063, "num_input_tokens_seen": 1685200, "step": 74, "time_per_iteration": 2.6883139610290527 }, { "auxiliary_loss_clip": 0.01915439, "auxiliary_loss_mlp": 0.01157771, "balance_loss_clip": 1.24206018, "balance_loss_mlp": 1.06807828, "epoch": 0.009018216797931822, "flos": 23075841908160.0, "grad_norm": 3.5232416188494957, "language_loss": 0.80040109, "learning_rate": 3.1277867780021663e-06, "loss": 0.83113319, "num_input_tokens_seen": 1701835, "step": 75, "time_per_iteration": 2.6442313194274902 }, { "auxiliary_loss_clip": 0.01882028, "auxiliary_loss_mlp": 0.01159309, "balance_loss_clip": 1.23059273, "balance_loss_mlp": 1.06370342, "epoch": 0.009138459688570914, "flos": 15918879807840.0, "grad_norm": 2.0571217106610105, "language_loss": 0.95446396, "learning_rate": 3.1373822288779824e-06, "loss": 0.98487723, "num_input_tokens_seen": 1718415, "step": 76, "time_per_iteration": 2.653449535369873 }, { "auxiliary_loss_clip": 0.01894908, "auxiliary_loss_mlp": 0.01181711, "balance_loss_clip": 1.23773289, "balance_loss_mlp": 1.09297132, "epoch": 0.009258702579210003, "flos": 27016535492160.0, "grad_norm": 2.3671834019851437, "language_loss": 0.79502517, "learning_rate": 3.1468522454274533e-06, "loss": 0.82579136, "num_input_tokens_seen": 1738770, "step": 77, "time_per_iteration": 2.6917288303375244 }, { "auxiliary_loss_clip": 0.01872165, "auxiliary_loss_mlp": 0.01150328, "balance_loss_clip": 1.22830045, "balance_loss_mlp": 1.06611919, "epoch": 0.009378945469849095, "flos": 26903200731360.0, "grad_norm": 1.9368022998262933, "language_loss": 0.91841793, "learning_rate": 3.15620006480197e-06, "loss": 0.94864285, "num_input_tokens_seen": 1758040, "step": 78, "time_per_iteration": 2.7052900791168213 }, { "auxiliary_loss_clip": 0.01856329, "auxiliary_loss_mlp": 0.01163897, "balance_loss_clip": 1.22166252, "balance_loss_mlp": 1.07835233, "epoch": 0.009499188360488187, "flos": 35694240705120.0, "grad_norm": 2.8313376307130804, "language_loss": 0.75113797, "learning_rate": 3.1654288004333087e-06, "loss": 0.78134024, "num_input_tokens_seen": 1776705, "step": 79, "time_per_iteration": 2.7718448638916016 }, { "auxiliary_loss_clip": 0.01835855, "auxiliary_loss_mlp": 0.01138479, "balance_loss_clip": 1.22079754, "balance_loss_mlp": 1.05460346, "epoch": 0.009619431251127276, "flos": 21503205176640.0, "grad_norm": 2.7703778500691585, "language_loss": 0.76189244, "learning_rate": 3.1745414482589353e-06, "loss": 0.79163575, "num_input_tokens_seen": 1795915, "step": 80, "time_per_iteration": 2.645050048828125 }, { "auxiliary_loss_clip": 0.0181887, "auxiliary_loss_mlp": 0.01134241, "balance_loss_clip": 1.21566963, "balance_loss_mlp": 1.05694532, "epoch": 0.009739674141766368, "flos": 17421060240480.0, "grad_norm": 2.5759852955879396, "language_loss": 0.87234032, "learning_rate": 3.1835408925606204e-06, "loss": 0.90187144, "num_input_tokens_seen": 1814055, "step": 81, "time_per_iteration": 2.6513705253601074 }, { "auxiliary_loss_clip": 0.01801404, "auxiliary_loss_mlp": 0.01160244, "balance_loss_clip": 1.21344745, "balance_loss_mlp": 1.07832384, "epoch": 0.00985991703240546, "flos": 27527113392960.0, "grad_norm": 2.3536812297115177, "language_loss": 0.89283246, "learning_rate": 3.1924299114448214e-06, "loss": 0.92244893, "num_input_tokens_seen": 1834535, "step": 82, "time_per_iteration": 2.7009735107421875 }, { "auxiliary_loss_clip": 0.01811562, "auxiliary_loss_mlp": 0.01152519, "balance_loss_clip": 1.21666336, "balance_loss_mlp": 1.07074177, "epoch": 0.00998015992304455, "flos": 13808087338560.0, "grad_norm": 2.4584719866246476, "language_loss": 0.83451748, "learning_rate": 3.2012111819909055e-06, "loss": 0.86415827, "num_input_tokens_seen": 1851865, "step": 83, "time_per_iteration": 2.6715939044952393 }, { "auxiliary_loss_clip": 0.01789468, "auxiliary_loss_mlp": 0.01138244, "balance_loss_clip": 1.20638633, "balance_loss_mlp": 1.0605669, "epoch": 0.010100402813683641, "flos": 20191388607360.0, "grad_norm": 9.079653495286319, "language_loss": 0.95158023, "learning_rate": 3.2098872850910627e-06, "loss": 0.98085737, "num_input_tokens_seen": 1868540, "step": 84, "time_per_iteration": 2.588625192642212 }, { "auxiliary_loss_clip": 0.01791468, "auxiliary_loss_mlp": 0.0113814, "balance_loss_clip": 1.20931661, "balance_loss_mlp": 1.0638963, "epoch": 0.010220645704322733, "flos": 17201645944800.0, "grad_norm": 2.0313959344363517, "language_loss": 0.89356893, "learning_rate": 3.2184607100038194e-06, "loss": 0.92286503, "num_input_tokens_seen": 1887180, "step": 85, "time_per_iteration": 2.6234209537506104 }, { "auxiliary_loss_clip": 0.01773606, "auxiliary_loss_mlp": 0.01119808, "balance_loss_clip": 1.20665693, "balance_loss_mlp": 1.05510139, "epoch": 0.010340888594961822, "flos": 21470419380480.0, "grad_norm": 2.1656162313873546, "language_loss": 0.93394411, "learning_rate": 3.2269338586412414e-06, "loss": 0.96287823, "num_input_tokens_seen": 1904765, "step": 86, "time_per_iteration": 2.6069495677948 }, { "auxiliary_loss_clip": 0.01753382, "auxiliary_loss_mlp": 0.01150414, "balance_loss_clip": 1.19696033, "balance_loss_mlp": 1.07769668, "epoch": 0.010461131485600914, "flos": 23002835505120.0, "grad_norm": 2.7707109598783295, "language_loss": 0.96472156, "learning_rate": 3.2353090496083106e-06, "loss": 0.99375939, "num_input_tokens_seen": 1922600, "step": 87, "time_per_iteration": 2.6807236671447754 }, { "auxiliary_loss_clip": 0.0174177, "auxiliary_loss_mlp": 0.01154358, "balance_loss_clip": 1.19476891, "balance_loss_mlp": 1.08164048, "epoch": 0.010581374376240005, "flos": 33546855241920.0, "grad_norm": 2.211027398841219, "language_loss": 0.81262159, "learning_rate": 3.2435885220114572e-06, "loss": 0.84158289, "num_input_tokens_seen": 1943950, "step": 88, "time_per_iteration": 2.784383535385132 }, { "auxiliary_loss_clip": 0.0175279, "auxiliary_loss_mlp": 0.01137144, "balance_loss_clip": 1.2017889, "balance_loss_mlp": 1.07048225, "epoch": 0.010701617266879095, "flos": 21763091497920.0, "grad_norm": 4.074474691723179, "language_loss": 0.93988842, "learning_rate": 3.2517744390519113e-06, "loss": 0.96878779, "num_input_tokens_seen": 1962815, "step": 89, "time_per_iteration": 2.651416063308716 }, { "auxiliary_loss_clip": 0.01727655, "auxiliary_loss_mlp": 0.0114025, "balance_loss_clip": 1.19078207, "balance_loss_mlp": 1.07396936, "epoch": 0.010821860157518187, "flos": 19060202405280.0, "grad_norm": 1.9598565377996118, "language_loss": 0.75102001, "learning_rate": 3.259868891418298e-06, "loss": 0.77969909, "num_input_tokens_seen": 1980580, "step": 90, "time_per_iteration": 2.6395556926727295 }, { "auxiliary_loss_clip": 0.01730335, "auxiliary_loss_mlp": 0.01123485, "balance_loss_clip": 1.19771624, "balance_loss_mlp": 1.05873013, "epoch": 0.010942103048157278, "flos": 25447382705760.0, "grad_norm": 2.0263056858444983, "language_loss": 0.85004663, "learning_rate": 3.2678739004917757e-06, "loss": 0.87858486, "num_input_tokens_seen": 2000315, "step": 91, "time_per_iteration": 2.690850019454956 }, { "auxiliary_loss_clip": 0.01715248, "auxiliary_loss_mlp": 0.01147162, "balance_loss_clip": 1.19420815, "balance_loss_mlp": 1.08503032, "epoch": 0.011062345938796368, "flos": 27493932510240.0, "grad_norm": 1.653953630216468, "language_loss": 0.92209661, "learning_rate": 3.275791421376029e-06, "loss": 0.95072067, "num_input_tokens_seen": 2023760, "step": 92, "time_per_iteration": 4.344743013381958 }, { "auxiliary_loss_clip": 0.01707741, "auxiliary_loss_mlp": 0.01132949, "balance_loss_clip": 1.18497074, "balance_loss_mlp": 1.06895733, "epoch": 0.01118258882943546, "flos": 16071213998880.0, "grad_norm": 2.390334756635815, "language_loss": 0.96134782, "learning_rate": 3.2836233457634622e-06, "loss": 0.98975468, "num_input_tokens_seen": 2041895, "step": 93, "time_per_iteration": 2.6108617782592773 }, { "auxiliary_loss_clip": 0.01705026, "auxiliary_loss_mlp": 0.0112772, "balance_loss_clip": 1.18206811, "balance_loss_mlp": 1.06415713, "epoch": 0.011302831720074551, "flos": 20668606040640.0, "grad_norm": 2.22627817711395, "language_loss": 0.85574687, "learning_rate": 3.2913715046481135e-06, "loss": 0.88407433, "num_input_tokens_seen": 2061640, "step": 94, "time_per_iteration": 3.528721332550049 }, { "auxiliary_loss_clip": 0.01696784, "auxiliary_loss_mlp": 0.01130288, "balance_loss_clip": 1.18344986, "balance_loss_mlp": 1.06748843, "epoch": 0.011423074610713641, "flos": 13072240677600.0, "grad_norm": 2.146880925410544, "language_loss": 0.8892684, "learning_rate": 3.299037670895023e-06, "loss": 0.91753918, "num_input_tokens_seen": 2078255, "step": 95, "time_per_iteration": 3.3987655639648438 }, { "auxiliary_loss_clip": 0.01694229, "auxiliary_loss_mlp": 0.01110691, "balance_loss_clip": 1.18418062, "balance_loss_mlp": 1.05502057, "epoch": 0.011543317501352733, "flos": 30335651016960.0, "grad_norm": 1.9283018255013737, "language_loss": 0.8054961, "learning_rate": 3.3066235616750667e-06, "loss": 0.83354533, "num_input_tokens_seen": 2099490, "step": 96, "time_per_iteration": 2.697235345840454 }, { "auxiliary_loss_clip": 0.01672277, "auxiliary_loss_mlp": 0.01134484, "balance_loss_clip": 1.17801452, "balance_loss_mlp": 1.07673895, "epoch": 0.011663560391991824, "flos": 15522965595360.0, "grad_norm": 2.3252619650166033, "language_loss": 0.92408645, "learning_rate": 3.3141308407736276e-06, "loss": 0.9521541, "num_input_tokens_seen": 2116125, "step": 97, "time_per_iteration": 2.6518750190734863 }, { "auxiliary_loss_clip": 0.01675285, "auxiliary_loss_mlp": 0.01111693, "balance_loss_clip": 1.17569566, "balance_loss_mlp": 1.05711913, "epoch": 0.011783803282630914, "flos": 19902128601120.0, "grad_norm": 1.9222841612518256, "language_loss": 0.86692083, "learning_rate": 3.321561120780869e-06, "loss": 0.89479059, "num_input_tokens_seen": 2134835, "step": 98, "time_per_iteration": 2.6276755332946777 }, { "auxiliary_loss_clip": 0.01665053, "auxiliary_loss_mlp": 0.01113791, "balance_loss_clip": 1.18055964, "balance_loss_mlp": 1.0621022, "epoch": 0.011904046173270006, "flos": 22340677669440.0, "grad_norm": 1.9822808825482605, "language_loss": 1.0166651, "learning_rate": 3.3289159651708192e-06, "loss": 1.04445362, "num_input_tokens_seen": 2152410, "step": 99, "time_per_iteration": 2.647322416305542 }, { "auxiliary_loss_clip": 0.01643798, "auxiliary_loss_mlp": 0.01129604, "balance_loss_clip": 1.17171514, "balance_loss_mlp": 1.07438624, "epoch": 0.012024289063909096, "flos": 19100063842560.0, "grad_norm": 2.436741397744935, "language_loss": 0.97658747, "learning_rate": 3.3361968902759768e-06, "loss": 1.00432146, "num_input_tokens_seen": 2172090, "step": 100, "time_per_iteration": 2.640697956085205 }, { "auxiliary_loss_clip": 0.01639623, "auxiliary_loss_mlp": 0.01103282, "balance_loss_clip": 1.1714983, "balance_loss_mlp": 1.0544064, "epoch": 0.012144531954548187, "flos": 15012208109760.0, "grad_norm": 2.4133022106506226, "language_loss": 0.93966925, "learning_rate": 3.343405367163663e-06, "loss": 0.9670983, "num_input_tokens_seen": 2189020, "step": 101, "time_per_iteration": 2.640946388244629 }, { "auxiliary_loss_clip": 0.01637109, "auxiliary_loss_mlp": 0.01100556, "balance_loss_clip": 1.17473078, "balance_loss_mlp": 1.05294371, "epoch": 0.012264774845187279, "flos": 15122023008480.0, "grad_norm": 2.518306130572046, "language_loss": 0.81317127, "learning_rate": 3.350542823419951e-06, "loss": 0.84054792, "num_input_tokens_seen": 2205620, "step": 102, "time_per_iteration": 2.617774248123169 }, { "auxiliary_loss_clip": 0.01629589, "auxiliary_loss_mlp": 0.01094724, "balance_loss_clip": 1.16765213, "balance_loss_mlp": 1.04654002, "epoch": 0.012385017735826368, "flos": 13949251355040.0, "grad_norm": 3.617546629226088, "language_loss": 0.87561083, "learning_rate": 3.3576106448465615e-06, "loss": 0.90285397, "num_input_tokens_seen": 2219000, "step": 103, "time_per_iteration": 2.623114824295044 }, { "auxiliary_loss_clip": 0.01626198, "auxiliary_loss_mlp": 0.01105841, "balance_loss_clip": 1.1707108, "balance_loss_mlp": 1.05539119, "epoch": 0.01250526062646546, "flos": 23623264221600.0, "grad_norm": 2.8666277182262814, "language_loss": 0.88324201, "learning_rate": 3.3646101770757797e-06, "loss": 0.9105624, "num_input_tokens_seen": 2237790, "step": 104, "time_per_iteration": 2.6787192821502686 }, { "auxiliary_loss_clip": 0.01609997, "auxiliary_loss_mlp": 0.01102656, "balance_loss_clip": 1.16515422, "balance_loss_mlp": 1.05583024, "epoch": 0.012625503517104552, "flos": 34640083605600.0, "grad_norm": 1.7059430054406823, "language_loss": 0.85658574, "learning_rate": 3.371542727108104e-06, "loss": 0.88371223, "num_input_tokens_seen": 2259965, "step": 105, "time_per_iteration": 2.7431509494781494 }, { "auxiliary_loss_clip": 0.01618026, "auxiliary_loss_mlp": 0.01111169, "balance_loss_clip": 1.16433477, "balance_loss_mlp": 1.06620288, "epoch": 0.012745746407743641, "flos": 17821895076480.0, "grad_norm": 3.2519752092259546, "language_loss": 0.89955366, "learning_rate": 3.3784095647770114e-06, "loss": 0.92684561, "num_input_tokens_seen": 2278610, "step": 106, "time_per_iteration": 2.6359031200408936 }, { "auxiliary_loss_clip": 0.01603634, "auxiliary_loss_mlp": 0.01104131, "balance_loss_clip": 1.16103387, "balance_loss_mlp": 1.058092, "epoch": 0.012865989298382733, "flos": 20595060883200.0, "grad_norm": 2.0439372769873265, "language_loss": 0.88615847, "learning_rate": 3.3852119241449547e-06, "loss": 0.91323608, "num_input_tokens_seen": 2297730, "step": 107, "time_per_iteration": 2.6236510276794434 }, { "auxiliary_loss_clip": 0.01600198, "auxiliary_loss_mlp": 0.01105843, "balance_loss_clip": 1.1608839, "balance_loss_mlp": 1.05727768, "epoch": 0.012986232189021825, "flos": 23948973553920.0, "grad_norm": 2.2993965247985186, "language_loss": 0.96213055, "learning_rate": 3.3919510048344295e-06, "loss": 0.98919094, "num_input_tokens_seen": 2315740, "step": 108, "time_per_iteration": 2.6841228008270264 }, { "auxiliary_loss_clip": 0.01594892, "auxiliary_loss_mlp": 0.01103835, "balance_loss_clip": 1.16091752, "balance_loss_mlp": 1.06146801, "epoch": 0.013106475079660914, "flos": 23725428807840.0, "grad_norm": 2.434932793643324, "language_loss": 0.86537874, "learning_rate": 3.3986279732976907e-06, "loss": 0.89236605, "num_input_tokens_seen": 2334215, "step": 109, "time_per_iteration": 2.6502323150634766 }, { "auxiliary_loss_clip": 0.01587095, "auxiliary_loss_mlp": 0.01113019, "balance_loss_clip": 1.15958428, "balance_loss_mlp": 1.07000852, "epoch": 0.013226717970300006, "flos": 21102442174560.0, "grad_norm": 1.9165314168017002, "language_loss": 0.95552194, "learning_rate": 3.4052439640284983e-06, "loss": 0.98252308, "num_input_tokens_seen": 2353130, "step": 110, "time_per_iteration": 2.6895768642425537 }, { "auxiliary_loss_clip": 0.01585841, "auxiliary_loss_mlp": 0.01105187, "balance_loss_clip": 1.16095066, "balance_loss_mlp": 1.06153202, "epoch": 0.013346960860939098, "flos": 24863906152800.0, "grad_norm": 1.7818951825461413, "language_loss": 0.80935615, "learning_rate": 3.4118000807190217e-06, "loss": 0.8362664, "num_input_tokens_seen": 2374010, "step": 111, "time_per_iteration": 2.6287996768951416 }, { "auxiliary_loss_clip": 0.01578175, "auxiliary_loss_mlp": 0.01111561, "balance_loss_clip": 1.1582402, "balance_loss_mlp": 1.07060075, "epoch": 0.013467203751578187, "flos": 28181944168800.0, "grad_norm": 1.8107643444663586, "language_loss": 0.76074827, "learning_rate": 3.4182973973648723e-06, "loss": 0.7876457, "num_input_tokens_seen": 2395220, "step": 112, "time_per_iteration": 2.8321030139923096 }, { "auxiliary_loss_clip": 0.01575859, "auxiliary_loss_mlp": 0.01108026, "balance_loss_clip": 1.15818453, "balance_loss_mlp": 1.06475282, "epoch": 0.013587446642217279, "flos": 18916236865920.0, "grad_norm": 2.624426613920711, "language_loss": 0.95159084, "learning_rate": 3.424736959321014e-06, "loss": 0.97842968, "num_input_tokens_seen": 2413025, "step": 113, "time_per_iteration": 2.615622043609619 }, { "auxiliary_loss_clip": 0.01580528, "auxiliary_loss_mlp": 0.01102429, "balance_loss_clip": 1.15960598, "balance_loss_mlp": 1.06468678, "epoch": 0.01370768953285637, "flos": 23988619489440.0, "grad_norm": 1.8940799995815456, "language_loss": 0.88704544, "learning_rate": 3.431119784311155e-06, "loss": 0.91387498, "num_input_tokens_seen": 2432700, "step": 114, "time_per_iteration": 2.684666395187378 }, { "auxiliary_loss_clip": 0.01556898, "auxiliary_loss_mlp": 0.01110751, "balance_loss_clip": 1.15362668, "balance_loss_mlp": 1.07186544, "epoch": 0.01382793242349546, "flos": 39202570750560.0, "grad_norm": 1.6108520641552193, "language_loss": 0.77558351, "learning_rate": 3.43744686339307e-06, "loss": 0.80225998, "num_input_tokens_seen": 2455020, "step": 115, "time_per_iteration": 2.777702808380127 }, { "auxiliary_loss_clip": 0.01557552, "auxiliary_loss_mlp": 0.01094714, "balance_loss_clip": 1.14814639, "balance_loss_mlp": 1.05344343, "epoch": 0.013948175314134552, "flos": 41353511992800.0, "grad_norm": 2.2116583412926816, "language_loss": 0.90863097, "learning_rate": 3.44371916188212e-06, "loss": 0.93515366, "num_input_tokens_seen": 2475775, "step": 116, "time_per_iteration": 2.800532341003418 }, { "auxiliary_loss_clip": 0.01549193, "auxiliary_loss_mlp": 0.01089601, "balance_loss_clip": 1.14885497, "balance_loss_mlp": 1.05162072, "epoch": 0.014068418204773643, "flos": 22453545509760.0, "grad_norm": 1.9724173004427206, "language_loss": 0.86265898, "learning_rate": 3.449937620235143e-06, "loss": 0.88904691, "num_input_tokens_seen": 2496370, "step": 117, "time_per_iteration": 3.4109866619110107 }, { "auxiliary_loss_clip": 0.01548567, "auxiliary_loss_mlp": 0.0110252, "balance_loss_clip": 1.15026164, "balance_loss_mlp": 1.064731, "epoch": 0.014188661095412733, "flos": 23805151682400.0, "grad_norm": 2.493836091865363, "language_loss": 0.89243013, "learning_rate": 3.456103154896722e-06, "loss": 0.91894102, "num_input_tokens_seen": 2517645, "step": 118, "time_per_iteration": 3.5204765796661377 }, { "auxiliary_loss_clip": 0.01538737, "auxiliary_loss_mlp": 0.01084085, "balance_loss_clip": 1.14797425, "balance_loss_mlp": 1.05020559, "epoch": 0.014308903986051825, "flos": 23660252302080.0, "grad_norm": 1.8350645382247661, "language_loss": 0.92340469, "learning_rate": 3.462216659109757e-06, "loss": 0.94963288, "num_input_tokens_seen": 2537825, "step": 119, "time_per_iteration": 3.545672655105591 }, { "auxiliary_loss_clip": 0.01548019, "auxiliary_loss_mlp": 0.01106334, "balance_loss_clip": 1.14961839, "balance_loss_mlp": 1.07002342, "epoch": 0.014429146876690916, "flos": 20667995452320.0, "grad_norm": 2.808106405821987, "language_loss": 0.8536942, "learning_rate": 3.4682790036921077e-06, "loss": 0.8802377, "num_input_tokens_seen": 2556485, "step": 120, "time_per_iteration": 3.4586966037750244 }, { "auxiliary_loss_clip": 0.01531721, "auxiliary_loss_mlp": 0.0109408, "balance_loss_clip": 1.14807546, "balance_loss_mlp": 1.06296599, "epoch": 0.014549389767330006, "flos": 20229202777920.0, "grad_norm": 2.1244221308977056, "language_loss": 0.83302748, "learning_rate": 3.4742910377810193e-06, "loss": 0.85928547, "num_input_tokens_seen": 2573945, "step": 121, "time_per_iteration": 2.6174237728118896 }, { "auxiliary_loss_clip": 0.01524885, "auxiliary_loss_mlp": 0.0110396, "balance_loss_clip": 1.14336395, "balance_loss_mlp": 1.07089114, "epoch": 0.014669632657969098, "flos": 18004177623840.0, "grad_norm": 2.061503841961702, "language_loss": 0.88868821, "learning_rate": 3.4802535895469042e-06, "loss": 0.91497672, "num_input_tokens_seen": 2592695, "step": 122, "time_per_iteration": 2.6318302154541016 }, { "auxiliary_loss_clip": 0.0152671, "auxiliary_loss_mlp": 0.01091396, "balance_loss_clip": 1.14571214, "balance_loss_mlp": 1.05842245, "epoch": 0.01478987554860819, "flos": 22741799841120.0, "grad_norm": 2.0655275834622513, "language_loss": 0.89538836, "learning_rate": 3.4861674668779934e-06, "loss": 0.92156935, "num_input_tokens_seen": 2610925, "step": 123, "time_per_iteration": 2.613194227218628 }, { "auxiliary_loss_clip": 0.01518331, "auxiliary_loss_mlp": 0.01083545, "balance_loss_clip": 1.14196885, "balance_loss_mlp": 1.0518117, "epoch": 0.01491011843924728, "flos": 17198593003200.0, "grad_norm": 1.9872640791831155, "language_loss": 0.84024107, "learning_rate": 3.492033458037272e-06, "loss": 0.86625981, "num_input_tokens_seen": 2629495, "step": 124, "time_per_iteration": 2.6273202896118164 }, { "auxiliary_loss_clip": 0.01513495, "auxiliary_loss_mlp": 0.01087048, "balance_loss_clip": 1.13785505, "balance_loss_mlp": 1.05521965, "epoch": 0.01503036132988637, "flos": 17673870920640.0, "grad_norm": 3.7995668952609107, "language_loss": 0.87113833, "learning_rate": 3.497852332293018e-06, "loss": 0.89714384, "num_input_tokens_seen": 2645070, "step": 125, "time_per_iteration": 2.578320264816284 }, { "auxiliary_loss_clip": 0.0151508, "auxiliary_loss_mlp": 0.01097892, "balance_loss_clip": 1.14199877, "balance_loss_mlp": 1.06590772, "epoch": 0.015150604220525462, "flos": 18878243110560.0, "grad_norm": 2.105240555627649, "language_loss": 0.96252024, "learning_rate": 3.5036248405242356e-06, "loss": 0.98864996, "num_input_tokens_seen": 2663825, "step": 126, "time_per_iteration": 2.636725664138794 }, { "auxiliary_loss_clip": 0.01510263, "auxiliary_loss_mlp": 0.01079714, "balance_loss_clip": 1.13936996, "balance_loss_mlp": 1.04988742, "epoch": 0.015270847111164552, "flos": 39420189198240.0, "grad_norm": 2.0749247888635196, "language_loss": 0.82688391, "learning_rate": 3.509351715802146e-06, "loss": 0.85278368, "num_input_tokens_seen": 2684710, "step": 127, "time_per_iteration": 2.788679361343384 }, { "auxiliary_loss_clip": 0.01511724, "auxiliary_loss_mlp": 0.01072572, "balance_loss_clip": 1.13778138, "balance_loss_mlp": 1.04371095, "epoch": 0.015391090001803644, "flos": 43762471874400.0, "grad_norm": 1.902994420428896, "language_loss": 0.78456616, "learning_rate": 3.5150336739488763e-06, "loss": 0.81040907, "num_input_tokens_seen": 2706995, "step": 128, "time_per_iteration": 2.794884443283081 }, { "auxiliary_loss_clip": 0.0149542, "auxiliary_loss_mlp": 0.01084627, "balance_loss_clip": 1.13798296, "balance_loss_mlp": 1.05337024, "epoch": 0.015511332892442733, "flos": 18916344616800.0, "grad_norm": 3.1332941663754434, "language_loss": 0.83997655, "learning_rate": 3.5206714140744143e-06, "loss": 0.86577702, "num_input_tokens_seen": 2727050, "step": 129, "time_per_iteration": 2.712477922439575 }, { "auxiliary_loss_clip": 0.01501934, "auxiliary_loss_mlp": 0.01093829, "balance_loss_clip": 1.1435647, "balance_loss_mlp": 1.06678045, "epoch": 0.015631575783081827, "flos": 24535287546720.0, "grad_norm": 2.55907089149707, "language_loss": 0.87467241, "learning_rate": 3.5262656190928208e-06, "loss": 0.90063006, "num_input_tokens_seen": 2745350, "step": 130, "time_per_iteration": 2.6503500938415527 }, { "auxiliary_loss_clip": 0.01512145, "auxiliary_loss_mlp": 0.01129817, "balance_loss_clip": 1.25555682, "balance_loss_mlp": 1.12318945, "epoch": 0.015751818673720917, "flos": 62328546346560.0, "grad_norm": 1.2164731839121286, "language_loss": 0.71529019, "learning_rate": 3.5318169562186737e-06, "loss": 0.74170983, "num_input_tokens_seen": 2814195, "step": 131, "time_per_iteration": 3.301440477371216 }, { "auxiliary_loss_clip": 0.01492609, "auxiliary_loss_mlp": 0.01090226, "balance_loss_clip": 1.13806248, "balance_loss_mlp": 1.06150889, "epoch": 0.015872061564360006, "flos": 23878553172000.0, "grad_norm": 3.5487832413592835, "language_loss": 0.8207891, "learning_rate": 3.5373260774446292e-06, "loss": 0.84661746, "num_input_tokens_seen": 2834645, "step": 132, "time_per_iteration": 2.6260602474212646 }, { "auxiliary_loss_clip": 0.01476264, "auxiliary_loss_mlp": 0.0107638, "balance_loss_clip": 1.13084865, "balance_loss_mlp": 1.04958129, "epoch": 0.0159923044549991, "flos": 23367903437280.0, "grad_norm": 2.0383212649921125, "language_loss": 0.90528101, "learning_rate": 3.542793620000961e-06, "loss": 0.93080747, "num_input_tokens_seen": 2854120, "step": 133, "time_per_iteration": 2.6288111209869385 }, { "auxiliary_loss_clip": 0.01485603, "auxiliary_loss_mlp": 0.01093045, "balance_loss_clip": 1.13688648, "balance_loss_mlp": 1.06659245, "epoch": 0.01611254734563819, "flos": 17858308485600.0, "grad_norm": 2.236034251289101, "language_loss": 0.86828816, "learning_rate": 3.5482202067978894e-06, "loss": 0.89407462, "num_input_tokens_seen": 2871330, "step": 134, "time_per_iteration": 2.5923595428466797 }, { "auxiliary_loss_clip": 0.01485732, "auxiliary_loss_mlp": 0.01082665, "balance_loss_clip": 1.13391674, "balance_loss_mlp": 1.05395913, "epoch": 0.01623279023627728, "flos": 20954777188320.0, "grad_norm": 2.080002971295935, "language_loss": 0.75951135, "learning_rate": 3.553606446851471e-06, "loss": 0.78519535, "num_input_tokens_seen": 2888070, "step": 135, "time_per_iteration": 2.640122890472412 }, { "auxiliary_loss_clip": 0.01479021, "auxiliary_loss_mlp": 0.01099558, "balance_loss_clip": 1.13500249, "balance_loss_mlp": 1.07265282, "epoch": 0.016353033126916373, "flos": 15742415808000.0, "grad_norm": 1.7450999342173072, "language_loss": 0.83371854, "learning_rate": 3.5589529356937613e-06, "loss": 0.85950434, "num_input_tokens_seen": 2906465, "step": 136, "time_per_iteration": 2.591466188430786 }, { "auxiliary_loss_clip": 0.01479621, "auxiliary_loss_mlp": 0.01086297, "balance_loss_clip": 1.13266778, "balance_loss_mlp": 1.05844998, "epoch": 0.016473276017555463, "flos": 18807283974240.0, "grad_norm": 10.01818864733358, "language_loss": 0.76825142, "learning_rate": 3.5642602557679627e-06, "loss": 0.79391062, "num_input_tokens_seen": 2924915, "step": 137, "time_per_iteration": 2.6275181770324707 }, { "auxiliary_loss_clip": 0.01474568, "auxiliary_loss_mlp": 0.01082395, "balance_loss_clip": 1.13865912, "balance_loss_mlp": 1.0583508, "epoch": 0.016593518908194552, "flos": 24352645829760.0, "grad_norm": 2.6275203742889417, "language_loss": 0.84001535, "learning_rate": 3.569528976809202e-06, "loss": 0.86558503, "num_input_tokens_seen": 2942130, "step": 138, "time_per_iteration": 2.6183922290802 }, { "auxiliary_loss_clip": 0.01470942, "auxiliary_loss_mlp": 0.01087239, "balance_loss_clip": 1.12824404, "balance_loss_mlp": 1.06255054, "epoch": 0.016713761798833646, "flos": 22346137047360.0, "grad_norm": 2.0294580903853547, "language_loss": 0.89856845, "learning_rate": 3.5747596562115522e-06, "loss": 0.92415023, "num_input_tokens_seen": 2962745, "step": 139, "time_per_iteration": 2.651902675628662 }, { "auxiliary_loss_clip": 0.01475366, "auxiliary_loss_mlp": 0.01085885, "balance_loss_clip": 1.13294554, "balance_loss_mlp": 1.05924165, "epoch": 0.016834004689472735, "flos": 17821823242560.0, "grad_norm": 3.089942004452782, "language_loss": 0.907812, "learning_rate": 3.5799528393819138e-06, "loss": 0.93342447, "num_input_tokens_seen": 2981825, "step": 140, "time_per_iteration": 2.6140267848968506 }, { "auxiliary_loss_clip": 0.01456862, "auxiliary_loss_mlp": 0.01077177, "balance_loss_clip": 1.12109232, "balance_loss_mlp": 1.05242956, "epoch": 0.016954247580111825, "flos": 20519504376000.0, "grad_norm": 2.081975873238969, "language_loss": 0.87941468, "learning_rate": 3.585109060081286e-06, "loss": 0.90475512, "num_input_tokens_seen": 3001625, "step": 141, "time_per_iteration": 2.6150050163269043 }, { "auxiliary_loss_clip": 0.01458788, "auxiliary_loss_mlp": 0.01088117, "balance_loss_clip": 1.12470484, "balance_loss_mlp": 1.06371486, "epoch": 0.017074490470750915, "flos": 22088872664160.0, "grad_norm": 1.8057041664241742, "language_loss": 0.78538859, "learning_rate": 3.590228840753992e-06, "loss": 0.81085753, "num_input_tokens_seen": 3022055, "step": 142, "time_per_iteration": 2.652402877807617 }, { "auxiliary_loss_clip": 0.01461129, "auxiliary_loss_mlp": 0.0108597, "balance_loss_clip": 1.1280427, "balance_loss_mlp": 1.06072187, "epoch": 0.01719473336139001, "flos": 15997273754880.0, "grad_norm": 2.6895006568971627, "language_loss": 0.87376285, "learning_rate": 3.5953126928453423e-06, "loss": 0.89923388, "num_input_tokens_seen": 3039605, "step": 143, "time_per_iteration": 3.3255090713500977 }, { "auxiliary_loss_clip": 0.01457103, "auxiliary_loss_mlp": 0.0109176, "balance_loss_clip": 1.12495995, "balance_loss_mlp": 1.0668577, "epoch": 0.017314976252029098, "flos": 22492042102560.0, "grad_norm": 1.9503216314748542, "language_loss": 0.80540729, "learning_rate": 3.600361117108239e-06, "loss": 0.83089584, "num_input_tokens_seen": 3059405, "step": 144, "time_per_iteration": 2.6401467323303223 }, { "auxiliary_loss_clip": 0.01451702, "auxiliary_loss_mlp": 0.01087928, "balance_loss_clip": 1.12135637, "balance_loss_mlp": 1.06263137, "epoch": 0.017435219142668188, "flos": 22018057195680.0, "grad_norm": 2.099408523623896, "language_loss": 0.97220743, "learning_rate": 3.6053746038991616e-06, "loss": 0.99760365, "num_input_tokens_seen": 3078490, "step": 145, "time_per_iteration": 3.514524221420288 }, { "auxiliary_loss_clip": 0.01448876, "auxiliary_loss_mlp": 0.01026794, "balance_loss_clip": 1.21540284, "balance_loss_mlp": 1.02135837, "epoch": 0.01755546203330728, "flos": 72240536188800.0, "grad_norm": 1.0784355882305068, "language_loss": 0.58465755, "learning_rate": 3.6103536334639843e-06, "loss": 0.60941434, "num_input_tokens_seen": 3131755, "step": 146, "time_per_iteration": 4.004445552825928 }, { "auxiliary_loss_clip": 0.01438559, "auxiliary_loss_mlp": 0.01065222, "balance_loss_clip": 1.11999667, "balance_loss_mlp": 1.04343057, "epoch": 0.01767570492394637, "flos": 25337064969600.0, "grad_norm": 2.257772766699115, "language_loss": 0.85642004, "learning_rate": 3.615298676214041e-06, "loss": 0.88145792, "num_input_tokens_seen": 3152035, "step": 147, "time_per_iteration": 3.3899195194244385 }, { "auxiliary_loss_clip": 0.01446028, "auxiliary_loss_mlp": 0.01070684, "balance_loss_clip": 1.12169898, "balance_loss_mlp": 1.049191, "epoch": 0.01779594781458546, "flos": 20449191744960.0, "grad_norm": 3.0497162605511137, "language_loss": 0.88860416, "learning_rate": 3.6202101929928317e-06, "loss": 0.91377121, "num_input_tokens_seen": 3170625, "step": 148, "time_per_iteration": 2.615971326828003 }, { "auxiliary_loss_clip": 0.01438357, "auxiliary_loss_mlp": 0.01097074, "balance_loss_clip": 1.11827588, "balance_loss_mlp": 1.07585526, "epoch": 0.017916190705224554, "flos": 16253604297120.0, "grad_norm": 1.9388044294218558, "language_loss": 0.88251537, "learning_rate": 3.6250886353337413e-06, "loss": 0.9078697, "num_input_tokens_seen": 3188155, "step": 149, "time_per_iteration": 2.6070926189422607 }, { "auxiliary_loss_clip": 0.01442043, "auxiliary_loss_mlp": 0.01075123, "balance_loss_clip": 1.11896205, "balance_loss_mlp": 1.05324769, "epoch": 0.018036433595863644, "flos": 23330592104160.0, "grad_norm": 9.934211133239272, "language_loss": 0.8623879, "learning_rate": 3.6299344457091488e-06, "loss": 0.88755953, "num_input_tokens_seen": 3209015, "step": 150, "time_per_iteration": 2.671457052230835 }, { "auxiliary_loss_clip": 0.01442157, "auxiliary_loss_mlp": 0.01066924, "balance_loss_clip": 1.12178206, "balance_loss_mlp": 1.04556203, "epoch": 0.018156676486502734, "flos": 18588013346400.0, "grad_norm": 2.715741061025007, "language_loss": 0.93588543, "learning_rate": 3.634748057771256e-06, "loss": 0.96097624, "num_input_tokens_seen": 3224955, "step": 151, "time_per_iteration": 2.5711803436279297 }, { "auxiliary_loss_clip": 0.01431126, "auxiliary_loss_mlp": 0.01086715, "balance_loss_clip": 1.11710072, "balance_loss_mlp": 1.06507874, "epoch": 0.018276919377141827, "flos": 25448711633280.0, "grad_norm": 1.682093278920507, "language_loss": 0.8570497, "learning_rate": 3.639529896584965e-06, "loss": 0.88222808, "num_input_tokens_seen": 3246330, "step": 152, "time_per_iteration": 2.6580348014831543 }, { "auxiliary_loss_clip": 0.01432309, "auxiliary_loss_mlp": 0.01069926, "balance_loss_clip": 1.11756432, "balance_loss_mlp": 1.04851651, "epoch": 0.018397162267780917, "flos": 20047315317120.0, "grad_norm": 3.2288231034386663, "language_loss": 0.88685024, "learning_rate": 3.6442803788531233e-06, "loss": 0.91187251, "num_input_tokens_seen": 3264290, "step": 153, "time_per_iteration": 2.581289529800415 }, { "auxiliary_loss_clip": 0.01433276, "auxiliary_loss_mlp": 0.01064068, "balance_loss_clip": 1.11806417, "balance_loss_mlp": 1.04340863, "epoch": 0.018517405158420007, "flos": 27565286733120.0, "grad_norm": 1.9553838186445869, "language_loss": 0.96291661, "learning_rate": 3.6489999131344357e-06, "loss": 0.98789006, "num_input_tokens_seen": 3287065, "step": 154, "time_per_iteration": 2.6729555130004883 }, { "auxiliary_loss_clip": 0.01421341, "auxiliary_loss_mlp": 0.01068447, "balance_loss_clip": 1.11782682, "balance_loss_mlp": 1.04865861, "epoch": 0.0186376480490591, "flos": 19354059782400.0, "grad_norm": 2.039051538986895, "language_loss": 0.90611857, "learning_rate": 3.653688900054313e-06, "loss": 0.93101639, "num_input_tokens_seen": 3305595, "step": 155, "time_per_iteration": 2.5882136821746826 }, { "auxiliary_loss_clip": 0.01427983, "auxiliary_loss_mlp": 0.01071771, "balance_loss_clip": 1.11139369, "balance_loss_mlp": 1.05089736, "epoch": 0.01875789093969819, "flos": 26687845052160.0, "grad_norm": 2.0450534015636967, "language_loss": 0.75969702, "learning_rate": 3.6583477325089526e-06, "loss": 0.78469461, "num_input_tokens_seen": 3326135, "step": 156, "time_per_iteration": 2.6633684635162354 }, { "auxiliary_loss_clip": 0.01422623, "auxiliary_loss_mlp": 0.01063361, "balance_loss_clip": 1.11469889, "balance_loss_mlp": 1.04335809, "epoch": 0.01887813383033728, "flos": 24353004999360.0, "grad_norm": 2.2275806497637216, "language_loss": 1.03950381, "learning_rate": 3.6629767958628916e-06, "loss": 1.06436372, "num_input_tokens_seen": 3343510, "step": 157, "time_per_iteration": 2.5954387187957764 }, { "auxiliary_loss_clip": 0.01427375, "auxiliary_loss_mlp": 0.01064774, "balance_loss_clip": 1.11681068, "balance_loss_mlp": 1.04453194, "epoch": 0.018998376720976373, "flos": 14647535264160.0, "grad_norm": 2.6416001542039442, "language_loss": 0.85382688, "learning_rate": 3.667576468140291e-06, "loss": 0.87874842, "num_input_tokens_seen": 3361325, "step": 158, "time_per_iteration": 2.5909407138824463 }, { "auxiliary_loss_clip": 0.01420929, "auxiliary_loss_mlp": 0.01070579, "balance_loss_clip": 1.11532009, "balance_loss_mlp": 1.05018198, "epoch": 0.019118619611615463, "flos": 29305264556640.0, "grad_norm": 2.2645495919181986, "language_loss": 0.89039165, "learning_rate": 3.672147120210184e-06, "loss": 0.91530675, "num_input_tokens_seen": 3377925, "step": 159, "time_per_iteration": 2.632502317428589 }, { "auxiliary_loss_clip": 0.01419953, "auxiliary_loss_mlp": 0.01069074, "balance_loss_clip": 1.11613142, "balance_loss_mlp": 1.05051267, "epoch": 0.019238862502254553, "flos": 20886727325760.0, "grad_norm": 3.234121907602069, "language_loss": 0.86307985, "learning_rate": 3.6766891159659177e-06, "loss": 0.88797009, "num_input_tokens_seen": 3396335, "step": 160, "time_per_iteration": 2.6374354362487793 }, { "auxiliary_loss_clip": 0.01417028, "auxiliary_loss_mlp": 0.01064422, "balance_loss_clip": 1.11295104, "balance_loss_mlp": 1.04558659, "epoch": 0.019359105392893646, "flos": 21360676315680.0, "grad_norm": 2.4375518301429793, "language_loss": 0.87601447, "learning_rate": 3.6812028124990075e-06, "loss": 0.90082896, "num_input_tokens_seen": 3413605, "step": 161, "time_per_iteration": 2.5842478275299072 }, { "auxiliary_loss_clip": 0.01411702, "auxiliary_loss_mlp": 0.01075281, "balance_loss_clip": 1.11240149, "balance_loss_mlp": 1.05550444, "epoch": 0.019479348283532736, "flos": 16283732238240.0, "grad_norm": 3.2262540142993172, "language_loss": 0.81679803, "learning_rate": 3.6856885602676016e-06, "loss": 0.84166789, "num_input_tokens_seen": 3429640, "step": 162, "time_per_iteration": 2.5986087322235107 }, { "auxiliary_loss_clip": 0.01414439, "auxiliary_loss_mlp": 0.01070089, "balance_loss_clip": 1.11213362, "balance_loss_mlp": 1.05102754, "epoch": 0.019599591174171826, "flos": 22091243183520.0, "grad_norm": 2.204047267732158, "language_loss": 0.9416517, "learning_rate": 3.6901467032597733e-06, "loss": 0.96649694, "num_input_tokens_seen": 3448125, "step": 163, "time_per_iteration": 2.5872750282287598 }, { "auxiliary_loss_clip": 0.01414158, "auxiliary_loss_mlp": 0.01072315, "balance_loss_clip": 1.10969925, "balance_loss_mlp": 1.05338514, "epoch": 0.01971983406481092, "flos": 19609671985440.0, "grad_norm": 2.2808588264034744, "language_loss": 0.87058246, "learning_rate": 3.694577579151804e-06, "loss": 0.89544725, "num_input_tokens_seen": 3466535, "step": 164, "time_per_iteration": 2.61564564704895 }, { "auxiliary_loss_clip": 0.01413968, "auxiliary_loss_mlp": 0.01062961, "balance_loss_clip": 1.11347055, "balance_loss_mlp": 1.04568768, "epoch": 0.01984007695545001, "flos": 19099345503360.0, "grad_norm": 2.0948592326987994, "language_loss": 0.73487222, "learning_rate": 3.6989815194616703e-06, "loss": 0.75964153, "num_input_tokens_seen": 3483730, "step": 165, "time_per_iteration": 2.567875623703003 }, { "auxiliary_loss_clip": 0.01414649, "auxiliary_loss_mlp": 0.01074546, "balance_loss_clip": 1.11131847, "balance_loss_mlp": 1.05488825, "epoch": 0.0199603198460891, "flos": 20848410317760.0, "grad_norm": 2.032888595969709, "language_loss": 0.79634476, "learning_rate": 3.703358849697888e-06, "loss": 0.82123673, "num_input_tokens_seen": 3503640, "step": 166, "time_per_iteration": 2.6178793907165527 }, { "auxiliary_loss_clip": 0.01402958, "auxiliary_loss_mlp": 0.01067422, "balance_loss_clip": 1.11029899, "balance_loss_mlp": 1.04894447, "epoch": 0.020080562736728192, "flos": 21870751379040.0, "grad_norm": 1.956172225300174, "language_loss": 0.82818115, "learning_rate": 3.7077098895038803e-06, "loss": 0.85288501, "num_input_tokens_seen": 3523010, "step": 167, "time_per_iteration": 2.593693971633911 }, { "auxiliary_loss_clip": 0.01402905, "auxiliary_loss_mlp": 0.01062292, "balance_loss_clip": 1.10938394, "balance_loss_mlp": 1.043612, "epoch": 0.020200805627367282, "flos": 21688792084320.0, "grad_norm": 2.048147282723413, "language_loss": 0.96991408, "learning_rate": 3.712034952798045e-06, "loss": 0.99456608, "num_input_tokens_seen": 3541125, "step": 168, "time_per_iteration": 2.6070761680603027 }, { "auxiliary_loss_clip": 0.01404685, "auxiliary_loss_mlp": 0.010863, "balance_loss_clip": 1.1093744, "balance_loss_mlp": 1.06658292, "epoch": 0.02032104851800637, "flos": 33543048044160.0, "grad_norm": 2.5532853632149255, "language_loss": 0.84694344, "learning_rate": 3.7163343479096656e-06, "loss": 0.87185329, "num_input_tokens_seen": 3562700, "step": 169, "time_per_iteration": 2.7079474925994873 }, { "auxiliary_loss_clip": 0.01398229, "auxiliary_loss_mlp": 0.01059152, "balance_loss_clip": 1.10603213, "balance_loss_mlp": 1.04226017, "epoch": 0.020441291408645465, "flos": 31686969853920.0, "grad_norm": 2.1160323212515215, "language_loss": 0.82771266, "learning_rate": 3.720608377710802e-06, "loss": 0.8522864, "num_input_tokens_seen": 3582790, "step": 170, "time_per_iteration": 3.367776393890381 }, { "auxiliary_loss_clip": 0.01396332, "auxiliary_loss_mlp": 0.01062135, "balance_loss_clip": 1.10466003, "balance_loss_mlp": 1.04467034, "epoch": 0.020561534299284555, "flos": 20886691408800.0, "grad_norm": 4.065605293768299, "language_loss": 0.86411536, "learning_rate": 3.7248573397443277e-06, "loss": 0.88870001, "num_input_tokens_seen": 3601715, "step": 171, "time_per_iteration": 3.376027822494507 }, { "auxiliary_loss_clip": 0.01397707, "auxiliary_loss_mlp": 0.01067229, "balance_loss_clip": 1.10952568, "balance_loss_mlp": 1.04968131, "epoch": 0.020681777189923645, "flos": 20996614058400.0, "grad_norm": 2.105426904302891, "language_loss": 0.97400177, "learning_rate": 3.729081526348224e-06, "loss": 0.99865115, "num_input_tokens_seen": 3620245, "step": 172, "time_per_iteration": 2.549234390258789 }, { "auxiliary_loss_clip": 0.01401443, "auxiliary_loss_mlp": 0.010654, "balance_loss_clip": 1.1098789, "balance_loss_mlp": 1.04792428, "epoch": 0.020802020080562738, "flos": 28257536592960.0, "grad_norm": 1.7771915993249825, "language_loss": 0.85175258, "learning_rate": 3.7332812247762777e-06, "loss": 0.87642097, "num_input_tokens_seen": 3641545, "step": 173, "time_per_iteration": 4.165031671524048 }, { "auxiliary_loss_clip": 0.01399742, "auxiliary_loss_mlp": 0.01056621, "balance_loss_clip": 1.10905719, "balance_loss_mlp": 1.03964567, "epoch": 0.020922262971201828, "flos": 19681277627040.0, "grad_norm": 2.576597089425312, "language_loss": 0.95594621, "learning_rate": 3.737456717315293e-06, "loss": 0.98050988, "num_input_tokens_seen": 3660510, "step": 174, "time_per_iteration": 2.6277098655700684 }, { "auxiliary_loss_clip": 0.01392773, "auxiliary_loss_mlp": 0.01069248, "balance_loss_clip": 1.107903, "balance_loss_mlp": 1.05254734, "epoch": 0.021042505861840918, "flos": 15666356463360.0, "grad_norm": 1.6847822066675833, "language_loss": 0.90637958, "learning_rate": 3.7416082813989552e-06, "loss": 0.93099982, "num_input_tokens_seen": 3677505, "step": 175, "time_per_iteration": 2.5673351287841797 }, { "auxiliary_loss_clip": 0.01401915, "auxiliary_loss_mlp": 0.01070146, "balance_loss_clip": 1.10974002, "balance_loss_mlp": 1.05237222, "epoch": 0.02116274875248001, "flos": 21142016276160.0, "grad_norm": 2.6693426299860494, "language_loss": 0.89317095, "learning_rate": 3.745736189718439e-06, "loss": 0.9178915, "num_input_tokens_seen": 3696760, "step": 176, "time_per_iteration": 2.6138112545013428 }, { "auxiliary_loss_clip": 0.01387659, "auxiliary_loss_mlp": 0.01065234, "balance_loss_clip": 1.10493827, "balance_loss_mlp": 1.04892588, "epoch": 0.0212829916431191, "flos": 24715774246080.0, "grad_norm": 2.5595051951505914, "language_loss": 0.72623628, "learning_rate": 3.749840710329894e-06, "loss": 0.7507652, "num_input_tokens_seen": 3717465, "step": 177, "time_per_iteration": 2.5995986461639404 }, { "auxiliary_loss_clip": 0.01399021, "auxiliary_loss_mlp": 0.01074495, "balance_loss_clip": 1.10975587, "balance_loss_mlp": 1.05644679, "epoch": 0.02140323453375819, "flos": 16645495810080.0, "grad_norm": 2.585154282129926, "language_loss": 0.98029912, "learning_rate": 3.7539221067588938e-06, "loss": 1.00503421, "num_input_tokens_seen": 3731440, "step": 178, "time_per_iteration": 2.606161594390869 }, { "auxiliary_loss_clip": 0.0139507, "auxiliary_loss_mlp": 0.01061423, "balance_loss_clip": 1.10589838, "balance_loss_mlp": 1.04269576, "epoch": 0.021523477424397284, "flos": 20299335824160.0, "grad_norm": 5.215364350531932, "language_loss": 0.93215531, "learning_rate": 3.757980638101964e-06, "loss": 0.95672023, "num_input_tokens_seen": 3744935, "step": 179, "time_per_iteration": 2.598646640777588 }, { "auxiliary_loss_clip": 0.01398001, "auxiliary_loss_mlp": 0.0106052, "balance_loss_clip": 1.10888267, "balance_loss_mlp": 1.04185176, "epoch": 0.021643720315036374, "flos": 26104009329600.0, "grad_norm": 2.2269254267045704, "language_loss": 0.89390844, "learning_rate": 3.7620165591252806e-06, "loss": 0.91849363, "num_input_tokens_seen": 3763035, "step": 180, "time_per_iteration": 2.6888222694396973 }, { "auxiliary_loss_clip": 0.01391809, "auxiliary_loss_mlp": 0.01075154, "balance_loss_clip": 1.10670519, "balance_loss_mlp": 1.05912042, "epoch": 0.021763963205675464, "flos": 24787667223360.0, "grad_norm": 1.9174189620578053, "language_loss": 0.94485742, "learning_rate": 3.766030120360636e-06, "loss": 0.96952707, "num_input_tokens_seen": 3782665, "step": 181, "time_per_iteration": 2.624023199081421 }, { "auxiliary_loss_clip": 0.01390766, "auxiliary_loss_mlp": 0.01061628, "balance_loss_clip": 1.10480678, "balance_loss_mlp": 1.04402137, "epoch": 0.021884206096314557, "flos": 25813564063680.0, "grad_norm": 2.092796374969199, "language_loss": 0.90105474, "learning_rate": 3.7700215681987578e-06, "loss": 0.92557871, "num_input_tokens_seen": 3802435, "step": 182, "time_per_iteration": 2.6785888671875 }, { "auxiliary_loss_clip": 0.01388016, "auxiliary_loss_mlp": 0.01064481, "balance_loss_clip": 1.10520697, "balance_loss_mlp": 1.04754174, "epoch": 0.022004448986953647, "flos": 20082723051360.0, "grad_norm": 2.1524117255880224, "language_loss": 0.82262433, "learning_rate": 3.7739911449800767e-06, "loss": 0.84714931, "num_input_tokens_seen": 3822490, "step": 183, "time_per_iteration": 2.6091864109039307 }, { "auxiliary_loss_clip": 0.01383734, "auxiliary_loss_mlp": 0.01051155, "balance_loss_clip": 1.10210299, "balance_loss_mlp": 1.03332138, "epoch": 0.022124691877592736, "flos": 20480612696640.0, "grad_norm": 2.17636623720557, "language_loss": 0.80629253, "learning_rate": 3.7779390890830114e-06, "loss": 0.83064139, "num_input_tokens_seen": 3841140, "step": 184, "time_per_iteration": 2.633136510848999 }, { "auxiliary_loss_clip": 0.01382279, "auxiliary_loss_mlp": 0.01062798, "balance_loss_clip": 1.10225725, "balance_loss_mlp": 1.0462873, "epoch": 0.02224493476823183, "flos": 23586850812480.0, "grad_norm": 1.7001519896989823, "language_loss": 0.85906744, "learning_rate": 3.7818656350098723e-06, "loss": 0.88351822, "num_input_tokens_seen": 3862090, "step": 185, "time_per_iteration": 2.643200635910034 }, { "auxiliary_loss_clip": 0.01386798, "auxiliary_loss_mlp": 0.01064703, "balance_loss_clip": 1.10535073, "balance_loss_mlp": 1.04716766, "epoch": 0.02236517765887092, "flos": 16909943585280.0, "grad_norm": 2.7217245898316307, "language_loss": 0.77165663, "learning_rate": 3.7857710134704447e-06, "loss": 0.79617161, "num_input_tokens_seen": 3881025, "step": 186, "time_per_iteration": 2.60864520072937 }, { "auxiliary_loss_clip": 0.0138302, "auxiliary_loss_mlp": 0.01063369, "balance_loss_clip": 1.10664773, "balance_loss_mlp": 1.04756224, "epoch": 0.02248542054951001, "flos": 43508188598880.0, "grad_norm": 2.1610552520479156, "language_loss": 0.79424381, "learning_rate": 3.7896554514633234e-06, "loss": 0.8187077, "num_input_tokens_seen": 3905310, "step": 187, "time_per_iteration": 2.8184266090393066 }, { "auxiliary_loss_clip": 0.01379502, "auxiliary_loss_mlp": 0.01052988, "balance_loss_clip": 1.10220778, "balance_loss_mlp": 1.03687072, "epoch": 0.022605663440149103, "flos": 23367652018560.0, "grad_norm": 2.0697482426042355, "language_loss": 0.84134203, "learning_rate": 3.7935191723550955e-06, "loss": 0.86566693, "num_input_tokens_seen": 3924265, "step": 188, "time_per_iteration": 2.6071524620056152 }, { "auxiliary_loss_clip": 0.01373461, "auxiliary_loss_mlp": 0.01054963, "balance_loss_clip": 1.09994256, "balance_loss_mlp": 1.03826189, "epoch": 0.022725906330788193, "flos": 29019919499040.0, "grad_norm": 2.0307521672145317, "language_loss": 0.88573951, "learning_rate": 3.797362395957408e-06, "loss": 0.91002381, "num_input_tokens_seen": 3944830, "step": 189, "time_per_iteration": 2.6863207817077637 }, { "auxiliary_loss_clip": 0.01385615, "auxiliary_loss_mlp": 0.01055956, "balance_loss_clip": 1.10616064, "balance_loss_mlp": 1.03844404, "epoch": 0.022846149221427282, "flos": 24496180365600.0, "grad_norm": 2.377309920464463, "language_loss": 0.78229415, "learning_rate": 3.8011853386020055e-06, "loss": 0.80670989, "num_input_tokens_seen": 3965735, "step": 190, "time_per_iteration": 2.628861427307129 }, { "auxiliary_loss_clip": 0.01381946, "auxiliary_loss_mlp": 0.01053705, "balance_loss_clip": 1.1052649, "balance_loss_mlp": 1.03733718, "epoch": 0.022966392112066376, "flos": 15523540266720.0, "grad_norm": 4.968131761169128, "language_loss": 0.89787328, "learning_rate": 3.804988213213804e-06, "loss": 0.92222989, "num_input_tokens_seen": 3983975, "step": 191, "time_per_iteration": 2.6103291511535645 }, { "auxiliary_loss_clip": 0.01404585, "auxiliary_loss_mlp": 0.01009391, "balance_loss_clip": 1.193964, "balance_loss_mlp": 1.00166643, "epoch": 0.023086635002705466, "flos": 55650417942720.0, "grad_norm": 1.025782873580317, "language_loss": 0.63223195, "learning_rate": 3.808771229382049e-06, "loss": 0.65637171, "num_input_tokens_seen": 4043440, "step": 192, "time_per_iteration": 3.0989415645599365 }, { "auxiliary_loss_clip": 0.01377131, "auxiliary_loss_mlp": 0.01065892, "balance_loss_clip": 1.10085821, "balance_loss_mlp": 1.05016804, "epoch": 0.023206877893344555, "flos": 19313444088960.0, "grad_norm": 2.2778660166491176, "language_loss": 0.84306455, "learning_rate": 3.8125345934296324e-06, "loss": 0.86749476, "num_input_tokens_seen": 4061750, "step": 193, "time_per_iteration": 2.6420395374298096 }, { "auxiliary_loss_clip": 0.01378692, "auxiliary_loss_mlp": 0.01071419, "balance_loss_clip": 1.10293484, "balance_loss_mlp": 1.05244112, "epoch": 0.02332712078398365, "flos": 23072968551360.0, "grad_norm": 1.9619199705631425, "language_loss": 0.87975109, "learning_rate": 3.81627850848061e-06, "loss": 0.90425217, "num_input_tokens_seen": 4082345, "step": 194, "time_per_iteration": 2.6344656944274902 }, { "auxiliary_loss_clip": 0.01372866, "auxiliary_loss_mlp": 0.01063915, "balance_loss_clip": 1.10196328, "balance_loss_mlp": 1.04726171, "epoch": 0.02344736367462274, "flos": 24425975485440.0, "grad_norm": 3.598762891656193, "language_loss": 0.8630634, "learning_rate": 3.820003174525994e-06, "loss": 0.88743114, "num_input_tokens_seen": 4101770, "step": 195, "time_per_iteration": 2.6231536865234375 }, { "auxiliary_loss_clip": 0.01376514, "auxiliary_loss_mlp": 0.01071987, "balance_loss_clip": 1.10286188, "balance_loss_mlp": 1.05669236, "epoch": 0.02356760656526183, "flos": 21579803275680.0, "grad_norm": 2.733710872307774, "language_loss": 0.8281129, "learning_rate": 3.823708788487851e-06, "loss": 0.85259789, "num_input_tokens_seen": 4118770, "step": 196, "time_per_iteration": 2.574467420578003 }, { "auxiliary_loss_clip": 0.01375826, "auxiliary_loss_mlp": 0.01067105, "balance_loss_clip": 1.10432994, "balance_loss_mlp": 1.05088079, "epoch": 0.02368784945590092, "flos": 25193602267680.0, "grad_norm": 2.1957784271900853, "language_loss": 0.84350884, "learning_rate": 3.827395544281781e-06, "loss": 0.86793804, "num_input_tokens_seen": 4141110, "step": 197, "time_per_iteration": 4.274305820465088 }, { "auxiliary_loss_clip": 0.01375604, "auxiliary_loss_mlp": 0.01074023, "balance_loss_clip": 1.10240173, "balance_loss_mlp": 1.05778646, "epoch": 0.02380809234654001, "flos": 27562485210240.0, "grad_norm": 1.795656564298872, "language_loss": 0.78819531, "learning_rate": 3.831063632877802e-06, "loss": 0.81269157, "num_input_tokens_seen": 4161430, "step": 198, "time_per_iteration": 2.6613502502441406 }, { "auxiliary_loss_clip": 0.01371409, "auxiliary_loss_mlp": 0.01055308, "balance_loss_clip": 1.10473919, "balance_loss_mlp": 1.04129481, "epoch": 0.0239283352371791, "flos": 18259789826880.0, "grad_norm": 2.6583662632480314, "language_loss": 0.75870717, "learning_rate": 3.834713242359712e-06, "loss": 0.7829743, "num_input_tokens_seen": 4179260, "step": 199, "time_per_iteration": 2.5488381385803223 }, { "auxiliary_loss_clip": 0.01372324, "auxiliary_loss_mlp": 0.01057706, "balance_loss_clip": 1.10016048, "balance_loss_mlp": 1.04127884, "epoch": 0.02404857812781819, "flos": 21395114292000.0, "grad_norm": 1.8535202188135167, "language_loss": 0.87308127, "learning_rate": 3.838344557982959e-06, "loss": 0.89738154, "num_input_tokens_seen": 4200640, "step": 200, "time_per_iteration": 4.208945274353027 }, { "auxiliary_loss_clip": 0.01367807, "auxiliary_loss_mlp": 0.01067888, "balance_loss_clip": 1.09690118, "balance_loss_mlp": 1.05152106, "epoch": 0.024168821018457284, "flos": 16654259548320.0, "grad_norm": 2.599939042116702, "language_loss": 0.84541959, "learning_rate": 3.841957762231063e-06, "loss": 0.86977655, "num_input_tokens_seen": 4218170, "step": 201, "time_per_iteration": 2.578536033630371 }, { "auxiliary_loss_clip": 0.01365547, "auxiliary_loss_mlp": 0.01067895, "balance_loss_clip": 1.09615541, "balance_loss_mlp": 1.05092001, "epoch": 0.024289063909096374, "flos": 22820876210400.0, "grad_norm": 2.0347744887698505, "language_loss": 0.87632012, "learning_rate": 3.8455530348706454e-06, "loss": 0.90065455, "num_input_tokens_seen": 4237770, "step": 202, "time_per_iteration": 2.5910587310791016 }, { "auxiliary_loss_clip": 0.01366466, "auxiliary_loss_mlp": 0.01054098, "balance_loss_clip": 1.10169983, "balance_loss_mlp": 1.03770685, "epoch": 0.024409306799735464, "flos": 17748601337760.0, "grad_norm": 1.8120308078297762, "language_loss": 0.77250642, "learning_rate": 3.849130553005099e-06, "loss": 0.79671204, "num_input_tokens_seen": 4255985, "step": 203, "time_per_iteration": 2.579857110977173 }, { "auxiliary_loss_clip": 0.01369245, "auxiliary_loss_mlp": 0.01060325, "balance_loss_clip": 1.10054171, "balance_loss_mlp": 1.04292083, "epoch": 0.024529549690374557, "flos": 21616216684800.0, "grad_norm": 1.7282506708662098, "language_loss": 0.83658522, "learning_rate": 3.852690491126933e-06, "loss": 0.86088091, "num_input_tokens_seen": 4276035, "step": 204, "time_per_iteration": 2.573577880859375 }, { "auxiliary_loss_clip": 0.01360848, "auxiliary_loss_mlp": 0.01060162, "balance_loss_clip": 1.0956589, "balance_loss_mlp": 1.04448056, "epoch": 0.024649792581013647, "flos": 25551666392640.0, "grad_norm": 4.779830878049353, "language_loss": 0.91114378, "learning_rate": 3.856233021168845e-06, "loss": 0.93535388, "num_input_tokens_seen": 4295730, "step": 205, "time_per_iteration": 2.632028818130493 }, { "auxiliary_loss_clip": 0.01358023, "auxiliary_loss_mlp": 0.01046509, "balance_loss_clip": 1.09680557, "balance_loss_mlp": 1.03193569, "epoch": 0.024770035471652737, "flos": 34495579311840.0, "grad_norm": 2.0794020341152892, "language_loss": 0.91253114, "learning_rate": 3.859758312553544e-06, "loss": 0.93657649, "num_input_tokens_seen": 4317950, "step": 206, "time_per_iteration": 2.674820899963379 }, { "auxiliary_loss_clip": 0.01369152, "auxiliary_loss_mlp": 0.01061001, "balance_loss_clip": 1.10256886, "balance_loss_mlp": 1.04495513, "epoch": 0.02489027836229183, "flos": 21505432028160.0, "grad_norm": 2.5279378550192857, "language_loss": 0.91945827, "learning_rate": 3.8632665322423735e-06, "loss": 0.9437598, "num_input_tokens_seen": 4337605, "step": 207, "time_per_iteration": 2.582911968231201 }, { "auxiliary_loss_clip": 0.01357011, "auxiliary_loss_mlp": 0.01058872, "balance_loss_clip": 1.09697366, "balance_loss_mlp": 1.04227829, "epoch": 0.02501052125293092, "flos": 23219017274400.0, "grad_norm": 1.684792213999402, "language_loss": 0.86071378, "learning_rate": 3.866757844782762e-06, "loss": 0.88487262, "num_input_tokens_seen": 4358110, "step": 208, "time_per_iteration": 2.614161729812622 }, { "auxiliary_loss_clip": 0.01363808, "auxiliary_loss_mlp": 0.01048292, "balance_loss_clip": 1.09844518, "balance_loss_mlp": 1.03271091, "epoch": 0.02513076414357001, "flos": 26388923383680.0, "grad_norm": 2.3252179162968214, "language_loss": 0.91417867, "learning_rate": 3.870232412354527e-06, "loss": 0.93829966, "num_input_tokens_seen": 4374955, "step": 209, "time_per_iteration": 2.6197807788848877 }, { "auxiliary_loss_clip": 0.01365073, "auxiliary_loss_mlp": 0.01056367, "balance_loss_clip": 1.09812498, "balance_loss_mlp": 1.04079843, "epoch": 0.025251007034209103, "flos": 13590432973920.0, "grad_norm": 2.2439028169074255, "language_loss": 0.9226889, "learning_rate": 3.873690394815086e-06, "loss": 0.94690335, "num_input_tokens_seen": 4391535, "step": 210, "time_per_iteration": 2.5468456745147705 }, { "auxiliary_loss_clip": 0.01362621, "auxiliary_loss_mlp": 0.01054862, "balance_loss_clip": 1.09863734, "balance_loss_mlp": 1.03915024, "epoch": 0.025371249924848193, "flos": 15049232107200.0, "grad_norm": 2.3168715707955885, "language_loss": 0.91074204, "learning_rate": 3.877131949743587e-06, "loss": 0.93491685, "num_input_tokens_seen": 4408400, "step": 211, "time_per_iteration": 2.5799362659454346 }, { "auxiliary_loss_clip": 0.01362727, "auxiliary_loss_mlp": 0.01065511, "balance_loss_clip": 1.09853232, "balance_loss_mlp": 1.04889333, "epoch": 0.025491492815487283, "flos": 25553857327200.0, "grad_norm": 2.476796077434094, "language_loss": 0.77864397, "learning_rate": 3.880557232483993e-06, "loss": 0.8029263, "num_input_tokens_seen": 4427840, "step": 212, "time_per_iteration": 2.616962432861328 }, { "auxiliary_loss_clip": 0.01355513, "auxiliary_loss_mlp": 0.01053583, "balance_loss_clip": 1.09387696, "balance_loss_mlp": 1.03660774, "epoch": 0.025611735706126376, "flos": 20630755953120.0, "grad_norm": 1.9785756345177155, "language_loss": 0.868864, "learning_rate": 3.883966396187164e-06, "loss": 0.89295495, "num_input_tokens_seen": 4447110, "step": 213, "time_per_iteration": 2.5845115184783936 }, { "auxiliary_loss_clip": 0.01360507, "auxiliary_loss_mlp": 0.01068426, "balance_loss_clip": 1.09920037, "balance_loss_mlp": 1.05369186, "epoch": 0.025731978596765466, "flos": 19062285588960.0, "grad_norm": 2.152418026267152, "language_loss": 0.90050626, "learning_rate": 3.887359591851937e-06, "loss": 0.92479563, "num_input_tokens_seen": 4464715, "step": 214, "time_per_iteration": 2.552445411682129 }, { "auxiliary_loss_clip": 0.01355835, "auxiliary_loss_mlp": 0.01058818, "balance_loss_clip": 1.09696031, "balance_loss_mlp": 1.04347539, "epoch": 0.025852221487404556, "flos": 22163818583040.0, "grad_norm": 1.9534735686564537, "language_loss": 0.9227016, "learning_rate": 3.890736968365265e-06, "loss": 0.94684809, "num_input_tokens_seen": 4485030, "step": 215, "time_per_iteration": 2.6023969650268555 }, { "auxiliary_loss_clip": 0.013534, "auxiliary_loss_mlp": 0.01060305, "balance_loss_clip": 1.09498584, "balance_loss_mlp": 1.04445004, "epoch": 0.02597246437804365, "flos": 26541975913920.0, "grad_norm": 2.0777583913928845, "language_loss": 0.85151511, "learning_rate": 3.894098672541412e-06, "loss": 0.87565219, "num_input_tokens_seen": 4505935, "step": 216, "time_per_iteration": 2.605844259262085 }, { "auxiliary_loss_clip": 0.01357169, "auxiliary_loss_mlp": 0.0104998, "balance_loss_clip": 1.09848309, "balance_loss_mlp": 1.03499532, "epoch": 0.02609270726868274, "flos": 32671676329440.0, "grad_norm": 1.787545449990298, "language_loss": 0.75492692, "learning_rate": 3.89744484916025e-06, "loss": 0.77899837, "num_input_tokens_seen": 4527045, "step": 217, "time_per_iteration": 2.6540017127990723 }, { "auxiliary_loss_clip": 0.01362328, "auxiliary_loss_mlp": 0.01043567, "balance_loss_clip": 1.09840631, "balance_loss_mlp": 1.02750981, "epoch": 0.02621295015932183, "flos": 26243557082880.0, "grad_norm": 1.8904955512092891, "language_loss": 0.87431157, "learning_rate": 3.900775641004673e-06, "loss": 0.8983705, "num_input_tokens_seen": 4546360, "step": 218, "time_per_iteration": 2.63987135887146 }, { "auxiliary_loss_clip": 0.01365716, "auxiliary_loss_mlp": 0.0105415, "balance_loss_clip": 1.10144734, "balance_loss_mlp": 1.03694797, "epoch": 0.026333193049960922, "flos": 42921407685600.0, "grad_norm": 2.6095231863755584, "language_loss": 0.73754573, "learning_rate": 3.904091188897156e-06, "loss": 0.76174438, "num_input_tokens_seen": 4565495, "step": 219, "time_per_iteration": 2.724677085876465 }, { "auxiliary_loss_clip": 0.01355488, "auxiliary_loss_mlp": 0.01063514, "balance_loss_clip": 1.09622681, "balance_loss_mlp": 1.04690194, "epoch": 0.026453435940600012, "flos": 17963849266080.0, "grad_norm": 2.1682768123449727, "language_loss": 0.81898904, "learning_rate": 3.90739163173548e-06, "loss": 0.84317911, "num_input_tokens_seen": 4583330, "step": 220, "time_per_iteration": 2.5970332622528076 }, { "auxiliary_loss_clip": 0.01352187, "auxiliary_loss_mlp": 0.01047766, "balance_loss_clip": 1.09347153, "balance_loss_mlp": 1.03254294, "epoch": 0.026573678831239102, "flos": 18984322645440.0, "grad_norm": 2.273324807548867, "language_loss": 0.8836683, "learning_rate": 3.910677106527646e-06, "loss": 0.90766788, "num_input_tokens_seen": 4600520, "step": 221, "time_per_iteration": 2.533687114715576 }, { "auxiliary_loss_clip": 0.01355207, "auxiliary_loss_mlp": 0.01068299, "balance_loss_clip": 1.09825087, "balance_loss_mlp": 1.05289698, "epoch": 0.026693921721878195, "flos": 29241452895360.0, "grad_norm": 3.111810321211162, "language_loss": 0.8409102, "learning_rate": 3.913947748426004e-06, "loss": 0.86514533, "num_input_tokens_seen": 4617340, "step": 222, "time_per_iteration": 2.640655994415283 }, { "auxiliary_loss_clip": 0.01356096, "auxiliary_loss_mlp": 0.0105954, "balance_loss_clip": 1.09992003, "balance_loss_mlp": 1.04508042, "epoch": 0.026814164612517285, "flos": 14128085874240.0, "grad_norm": 2.412855659226124, "language_loss": 0.76626682, "learning_rate": 3.9172036907606136e-06, "loss": 0.79042315, "num_input_tokens_seen": 4630820, "step": 223, "time_per_iteration": 3.3805618286132812 }, { "auxiliary_loss_clip": 0.01351296, "auxiliary_loss_mlp": 0.0106135, "balance_loss_clip": 1.09339416, "balance_loss_mlp": 1.04623985, "epoch": 0.026934407503156375, "flos": 23511976727520.0, "grad_norm": 1.7911201198074327, "language_loss": 0.95055044, "learning_rate": 3.920445065071855e-06, "loss": 0.97467691, "num_input_tokens_seen": 4651985, "step": 224, "time_per_iteration": 3.3325040340423584 }, { "auxiliary_loss_clip": 0.01349723, "auxiliary_loss_mlp": 0.0105228, "balance_loss_clip": 1.09443724, "balance_loss_mlp": 1.03748608, "epoch": 0.027054650393795468, "flos": 28950361124160.0, "grad_norm": 2.1906252487487077, "language_loss": 0.79793549, "learning_rate": 3.923672001142322e-06, "loss": 0.82195544, "num_input_tokens_seen": 4672295, "step": 225, "time_per_iteration": 2.643852472305298 }, { "auxiliary_loss_clip": 0.01349144, "auxiliary_loss_mlp": 0.01064321, "balance_loss_clip": 1.09504509, "balance_loss_mlp": 1.04949188, "epoch": 0.027174893284434558, "flos": 31431573152640.0, "grad_norm": 1.927269258478074, "language_loss": 0.84484196, "learning_rate": 3.926884627027996e-06, "loss": 0.86897659, "num_input_tokens_seen": 4696065, "step": 226, "time_per_iteration": 3.5281753540039062 }, { "auxiliary_loss_clip": 0.01347494, "auxiliary_loss_mlp": 0.01062161, "balance_loss_clip": 1.09428871, "balance_loss_mlp": 1.04646111, "epoch": 0.027295136175073648, "flos": 22054470604800.0, "grad_norm": 2.1809657545530214, "language_loss": 0.77358949, "learning_rate": 3.930083069088744e-06, "loss": 0.79768604, "num_input_tokens_seen": 4716065, "step": 227, "time_per_iteration": 2.5985074043273926 }, { "auxiliary_loss_clip": 0.01358175, "auxiliary_loss_mlp": 0.01017318, "balance_loss_clip": 1.16196585, "balance_loss_mlp": 1.00902104, "epoch": 0.02741537906571274, "flos": 60800763509760.0, "grad_norm": 0.9860957981806727, "language_loss": 0.59330869, "learning_rate": 3.933267452018137e-06, "loss": 0.61706364, "num_input_tokens_seen": 4775860, "step": 228, "time_per_iteration": 3.152433395385742 }, { "auxiliary_loss_clip": 0.01347321, "auxiliary_loss_mlp": 0.01055659, "balance_loss_clip": 1.09563673, "balance_loss_mlp": 1.04073381, "epoch": 0.02753562195635183, "flos": 24606282600000.0, "grad_norm": 2.077358190837518, "language_loss": 0.84453356, "learning_rate": 3.936437898872622e-06, "loss": 0.86856341, "num_input_tokens_seen": 4795835, "step": 229, "time_per_iteration": 2.612440824508667 }, { "auxiliary_loss_clip": 0.01344854, "auxiliary_loss_mlp": 0.01050965, "balance_loss_clip": 1.09229493, "balance_loss_mlp": 1.03626692, "epoch": 0.02765586484699092, "flos": 34094241638400.0, "grad_norm": 2.4007894792106943, "language_loss": 0.79605234, "learning_rate": 3.9395945311000525e-06, "loss": 0.8200106, "num_input_tokens_seen": 4817460, "step": 230, "time_per_iteration": 2.6816673278808594 }, { "auxiliary_loss_clip": 0.01347096, "auxiliary_loss_mlp": 0.01066617, "balance_loss_clip": 1.09427297, "balance_loss_mlp": 1.05173934, "epoch": 0.027776107737630014, "flos": 14829925562400.0, "grad_norm": 2.0789448077868813, "language_loss": 0.90668631, "learning_rate": 3.942737468567608e-06, "loss": 0.93082345, "num_input_tokens_seen": 4835475, "step": 231, "time_per_iteration": 2.5630500316619873 }, { "auxiliary_loss_clip": 0.01344198, "auxiliary_loss_mlp": 0.01050784, "balance_loss_clip": 1.09092855, "balance_loss_mlp": 1.03661036, "epoch": 0.027896350628269104, "flos": 47920352902560.0, "grad_norm": 1.90949085897355, "language_loss": 0.85986048, "learning_rate": 3.9458668295891026e-06, "loss": 0.88381034, "num_input_tokens_seen": 4857760, "step": 232, "time_per_iteration": 2.8108010292053223 }, { "auxiliary_loss_clip": 0.01344327, "auxiliary_loss_mlp": 0.01059405, "balance_loss_clip": 1.09120488, "balance_loss_mlp": 1.04401529, "epoch": 0.028016593518908194, "flos": 21684553883040.0, "grad_norm": 2.5228843415779623, "language_loss": 0.86631322, "learning_rate": 3.948982730951712e-06, "loss": 0.89035058, "num_input_tokens_seen": 4875855, "step": 233, "time_per_iteration": 2.5683093070983887 }, { "auxiliary_loss_clip": 0.01352997, "auxiliary_loss_mlp": 0.01058921, "balance_loss_clip": 1.09606528, "balance_loss_mlp": 1.04325724, "epoch": 0.028136836409547287, "flos": 18439486353120.0, "grad_norm": 2.8165918796227247, "language_loss": 0.81852591, "learning_rate": 3.9520852879421254e-06, "loss": 0.84264517, "num_input_tokens_seen": 4893200, "step": 234, "time_per_iteration": 2.596571207046509 }, { "auxiliary_loss_clip": 0.01345336, "auxiliary_loss_mlp": 0.0105625, "balance_loss_clip": 1.09382761, "balance_loss_mlp": 1.0414561, "epoch": 0.028257079300186377, "flos": 31576939453440.0, "grad_norm": 2.23296594993644, "language_loss": 0.81809741, "learning_rate": 3.955174614372137e-06, "loss": 0.84211326, "num_input_tokens_seen": 4912965, "step": 235, "time_per_iteration": 2.632941961288452 }, { "auxiliary_loss_clip": 0.01344387, "auxiliary_loss_mlp": 0.01060575, "balance_loss_clip": 1.09578133, "balance_loss_mlp": 1.04648423, "epoch": 0.028377322190825467, "flos": 23513341572000.0, "grad_norm": 2.6136863101417314, "language_loss": 0.84448093, "learning_rate": 3.9582508226037045e-06, "loss": 0.86853051, "num_input_tokens_seen": 4933105, "step": 236, "time_per_iteration": 2.608247995376587 }, { "auxiliary_loss_clip": 0.01347809, "auxiliary_loss_mlp": 0.01045075, "balance_loss_clip": 1.0955739, "balance_loss_mlp": 1.02926767, "epoch": 0.02849756508146456, "flos": 20479607021760.0, "grad_norm": 2.68252277461749, "language_loss": 0.94106781, "learning_rate": 3.9613140235734636e-06, "loss": 0.9649967, "num_input_tokens_seen": 4950085, "step": 237, "time_per_iteration": 2.5397636890411377 }, { "auxiliary_loss_clip": 0.01344215, "auxiliary_loss_mlp": 0.01063329, "balance_loss_clip": 1.09256864, "balance_loss_mlp": 1.0474031, "epoch": 0.02861780797210365, "flos": 14283365256000.0, "grad_norm": 1.886812930991703, "language_loss": 0.81383514, "learning_rate": 3.96436432681674e-06, "loss": 0.83791053, "num_input_tokens_seen": 4968075, "step": 238, "time_per_iteration": 2.572567939758301 }, { "auxiliary_loss_clip": 0.01344095, "auxiliary_loss_mlp": 0.01059082, "balance_loss_clip": 1.09305167, "balance_loss_mlp": 1.04426432, "epoch": 0.02873805086274274, "flos": 25808535689280.0, "grad_norm": 2.311624750892354, "language_loss": 0.89233696, "learning_rate": 3.967401840491044e-06, "loss": 0.91636872, "num_input_tokens_seen": 4987355, "step": 239, "time_per_iteration": 2.5965189933776855 }, { "auxiliary_loss_clip": 0.01343532, "auxiliary_loss_mlp": 0.01050507, "balance_loss_clip": 1.09440994, "balance_loss_mlp": 1.03654814, "epoch": 0.028858293753381833, "flos": 17304241534560.0, "grad_norm": 2.1364606092848053, "language_loss": 0.87358564, "learning_rate": 3.97042667139909e-06, "loss": 0.89752603, "num_input_tokens_seen": 5004680, "step": 240, "time_per_iteration": 2.5685231685638428 }, { "auxiliary_loss_clip": 0.01338994, "auxiliary_loss_mlp": 0.01043707, "balance_loss_clip": 1.09245384, "balance_loss_mlp": 1.02769673, "epoch": 0.028978536644020923, "flos": 23038350990240.0, "grad_norm": 2.0362940667554277, "language_loss": 0.87297404, "learning_rate": 3.973438925011327e-06, "loss": 0.89680111, "num_input_tokens_seen": 5022965, "step": 241, "time_per_iteration": 2.5619876384735107 }, { "auxiliary_loss_clip": 0.01341481, "auxiliary_loss_mlp": 0.0105179, "balance_loss_clip": 1.0903759, "balance_loss_mlp": 1.03637052, "epoch": 0.029098779534660012, "flos": 28329716905920.0, "grad_norm": 6.432925234558297, "language_loss": 0.91225213, "learning_rate": 3.976438705488002e-06, "loss": 0.93618482, "num_input_tokens_seen": 5042625, "step": 242, "time_per_iteration": 2.6342198848724365 }, { "auxiliary_loss_clip": 0.01344022, "auxiliary_loss_mlp": 0.01055735, "balance_loss_clip": 1.0950315, "balance_loss_mlp": 1.04073822, "epoch": 0.029219022425299106, "flos": 13881668412960.0, "grad_norm": 2.454339335372096, "language_loss": 0.93099618, "learning_rate": 3.9794261157007744e-06, "loss": 0.95499372, "num_input_tokens_seen": 5060380, "step": 243, "time_per_iteration": 2.5276734828948975 }, { "auxiliary_loss_clip": 0.01350409, "auxiliary_loss_mlp": 0.01056414, "balance_loss_clip": 1.09665811, "balance_loss_mlp": 1.03992724, "epoch": 0.029339265315938196, "flos": 19422504731520.0, "grad_norm": 3.1898288403707395, "language_loss": 0.84736741, "learning_rate": 3.982401257253887e-06, "loss": 0.87143564, "num_input_tokens_seen": 5078720, "step": 244, "time_per_iteration": 2.575969696044922 }, { "auxiliary_loss_clip": 0.01342591, "auxiliary_loss_mlp": 0.0104689, "balance_loss_clip": 1.09482539, "balance_loss_mlp": 1.03270364, "epoch": 0.029459508206577285, "flos": 15669553072800.0, "grad_norm": 2.4375126957878956, "language_loss": 0.89761198, "learning_rate": 3.985364230504893e-06, "loss": 0.92150676, "num_input_tokens_seen": 5096605, "step": 245, "time_per_iteration": 2.527632474899292 }, { "auxiliary_loss_clip": 0.01345548, "auxiliary_loss_mlp": 0.01058294, "balance_loss_clip": 1.0943135, "balance_loss_mlp": 1.04354167, "epoch": 0.02957975109721638, "flos": 28220979516000.0, "grad_norm": 2.880130603321742, "language_loss": 0.84103978, "learning_rate": 3.988315134584976e-06, "loss": 0.86507821, "num_input_tokens_seen": 5116285, "step": 246, "time_per_iteration": 2.6195855140686035 }, { "auxiliary_loss_clip": 0.01344047, "auxiliary_loss_mlp": 0.01051255, "balance_loss_clip": 1.09442711, "balance_loss_mlp": 1.03593683, "epoch": 0.02969999398785547, "flos": 24315873251040.0, "grad_norm": 2.0367077751448814, "language_loss": 0.80371135, "learning_rate": 3.991254067418851e-06, "loss": 0.82766432, "num_input_tokens_seen": 5136825, "step": 247, "time_per_iteration": 2.584259271621704 }, { "auxiliary_loss_clip": 0.01339338, "auxiliary_loss_mlp": 0.01056257, "balance_loss_clip": 1.09322238, "balance_loss_mlp": 1.04047322, "epoch": 0.02982023687849456, "flos": 35078588944320.0, "grad_norm": 2.165560060754702, "language_loss": 0.82840592, "learning_rate": 3.994181125744254e-06, "loss": 0.8523618, "num_input_tokens_seen": 5158630, "step": 248, "time_per_iteration": 2.697114944458008 }, { "auxiliary_loss_clip": 0.01339507, "auxiliary_loss_mlp": 0.01054683, "balance_loss_clip": 1.09188008, "balance_loss_mlp": 1.04049683, "epoch": 0.02994047976913365, "flos": 26177159400480.0, "grad_norm": 2.5827070427231993, "language_loss": 0.738428, "learning_rate": 3.99709640513106e-06, "loss": 0.76236987, "num_input_tokens_seen": 5179510, "step": 249, "time_per_iteration": 2.588961362838745 }, { "auxiliary_loss_clip": 0.01347342, "auxiliary_loss_mlp": 0.01048609, "balance_loss_clip": 1.09324825, "balance_loss_mlp": 1.03320718, "epoch": 0.03006072265977274, "flos": 25625031965280.0, "grad_norm": 2.938669847459281, "language_loss": 0.85622048, "learning_rate": 4e-06, "loss": 0.88018012, "num_input_tokens_seen": 5199345, "step": 250, "time_per_iteration": 3.317899703979492 }, { "auxiliary_loss_clip": 0.01341293, "auxiliary_loss_mlp": 0.01060272, "balance_loss_clip": 1.09463918, "balance_loss_mlp": 1.04574001, "epoch": 0.03018096555041183, "flos": 22127081921280.0, "grad_norm": 14.686570434665605, "language_loss": 0.88434184, "learning_rate": 3.999999848300794e-06, "loss": 0.90835744, "num_input_tokens_seen": 5218330, "step": 251, "time_per_iteration": 3.4295175075531006 }, { "auxiliary_loss_clip": 0.01333278, "auxiliary_loss_mlp": 0.01053757, "balance_loss_clip": 1.08767116, "balance_loss_mlp": 1.03852224, "epoch": 0.030301208441050925, "flos": 30188201532480.0, "grad_norm": 1.6236892132495664, "language_loss": 0.8919003, "learning_rate": 3.999999393203203e-06, "loss": 0.91577065, "num_input_tokens_seen": 5240740, "step": 252, "time_per_iteration": 3.3641209602355957 }, { "auxiliary_loss_clip": 0.01338537, "auxiliary_loss_mlp": 0.01046156, "balance_loss_clip": 1.09178591, "balance_loss_mlp": 1.03231621, "epoch": 0.030421451331690014, "flos": 23621396539680.0, "grad_norm": 1.9461189008075146, "language_loss": 0.84946287, "learning_rate": 3.999998634707293e-06, "loss": 0.87330985, "num_input_tokens_seen": 5260290, "step": 253, "time_per_iteration": 2.604768753051758 }, { "auxiliary_loss_clip": 0.01343177, "auxiliary_loss_mlp": 0.0106094, "balance_loss_clip": 1.09565842, "balance_loss_mlp": 1.04527617, "epoch": 0.030541694222329104, "flos": 27928451066400.0, "grad_norm": 2.6224456549413815, "language_loss": 0.96596724, "learning_rate": 3.999997572813182e-06, "loss": 0.99000835, "num_input_tokens_seen": 5278100, "step": 254, "time_per_iteration": 3.5222203731536865 }, { "auxiliary_loss_clip": 0.01334877, "auxiliary_loss_mlp": 0.01054184, "balance_loss_clip": 1.09128785, "balance_loss_mlp": 1.0387702, "epoch": 0.030661937112968194, "flos": 18588444349920.0, "grad_norm": 2.017560495692821, "language_loss": 0.87322581, "learning_rate": 3.999996207521028e-06, "loss": 0.89711636, "num_input_tokens_seen": 5296810, "step": 255, "time_per_iteration": 2.528486490249634 }, { "auxiliary_loss_clip": 0.01339047, "auxiliary_loss_mlp": 0.01054655, "balance_loss_clip": 1.08871436, "balance_loss_mlp": 1.0382638, "epoch": 0.030782180003607287, "flos": 12969142250400.0, "grad_norm": 2.8515955788887974, "language_loss": 0.82103717, "learning_rate": 3.999994538831039e-06, "loss": 0.84497422, "num_input_tokens_seen": 5313395, "step": 256, "time_per_iteration": 2.562865734100342 }, { "auxiliary_loss_clip": 0.01337895, "auxiliary_loss_mlp": 0.01061803, "balance_loss_clip": 1.08966267, "balance_loss_mlp": 1.04718769, "epoch": 0.030902422894246377, "flos": 23335369059840.0, "grad_norm": 2.211974469376792, "language_loss": 0.85736763, "learning_rate": 3.99999256674347e-06, "loss": 0.88136458, "num_input_tokens_seen": 5333545, "step": 257, "time_per_iteration": 2.580204963684082 }, { "auxiliary_loss_clip": 0.01319032, "auxiliary_loss_mlp": 0.01017056, "balance_loss_clip": 1.12708879, "balance_loss_mlp": 1.01042783, "epoch": 0.031022665784885467, "flos": 55094160057120.0, "grad_norm": 1.0085002493745936, "language_loss": 0.5351733, "learning_rate": 3.999990291258618e-06, "loss": 0.5585342, "num_input_tokens_seen": 5392235, "step": 258, "time_per_iteration": 3.1626622676849365 }, { "auxiliary_loss_clip": 0.01340217, "auxiliary_loss_mlp": 0.01049584, "balance_loss_clip": 1.09430516, "balance_loss_mlp": 1.03512406, "epoch": 0.03114290867552456, "flos": 19317790041120.0, "grad_norm": 3.26391166893829, "language_loss": 0.86781728, "learning_rate": 3.999987712376829e-06, "loss": 0.89171529, "num_input_tokens_seen": 5410555, "step": 259, "time_per_iteration": 2.536162853240967 }, { "auxiliary_loss_clip": 0.01331775, "auxiliary_loss_mlp": 0.01057856, "balance_loss_clip": 1.09009373, "balance_loss_mlp": 1.04377699, "epoch": 0.031263151566163654, "flos": 20959446393120.0, "grad_norm": 1.9327316233475564, "language_loss": 0.82171673, "learning_rate": 3.999984830098494e-06, "loss": 0.845613, "num_input_tokens_seen": 5430135, "step": 260, "time_per_iteration": 2.566650629043579 }, { "auxiliary_loss_clip": 0.01333981, "auxiliary_loss_mlp": 0.01056866, "balance_loss_clip": 1.08848321, "balance_loss_mlp": 1.0413692, "epoch": 0.03138339445680274, "flos": 14793009315840.0, "grad_norm": 2.682483681480099, "language_loss": 0.97892213, "learning_rate": 3.999981644424051e-06, "loss": 1.00283062, "num_input_tokens_seen": 5444935, "step": 261, "time_per_iteration": 2.475956916809082 }, { "auxiliary_loss_clip": 0.01335056, "auxiliary_loss_mlp": 0.01054583, "balance_loss_clip": 1.08947372, "balance_loss_mlp": 1.03999209, "epoch": 0.03150363734744183, "flos": 11655601667040.0, "grad_norm": 2.303859246698878, "language_loss": 0.86215049, "learning_rate": 3.999978155353982e-06, "loss": 0.88604689, "num_input_tokens_seen": 5462080, "step": 262, "time_per_iteration": 2.503742218017578 }, { "auxiliary_loss_clip": 0.01331612, "auxiliary_loss_mlp": 0.01052084, "balance_loss_clip": 1.08667374, "balance_loss_mlp": 1.03690863, "epoch": 0.03162388023808092, "flos": 33727737027840.0, "grad_norm": 2.104954752886585, "language_loss": 0.80249637, "learning_rate": 3.9999743628888186e-06, "loss": 0.82633334, "num_input_tokens_seen": 5483870, "step": 263, "time_per_iteration": 2.664907932281494 }, { "auxiliary_loss_clip": 0.01334697, "auxiliary_loss_mlp": 0.01052372, "balance_loss_clip": 1.08753538, "balance_loss_mlp": 1.03716707, "epoch": 0.03174412312872001, "flos": 20810955316800.0, "grad_norm": 2.515726644194384, "language_loss": 0.89533013, "learning_rate": 3.999970267029133e-06, "loss": 0.91920078, "num_input_tokens_seen": 5502830, "step": 264, "time_per_iteration": 2.5942656993865967 }, { "auxiliary_loss_clip": 0.0132975, "auxiliary_loss_mlp": 0.01057586, "balance_loss_clip": 1.08735919, "balance_loss_mlp": 1.04405582, "epoch": 0.0318643660193591, "flos": 23727942995040.0, "grad_norm": 2.385073205044612, "language_loss": 0.80023968, "learning_rate": 3.999965867775548e-06, "loss": 0.82411307, "num_input_tokens_seen": 5523225, "step": 265, "time_per_iteration": 2.573023796081543 }, { "auxiliary_loss_clip": 0.01333364, "auxiliary_loss_mlp": 0.01050322, "balance_loss_clip": 1.09027028, "balance_loss_mlp": 1.03475308, "epoch": 0.0319846089099982, "flos": 13917866320320.0, "grad_norm": 2.580439984313093, "language_loss": 0.86605155, "learning_rate": 3.9999611651287315e-06, "loss": 0.88988841, "num_input_tokens_seen": 5541380, "step": 266, "time_per_iteration": 2.5627317428588867 }, { "auxiliary_loss_clip": 0.01337728, "auxiliary_loss_mlp": 0.0105392, "balance_loss_clip": 1.09250617, "balance_loss_mlp": 1.03874505, "epoch": 0.03210485180063729, "flos": 14753255629440.0, "grad_norm": 2.5594297722237243, "language_loss": 0.78583181, "learning_rate": 3.999956159089396e-06, "loss": 0.80974823, "num_input_tokens_seen": 5558830, "step": 267, "time_per_iteration": 2.526312828063965 }, { "auxiliary_loss_clip": 0.01330864, "auxiliary_loss_mlp": 0.01053683, "balance_loss_clip": 1.08964813, "balance_loss_mlp": 1.03937221, "epoch": 0.03222509469127638, "flos": 28913157541920.0, "grad_norm": 1.9962759762663729, "language_loss": 0.79350185, "learning_rate": 3.999950849658302e-06, "loss": 0.81734729, "num_input_tokens_seen": 5577750, "step": 268, "time_per_iteration": 2.6132254600524902 }, { "auxiliary_loss_clip": 0.01332274, "auxiliary_loss_mlp": 0.01054142, "balance_loss_clip": 1.08986926, "balance_loss_mlp": 1.0381974, "epoch": 0.03234533758191547, "flos": 16946392911360.0, "grad_norm": 2.9271031421670015, "language_loss": 0.84061694, "learning_rate": 3.999945236836254e-06, "loss": 0.86448109, "num_input_tokens_seen": 5596715, "step": 269, "time_per_iteration": 2.532719612121582 }, { "auxiliary_loss_clip": 0.01341671, "auxiliary_loss_mlp": 0.01056203, "balance_loss_clip": 1.09314489, "balance_loss_mlp": 1.03973961, "epoch": 0.03246558047255456, "flos": 18989099601120.0, "grad_norm": 2.773696936557044, "language_loss": 0.9479152, "learning_rate": 3.999939320624103e-06, "loss": 0.97189397, "num_input_tokens_seen": 5611865, "step": 270, "time_per_iteration": 2.5447282791137695 }, { "auxiliary_loss_clip": 0.01333769, "auxiliary_loss_mlp": 0.01051533, "balance_loss_clip": 1.08976841, "balance_loss_mlp": 1.03527272, "epoch": 0.03258582336319365, "flos": 23728338081600.0, "grad_norm": 2.063090449357795, "language_loss": 0.90228856, "learning_rate": 3.999933101022749e-06, "loss": 0.92614156, "num_input_tokens_seen": 5632270, "step": 271, "time_per_iteration": 2.5779366493225098 }, { "auxiliary_loss_clip": 0.01331496, "auxiliary_loss_mlp": 0.01058991, "balance_loss_clip": 1.09013295, "balance_loss_mlp": 1.04388738, "epoch": 0.032706066253832745, "flos": 27670827513600.0, "grad_norm": 1.9141109710699775, "language_loss": 0.86569273, "learning_rate": 3.999926578033132e-06, "loss": 0.88959765, "num_input_tokens_seen": 5652085, "step": 272, "time_per_iteration": 2.632751703262329 }, { "auxiliary_loss_clip": 0.01335819, "auxiliary_loss_mlp": 0.01053983, "balance_loss_clip": 1.08900654, "balance_loss_mlp": 1.03804445, "epoch": 0.032826309144471835, "flos": 45624691864800.0, "grad_norm": 2.1081852535814414, "language_loss": 0.63075566, "learning_rate": 3.999919751656244e-06, "loss": 0.65465367, "num_input_tokens_seen": 5678985, "step": 273, "time_per_iteration": 2.7764742374420166 }, { "auxiliary_loss_clip": 0.01331948, "auxiliary_loss_mlp": 0.0105113, "balance_loss_clip": 1.08816051, "balance_loss_mlp": 1.0373019, "epoch": 0.032946552035110925, "flos": 25812378804000.0, "grad_norm": 2.466115106123014, "language_loss": 0.75858039, "learning_rate": 3.9999126218931195e-06, "loss": 0.78241122, "num_input_tokens_seen": 5697020, "step": 274, "time_per_iteration": 2.608823299407959 }, { "auxiliary_loss_clip": 0.01335257, "auxiliary_loss_mlp": 0.01049669, "balance_loss_clip": 1.09156585, "balance_loss_mlp": 1.03418946, "epoch": 0.033066794925750015, "flos": 15121987091520.0, "grad_norm": 2.2962386832577537, "language_loss": 0.89339924, "learning_rate": 3.99990518874484e-06, "loss": 0.91724855, "num_input_tokens_seen": 5713460, "step": 275, "time_per_iteration": 2.5625574588775635 }, { "auxiliary_loss_clip": 0.01327991, "auxiliary_loss_mlp": 0.01050926, "balance_loss_clip": 1.08870745, "balance_loss_mlp": 1.03536892, "epoch": 0.033187037816389105, "flos": 22776596987040.0, "grad_norm": 2.583743357532601, "language_loss": 0.92501748, "learning_rate": 3.999897452212534e-06, "loss": 0.94880664, "num_input_tokens_seen": 5730790, "step": 276, "time_per_iteration": 3.2961225509643555 }, { "auxiliary_loss_clip": 0.01330364, "auxiliary_loss_mlp": 0.01051683, "balance_loss_clip": 1.08774066, "balance_loss_mlp": 1.03694916, "epoch": 0.033307280707028195, "flos": 23331418194240.0, "grad_norm": 3.001788982515616, "language_loss": 1.0008595, "learning_rate": 3.999889412297374e-06, "loss": 1.0246799, "num_input_tokens_seen": 5750215, "step": 277, "time_per_iteration": 3.4553730487823486 }, { "auxiliary_loss_clip": 0.01329346, "auxiliary_loss_mlp": 0.01046937, "balance_loss_clip": 1.08676732, "balance_loss_mlp": 1.0316782, "epoch": 0.03342752359766729, "flos": 28840294806720.0, "grad_norm": 2.7776798411590438, "language_loss": 0.78840601, "learning_rate": 3.999881069000581e-06, "loss": 0.81216884, "num_input_tokens_seen": 5769945, "step": 278, "time_per_iteration": 2.631734848022461 }, { "auxiliary_loss_clip": 0.01328855, "auxiliary_loss_mlp": 0.01049886, "balance_loss_clip": 1.08777094, "balance_loss_mlp": 1.03485382, "epoch": 0.03354776648830638, "flos": 19384546893120.0, "grad_norm": 2.5101153759041757, "language_loss": 0.8650347, "learning_rate": 3.99987242232342e-06, "loss": 0.88882208, "num_input_tokens_seen": 5784950, "step": 279, "time_per_iteration": 3.2812814712524414 }, { "auxiliary_loss_clip": 0.01333179, "auxiliary_loss_mlp": 0.01051088, "balance_loss_clip": 1.09157217, "balance_loss_mlp": 1.03542399, "epoch": 0.03366800937894547, "flos": 17858631738240.0, "grad_norm": 2.254038143911953, "language_loss": 0.79568291, "learning_rate": 3.9998634722672026e-06, "loss": 0.8195256, "num_input_tokens_seen": 5805005, "step": 280, "time_per_iteration": 3.4432320594787598 }, { "auxiliary_loss_clip": 0.01332911, "auxiliary_loss_mlp": 0.01058852, "balance_loss_clip": 1.09242368, "balance_loss_mlp": 1.04303312, "epoch": 0.03378825226958456, "flos": 35951038167840.0, "grad_norm": 4.893566849296687, "language_loss": 0.78525394, "learning_rate": 3.999854218833286e-06, "loss": 0.80917156, "num_input_tokens_seen": 5825825, "step": 281, "time_per_iteration": 2.663093090057373 }, { "auxiliary_loss_clip": 0.01332524, "auxiliary_loss_mlp": 0.01056836, "balance_loss_clip": 1.09096372, "balance_loss_mlp": 1.04298377, "epoch": 0.03390849516022365, "flos": 25702492071360.0, "grad_norm": 2.0394395039609754, "language_loss": 0.82072324, "learning_rate": 3.999844662023075e-06, "loss": 0.84461683, "num_input_tokens_seen": 5845700, "step": 282, "time_per_iteration": 2.60213041305542 }, { "auxiliary_loss_clip": 0.01320281, "auxiliary_loss_mlp": 0.0105668, "balance_loss_clip": 1.08453369, "balance_loss_mlp": 1.04251242, "epoch": 0.03402873805086274, "flos": 21284509220160.0, "grad_norm": 1.9479703392297816, "language_loss": 0.92116135, "learning_rate": 3.999834801838018e-06, "loss": 0.94493103, "num_input_tokens_seen": 5864680, "step": 283, "time_per_iteration": 2.5602307319641113 }, { "auxiliary_loss_clip": 0.01324101, "auxiliary_loss_mlp": 0.01061932, "balance_loss_clip": 1.08840871, "balance_loss_mlp": 1.04724598, "epoch": 0.03414898094150183, "flos": 22710917643840.0, "grad_norm": 4.100060207418749, "language_loss": 0.74089324, "learning_rate": 3.9998246382796115e-06, "loss": 0.76475358, "num_input_tokens_seen": 5884260, "step": 284, "time_per_iteration": 2.578204393386841 }, { "auxiliary_loss_clip": 0.01331879, "auxiliary_loss_mlp": 0.01055137, "balance_loss_clip": 1.08682752, "balance_loss_mlp": 1.03844821, "epoch": 0.03426922383214093, "flos": 18879931207680.0, "grad_norm": 2.990672812337655, "language_loss": 0.90823644, "learning_rate": 3.999814171349399e-06, "loss": 0.93210661, "num_input_tokens_seen": 5902120, "step": 285, "time_per_iteration": 2.5438458919525146 }, { "auxiliary_loss_clip": 0.01325965, "auxiliary_loss_mlp": 0.0105662, "balance_loss_clip": 1.08783674, "balance_loss_mlp": 1.04232669, "epoch": 0.03438946672278002, "flos": 34752017604960.0, "grad_norm": 1.9520252331268846, "language_loss": 0.73532671, "learning_rate": 3.9998034010489655e-06, "loss": 0.75915247, "num_input_tokens_seen": 5925810, "step": 286, "time_per_iteration": 2.681288957595825 }, { "auxiliary_loss_clip": 0.01329245, "auxiliary_loss_mlp": 0.01060494, "balance_loss_clip": 1.08979988, "balance_loss_mlp": 1.0449667, "epoch": 0.03450970961341911, "flos": 22164105918720.0, "grad_norm": 3.1516002625034227, "language_loss": 0.75639999, "learning_rate": 3.999792327379946e-06, "loss": 0.78029734, "num_input_tokens_seen": 5945185, "step": 287, "time_per_iteration": 2.5627059936523438 }, { "auxiliary_loss_clip": 0.01335036, "auxiliary_loss_mlp": 0.01059208, "balance_loss_clip": 1.09315205, "balance_loss_mlp": 1.04387784, "epoch": 0.034629952504058197, "flos": 21725744247840.0, "grad_norm": 2.425383795323513, "language_loss": 0.9630217, "learning_rate": 3.999780950344021e-06, "loss": 0.98696411, "num_input_tokens_seen": 5963375, "step": 288, "time_per_iteration": 2.571099042892456 }, { "auxiliary_loss_clip": 0.01329699, "auxiliary_loss_mlp": 0.01052892, "balance_loss_clip": 1.08877659, "balance_loss_mlp": 1.03721559, "epoch": 0.034750195394697286, "flos": 20047997739360.0, "grad_norm": 1.8192832107331145, "language_loss": 0.82711041, "learning_rate": 3.999769269942916e-06, "loss": 0.85093629, "num_input_tokens_seen": 5983415, "step": 289, "time_per_iteration": 2.561180353164673 }, { "auxiliary_loss_clip": 0.01327574, "auxiliary_loss_mlp": 0.01056803, "balance_loss_clip": 1.08990312, "balance_loss_mlp": 1.04079366, "epoch": 0.034870438285336376, "flos": 27965870150400.0, "grad_norm": 2.0946208017335985, "language_loss": 0.81022733, "learning_rate": 3.999757286178402e-06, "loss": 0.83407104, "num_input_tokens_seen": 6005850, "step": 290, "time_per_iteration": 2.623450517654419 }, { "auxiliary_loss_clip": 0.01327159, "auxiliary_loss_mlp": 0.01048613, "balance_loss_clip": 1.08764195, "balance_loss_mlp": 1.03387833, "epoch": 0.03499068117597547, "flos": 22017518441280.0, "grad_norm": 2.219827229834086, "language_loss": 0.90498322, "learning_rate": 3.999744999052299e-06, "loss": 0.92874098, "num_input_tokens_seen": 6027240, "step": 291, "time_per_iteration": 2.598647117614746 }, { "auxiliary_loss_clip": 0.01325279, "auxiliary_loss_mlp": 0.0101812, "balance_loss_clip": 1.13460827, "balance_loss_mlp": 1.01149201, "epoch": 0.03511092406661456, "flos": 57242156108640.0, "grad_norm": 0.9616036049267507, "language_loss": 0.61213624, "learning_rate": 3.9997324085664675e-06, "loss": 0.63557023, "num_input_tokens_seen": 6087470, "step": 292, "time_per_iteration": 3.1418774127960205 }, { "auxiliary_loss_clip": 0.01328909, "auxiliary_loss_mlp": 0.01053439, "balance_loss_clip": 1.08844626, "balance_loss_mlp": 1.03816831, "epoch": 0.03523116695725365, "flos": 22928069171040.0, "grad_norm": 2.12507199704079, "language_loss": 0.91977179, "learning_rate": 3.999719514722821e-06, "loss": 0.94359529, "num_input_tokens_seen": 6107600, "step": 293, "time_per_iteration": 2.5897328853607178 }, { "auxiliary_loss_clip": 0.01318432, "auxiliary_loss_mlp": 0.01054261, "balance_loss_clip": 1.085186, "balance_loss_mlp": 1.03948462, "epoch": 0.03535140984789274, "flos": 36903246182880.0, "grad_norm": 2.458177806400054, "language_loss": 0.74940425, "learning_rate": 3.999706317523314e-06, "loss": 0.77313113, "num_input_tokens_seen": 6126160, "step": 294, "time_per_iteration": 2.6564149856567383 }, { "auxiliary_loss_clip": 0.01324452, "auxiliary_loss_mlp": 0.01058677, "balance_loss_clip": 1.08692014, "balance_loss_mlp": 1.04548061, "epoch": 0.03547165273853183, "flos": 20449155828000.0, "grad_norm": 1.8948041856841316, "language_loss": 0.86215997, "learning_rate": 3.999692816969948e-06, "loss": 0.88599122, "num_input_tokens_seen": 6145695, "step": 295, "time_per_iteration": 2.5793216228485107 }, { "auxiliary_loss_clip": 0.0131167, "auxiliary_loss_mlp": 0.01005603, "balance_loss_clip": 1.12530053, "balance_loss_mlp": 0.99902266, "epoch": 0.03559189562917092, "flos": 69850576379040.0, "grad_norm": 1.0842476486483636, "language_loss": 0.69474638, "learning_rate": 3.999679013064772e-06, "loss": 0.71791911, "num_input_tokens_seen": 6212440, "step": 296, "time_per_iteration": 3.200241804122925 }, { "auxiliary_loss_clip": 0.01320881, "auxiliary_loss_mlp": 0.01053602, "balance_loss_clip": 1.08698201, "balance_loss_mlp": 1.03952301, "epoch": 0.03571213851981002, "flos": 21651947671680.0, "grad_norm": 2.741616260626534, "language_loss": 0.85679746, "learning_rate": 3.99966490580988e-06, "loss": 0.88054234, "num_input_tokens_seen": 6229800, "step": 297, "time_per_iteration": 2.5962424278259277 }, { "auxiliary_loss_clip": 0.01327427, "auxiliary_loss_mlp": 0.01052003, "balance_loss_clip": 1.09007955, "balance_loss_mlp": 1.03786492, "epoch": 0.03583238141044911, "flos": 43945616428800.0, "grad_norm": 2.2404022991102694, "language_loss": 0.65812826, "learning_rate": 3.999650495207411e-06, "loss": 0.68192261, "num_input_tokens_seen": 6255825, "step": 298, "time_per_iteration": 2.755303144454956 }, { "auxiliary_loss_clip": 0.01325996, "auxiliary_loss_mlp": 0.01053357, "balance_loss_clip": 1.09079158, "balance_loss_mlp": 1.037884, "epoch": 0.0359526243010882, "flos": 18910813404960.0, "grad_norm": 2.5727517817889627, "language_loss": 0.90360236, "learning_rate": 3.999635781259553e-06, "loss": 0.92739588, "num_input_tokens_seen": 6271090, "step": 299, "time_per_iteration": 2.5600392818450928 }, { "auxiliary_loss_clip": 0.01295147, "auxiliary_loss_mlp": 0.01006366, "balance_loss_clip": 1.11377978, "balance_loss_mlp": 0.99983358, "epoch": 0.03607286719172729, "flos": 61668902698080.0, "grad_norm": 0.9198174521360335, "language_loss": 0.52285248, "learning_rate": 3.999620763968535e-06, "loss": 0.54586762, "num_input_tokens_seen": 6329965, "step": 300, "time_per_iteration": 3.023345947265625 }, { "auxiliary_loss_clip": 0.0132117, "auxiliary_loss_mlp": 0.01050584, "balance_loss_clip": 1.08746827, "balance_loss_mlp": 1.03729796, "epoch": 0.03619311008236638, "flos": 27819067171200.0, "grad_norm": 1.785763938823292, "language_loss": 0.86457741, "learning_rate": 3.999605443336638e-06, "loss": 0.88829494, "num_input_tokens_seen": 6352095, "step": 301, "time_per_iteration": 2.599754810333252 }, { "auxiliary_loss_clip": 0.01326085, "auxiliary_loss_mlp": 0.010473, "balance_loss_clip": 1.08828783, "balance_loss_mlp": 1.03238726, "epoch": 0.03631335297300547, "flos": 13621135586400.0, "grad_norm": 2.4042281295553587, "language_loss": 0.89308816, "learning_rate": 3.999589819366185e-06, "loss": 0.91682202, "num_input_tokens_seen": 6365885, "step": 302, "time_per_iteration": 2.5649399757385254 }, { "auxiliary_loss_clip": 0.01327983, "auxiliary_loss_mlp": 0.01068069, "balance_loss_clip": 1.08980942, "balance_loss_mlp": 1.05227351, "epoch": 0.036433595863644565, "flos": 27631792166400.0, "grad_norm": 1.9961496469761073, "language_loss": 0.8465699, "learning_rate": 3.999573892059547e-06, "loss": 0.87053049, "num_input_tokens_seen": 6385015, "step": 303, "time_per_iteration": 3.5047616958618164 }, { "auxiliary_loss_clip": 0.01332296, "auxiliary_loss_mlp": 0.01053287, "balance_loss_clip": 1.09003592, "balance_loss_mlp": 1.03889823, "epoch": 0.036553838754283655, "flos": 24572024208480.0, "grad_norm": 1.863169414106308, "language_loss": 0.81035686, "learning_rate": 3.999557661419138e-06, "loss": 0.83421272, "num_input_tokens_seen": 6405165, "step": 304, "time_per_iteration": 2.6182801723480225 }, { "auxiliary_loss_clip": 0.01330226, "auxiliary_loss_mlp": 0.01054116, "balance_loss_clip": 1.09241152, "balance_loss_mlp": 1.03928041, "epoch": 0.036674081644922744, "flos": 23404316846400.0, "grad_norm": 2.575833904467367, "language_loss": 0.81371164, "learning_rate": 3.9995411274474225e-06, "loss": 0.83755505, "num_input_tokens_seen": 6424445, "step": 305, "time_per_iteration": 3.3116321563720703 }, { "auxiliary_loss_clip": 0.01326776, "auxiliary_loss_mlp": 0.01059804, "balance_loss_clip": 1.09005725, "balance_loss_mlp": 1.04452157, "epoch": 0.036794324535561834, "flos": 27489694308960.0, "grad_norm": 2.6421434470698433, "language_loss": 0.81222314, "learning_rate": 3.999524290146908e-06, "loss": 0.8360889, "num_input_tokens_seen": 6444650, "step": 306, "time_per_iteration": 3.5089542865753174 }, { "auxiliary_loss_clip": 0.01322985, "auxiliary_loss_mlp": 0.01053501, "balance_loss_clip": 1.08838844, "balance_loss_mlp": 1.03973222, "epoch": 0.036914567426200924, "flos": 19463479594560.0, "grad_norm": 2.248717331214705, "language_loss": 0.92541528, "learning_rate": 3.9995071495201485e-06, "loss": 0.94918013, "num_input_tokens_seen": 6461755, "step": 307, "time_per_iteration": 2.5434818267822266 }, { "auxiliary_loss_clip": 0.01323977, "auxiliary_loss_mlp": 0.01052501, "balance_loss_clip": 1.09099936, "balance_loss_mlp": 1.03895283, "epoch": 0.037034810316840014, "flos": 22309328551680.0, "grad_norm": 2.7157073322914966, "language_loss": 0.97454101, "learning_rate": 3.999489705569744e-06, "loss": 0.9983058, "num_input_tokens_seen": 6479455, "step": 308, "time_per_iteration": 2.550619602203369 }, { "auxiliary_loss_clip": 0.01323531, "auxiliary_loss_mlp": 0.0104961, "balance_loss_clip": 1.08890104, "balance_loss_mlp": 1.03521538, "epoch": 0.03715505320747911, "flos": 18588336599040.0, "grad_norm": 1.953800859301433, "language_loss": 0.86314172, "learning_rate": 3.999471958298341e-06, "loss": 0.88687313, "num_input_tokens_seen": 6498365, "step": 309, "time_per_iteration": 2.5646026134490967 }, { "auxiliary_loss_clip": 0.01327493, "auxiliary_loss_mlp": 0.01060446, "balance_loss_clip": 1.09024382, "balance_loss_mlp": 1.0459919, "epoch": 0.0372752960981182, "flos": 35955348203040.0, "grad_norm": 1.80273045419176, "language_loss": 0.76007509, "learning_rate": 3.999453907708631e-06, "loss": 0.7839545, "num_input_tokens_seen": 6520770, "step": 310, "time_per_iteration": 2.6620023250579834 }, { "auxiliary_loss_clip": 0.01320359, "auxiliary_loss_mlp": 0.01053282, "balance_loss_clip": 1.08825588, "balance_loss_mlp": 1.03894079, "epoch": 0.03739553898875729, "flos": 20814044175360.0, "grad_norm": 1.7333611495289072, "language_loss": 0.81239152, "learning_rate": 3.999435553803353e-06, "loss": 0.83612788, "num_input_tokens_seen": 6540170, "step": 311, "time_per_iteration": 2.570380687713623 }, { "auxiliary_loss_clip": 0.01318152, "auxiliary_loss_mlp": 0.01058534, "balance_loss_clip": 1.0852387, "balance_loss_mlp": 1.04459834, "epoch": 0.03751578187939638, "flos": 20264143591680.0, "grad_norm": 2.313089779654683, "language_loss": 0.83253789, "learning_rate": 3.999416896585292e-06, "loss": 0.85630476, "num_input_tokens_seen": 6557200, "step": 312, "time_per_iteration": 2.548006534576416 }, { "auxiliary_loss_clip": 0.01320125, "auxiliary_loss_mlp": 0.01045743, "balance_loss_clip": 1.08742952, "balance_loss_mlp": 1.0315814, "epoch": 0.03763602477003547, "flos": 20668067286240.0, "grad_norm": 10.972219754133912, "language_loss": 0.85809529, "learning_rate": 3.9993979360572775e-06, "loss": 0.88175392, "num_input_tokens_seen": 6577340, "step": 313, "time_per_iteration": 2.6125741004943848 }, { "auxiliary_loss_clip": 0.01326318, "auxiliary_loss_mlp": 0.01056166, "balance_loss_clip": 1.09242272, "balance_loss_mlp": 1.04116368, "epoch": 0.03775626766067456, "flos": 16691355379680.0, "grad_norm": 2.141402149701059, "language_loss": 0.82833552, "learning_rate": 3.999378672222185e-06, "loss": 0.85216033, "num_input_tokens_seen": 6595125, "step": 314, "time_per_iteration": 2.5257837772369385 }, { "auxiliary_loss_clip": 0.01320887, "auxiliary_loss_mlp": 0.01054683, "balance_loss_clip": 1.09081161, "balance_loss_mlp": 1.04004455, "epoch": 0.03787651055131366, "flos": 21141800774400.0, "grad_norm": 2.7213784903896805, "language_loss": 0.82302457, "learning_rate": 3.9993591050829385e-06, "loss": 0.8467803, "num_input_tokens_seen": 6612990, "step": 315, "time_per_iteration": 2.593708038330078 }, { "auxiliary_loss_clip": 0.01324156, "auxiliary_loss_mlp": 0.01050876, "balance_loss_clip": 1.09201467, "balance_loss_mlp": 1.03634453, "epoch": 0.037996753441952746, "flos": 22018093112640.0, "grad_norm": 1.9449180311272067, "language_loss": 0.79188901, "learning_rate": 3.999339234642506e-06, "loss": 0.81563938, "num_input_tokens_seen": 6632740, "step": 316, "time_per_iteration": 2.5477826595306396 }, { "auxiliary_loss_clip": 0.01323097, "auxiliary_loss_mlp": 0.01055069, "balance_loss_clip": 1.09044909, "balance_loss_mlp": 1.03992987, "epoch": 0.038116996332591836, "flos": 27709395940320.0, "grad_norm": 1.9372727164170824, "language_loss": 0.83891153, "learning_rate": 3.9993190609038994e-06, "loss": 0.86269325, "num_input_tokens_seen": 6651505, "step": 317, "time_per_iteration": 2.6138148307800293 }, { "auxiliary_loss_clip": 0.0131846, "auxiliary_loss_mlp": 0.01047307, "balance_loss_clip": 1.08707619, "balance_loss_mlp": 1.03263223, "epoch": 0.038237239223230926, "flos": 21178070515680.0, "grad_norm": 2.080933434828261, "language_loss": 0.83065194, "learning_rate": 3.999298583870182e-06, "loss": 0.85430962, "num_input_tokens_seen": 6671090, "step": 318, "time_per_iteration": 2.550985097885132 }, { "auxiliary_loss_clip": 0.01315438, "auxiliary_loss_mlp": 0.01062169, "balance_loss_clip": 1.08452916, "balance_loss_mlp": 1.04825163, "epoch": 0.038357482113870016, "flos": 25556622933120.0, "grad_norm": 1.9075691961624435, "language_loss": 0.77529472, "learning_rate": 3.999277803544458e-06, "loss": 0.79907072, "num_input_tokens_seen": 6691245, "step": 319, "time_per_iteration": 2.606800079345703 }, { "auxiliary_loss_clip": 0.01250794, "auxiliary_loss_mlp": 0.01044066, "balance_loss_clip": 1.08083463, "balance_loss_mlp": 1.03677034, "epoch": 0.038477725004509106, "flos": 59227588023840.0, "grad_norm": 0.9671437894867687, "language_loss": 0.62378764, "learning_rate": 3.999256719929882e-06, "loss": 0.64673626, "num_input_tokens_seen": 6752520, "step": 320, "time_per_iteration": 3.1340928077697754 }, { "auxiliary_loss_clip": 0.01248293, "auxiliary_loss_mlp": 0.0103837, "balance_loss_clip": 1.07895625, "balance_loss_mlp": 1.03116977, "epoch": 0.0385979678951482, "flos": 67317686233440.0, "grad_norm": 1.2255821681285668, "language_loss": 0.67178047, "learning_rate": 3.999235333029651e-06, "loss": 0.69464713, "num_input_tokens_seen": 6806460, "step": 321, "time_per_iteration": 3.0503506660461426 }, { "auxiliary_loss_clip": 0.01318591, "auxiliary_loss_mlp": 0.01055664, "balance_loss_clip": 1.09021485, "balance_loss_mlp": 1.04092956, "epoch": 0.03871821078578729, "flos": 22746756381600.0, "grad_norm": 2.041209570014877, "language_loss": 0.82038146, "learning_rate": 3.999213642847009e-06, "loss": 0.84412402, "num_input_tokens_seen": 6827045, "step": 322, "time_per_iteration": 2.637996196746826 }, { "auxiliary_loss_clip": 0.01316777, "auxiliary_loss_mlp": 0.01055104, "balance_loss_clip": 1.08704376, "balance_loss_mlp": 1.0415256, "epoch": 0.03883845367642638, "flos": 26280616997280.0, "grad_norm": 1.7282970416055998, "language_loss": 0.90986025, "learning_rate": 3.999191649385247e-06, "loss": 0.93357909, "num_input_tokens_seen": 6848220, "step": 323, "time_per_iteration": 2.594404458999634 }, { "auxiliary_loss_clip": 0.01239332, "auxiliary_loss_mlp": 0.01005898, "balance_loss_clip": 1.07198644, "balance_loss_mlp": 0.99917442, "epoch": 0.03895869656706547, "flos": 56962845100320.0, "grad_norm": 1.2895678069788044, "language_loss": 0.59877688, "learning_rate": 3.999169352647702e-06, "loss": 0.62122917, "num_input_tokens_seen": 6909400, "step": 324, "time_per_iteration": 3.0858829021453857 }, { "auxiliary_loss_clip": 0.01320917, "auxiliary_loss_mlp": 0.01053582, "balance_loss_clip": 1.08969212, "balance_loss_mlp": 1.03947973, "epoch": 0.03907893945770456, "flos": 24863367398400.0, "grad_norm": 1.7868327327261853, "language_loss": 0.82753611, "learning_rate": 3.999146752637755e-06, "loss": 0.85128105, "num_input_tokens_seen": 6930445, "step": 325, "time_per_iteration": 2.5962231159210205 }, { "auxiliary_loss_clip": 0.01317417, "auxiliary_loss_mlp": 0.01051441, "balance_loss_clip": 1.08660257, "balance_loss_mlp": 1.03770816, "epoch": 0.03919918234834365, "flos": 18368599050720.0, "grad_norm": 2.6466928354578383, "language_loss": 0.89904016, "learning_rate": 3.999123849358836e-06, "loss": 0.92272878, "num_input_tokens_seen": 6948110, "step": 326, "time_per_iteration": 2.5523555278778076 }, { "auxiliary_loss_clip": 0.01314733, "auxiliary_loss_mlp": 0.01040319, "balance_loss_clip": 1.08552122, "balance_loss_mlp": 1.02735472, "epoch": 0.03931942523898275, "flos": 25225418305920.0, "grad_norm": 5.797491409577405, "language_loss": 0.74526769, "learning_rate": 3.999100642814418e-06, "loss": 0.76881826, "num_input_tokens_seen": 6968550, "step": 327, "time_per_iteration": 2.5723369121551514 }, { "auxiliary_loss_clip": 0.01316421, "auxiliary_loss_mlp": 0.01045813, "balance_loss_clip": 1.08666277, "balance_loss_mlp": 1.03332567, "epoch": 0.03943966812962184, "flos": 23257908953760.0, "grad_norm": 2.210743837745892, "language_loss": 0.88615108, "learning_rate": 3.999077133008022e-06, "loss": 0.90977347, "num_input_tokens_seen": 6987135, "step": 328, "time_per_iteration": 2.566403388977051 }, { "auxiliary_loss_clip": 0.01320886, "auxiliary_loss_mlp": 0.01073851, "balance_loss_clip": 1.08943415, "balance_loss_mlp": 1.05918837, "epoch": 0.03955991102026093, "flos": 29168841578880.0, "grad_norm": 1.7591936620466895, "language_loss": 0.90656722, "learning_rate": 3.9990533199432145e-06, "loss": 0.93051457, "num_input_tokens_seen": 7008630, "step": 329, "time_per_iteration": 3.405320882797241 }, { "auxiliary_loss_clip": 0.0131913, "auxiliary_loss_mlp": 0.01055366, "balance_loss_clip": 1.08919048, "balance_loss_mlp": 1.04045248, "epoch": 0.03968015391090002, "flos": 17602444863840.0, "grad_norm": 2.0891787909835964, "language_loss": 0.75732017, "learning_rate": 3.999029203623608e-06, "loss": 0.78106511, "num_input_tokens_seen": 7026350, "step": 330, "time_per_iteration": 2.5236194133758545 }, { "auxiliary_loss_clip": 0.01316258, "auxiliary_loss_mlp": 0.01064949, "balance_loss_clip": 1.08636403, "balance_loss_mlp": 1.05079281, "epoch": 0.03980039680153911, "flos": 21799289405280.0, "grad_norm": 1.922740559037425, "language_loss": 0.86790317, "learning_rate": 3.99900478405286e-06, "loss": 0.89171523, "num_input_tokens_seen": 7045660, "step": 331, "time_per_iteration": 4.230665445327759 }, { "auxiliary_loss_clip": 0.01317328, "auxiliary_loss_mlp": 0.01061015, "balance_loss_clip": 1.0870142, "balance_loss_mlp": 1.04656076, "epoch": 0.0399206396921782, "flos": 15195137162400.0, "grad_norm": 5.959956016040264, "language_loss": 0.82233554, "learning_rate": 3.998980061234676e-06, "loss": 0.84611899, "num_input_tokens_seen": 7063575, "step": 332, "time_per_iteration": 2.5248279571533203 }, { "auxiliary_loss_clip": 0.0131819, "auxiliary_loss_mlp": 0.01053892, "balance_loss_clip": 1.08779716, "balance_loss_mlp": 1.03946221, "epoch": 0.040040882582817294, "flos": 14422913009280.0, "grad_norm": 2.481380082085373, "language_loss": 0.75657827, "learning_rate": 3.9989550351728055e-06, "loss": 0.78029907, "num_input_tokens_seen": 7080505, "step": 333, "time_per_iteration": 3.4194934368133545 }, { "auxiliary_loss_clip": 0.01316484, "auxiliary_loss_mlp": 0.01059846, "balance_loss_clip": 1.08774209, "balance_loss_mlp": 1.04636943, "epoch": 0.040161125473456384, "flos": 19280909711520.0, "grad_norm": 3.007947665271289, "language_loss": 0.84800518, "learning_rate": 3.998929705871046e-06, "loss": 0.87176847, "num_input_tokens_seen": 7097860, "step": 334, "time_per_iteration": 2.548095941543579 }, { "auxiliary_loss_clip": 0.01314274, "auxiliary_loss_mlp": 0.01046736, "balance_loss_clip": 1.08727503, "balance_loss_mlp": 1.03259802, "epoch": 0.040281368364095474, "flos": 17821104903360.0, "grad_norm": 2.5571508321306227, "language_loss": 0.89088607, "learning_rate": 3.99890407333324e-06, "loss": 0.91449618, "num_input_tokens_seen": 7116390, "step": 335, "time_per_iteration": 2.528686285018921 }, { "auxiliary_loss_clip": 0.01310988, "auxiliary_loss_mlp": 0.01057165, "balance_loss_clip": 1.0822283, "balance_loss_mlp": 1.04341388, "epoch": 0.040401611254734564, "flos": 19573761413760.0, "grad_norm": 2.1701855894727435, "language_loss": 0.87026149, "learning_rate": 3.998878137563275e-06, "loss": 0.89394307, "num_input_tokens_seen": 7135940, "step": 336, "time_per_iteration": 2.5495855808258057 }, { "auxiliary_loss_clip": 0.01315825, "auxiliary_loss_mlp": 0.0104804, "balance_loss_clip": 1.08696842, "balance_loss_mlp": 1.03405666, "epoch": 0.040521854145373654, "flos": 22054470604800.0, "grad_norm": 2.4516065354254484, "language_loss": 0.8535732, "learning_rate": 3.998851898565085e-06, "loss": 0.87721181, "num_input_tokens_seen": 7155745, "step": 337, "time_per_iteration": 2.545060634613037 }, { "auxiliary_loss_clip": 0.01311502, "auxiliary_loss_mlp": 0.01045297, "balance_loss_clip": 1.08370304, "balance_loss_mlp": 1.03128445, "epoch": 0.04064209703601274, "flos": 22674647902560.0, "grad_norm": 1.8683438364753258, "language_loss": 0.8312006, "learning_rate": 3.998825356342653e-06, "loss": 0.85476857, "num_input_tokens_seen": 7175920, "step": 338, "time_per_iteration": 2.578061819076538 }, { "auxiliary_loss_clip": 0.01309533, "auxiliary_loss_mlp": 0.0105123, "balance_loss_clip": 1.08325958, "balance_loss_mlp": 1.03729463, "epoch": 0.04076233992665183, "flos": 38582177951040.0, "grad_norm": 2.2250917763082647, "language_loss": 0.72903967, "learning_rate": 3.998798510900003e-06, "loss": 0.75264728, "num_input_tokens_seen": 7198720, "step": 339, "time_per_iteration": 2.6795549392700195 }, { "auxiliary_loss_clip": 0.01309266, "auxiliary_loss_mlp": 0.01045542, "balance_loss_clip": 1.08283114, "balance_loss_mlp": 1.03139806, "epoch": 0.04088258281729093, "flos": 25885313373120.0, "grad_norm": 2.1085189094829633, "language_loss": 0.83770597, "learning_rate": 3.998771362241207e-06, "loss": 0.8612541, "num_input_tokens_seen": 7219125, "step": 340, "time_per_iteration": 2.582773208618164 }, { "auxiliary_loss_clip": 0.01310822, "auxiliary_loss_mlp": 0.01044687, "balance_loss_clip": 1.08491898, "balance_loss_mlp": 1.03092384, "epoch": 0.04100282570793002, "flos": 19789835432160.0, "grad_norm": 1.8454576586943225, "language_loss": 0.87964165, "learning_rate": 3.998743910370385e-06, "loss": 0.90319669, "num_input_tokens_seen": 7237985, "step": 341, "time_per_iteration": 2.5343306064605713 }, { "auxiliary_loss_clip": 0.01318227, "auxiliary_loss_mlp": 0.0106322, "balance_loss_clip": 1.09019184, "balance_loss_mlp": 1.05000544, "epoch": 0.04112306859856911, "flos": 22565156256480.0, "grad_norm": 2.1585820169328787, "language_loss": 0.73199522, "learning_rate": 3.998716155291702e-06, "loss": 0.75580966, "num_input_tokens_seen": 7255825, "step": 342, "time_per_iteration": 2.5584709644317627 }, { "auxiliary_loss_clip": 0.01314034, "auxiliary_loss_mlp": 0.0105351, "balance_loss_clip": 1.08749628, "balance_loss_mlp": 1.04051042, "epoch": 0.0412433114892082, "flos": 25040657488320.0, "grad_norm": 1.6716524914385258, "language_loss": 0.90505439, "learning_rate": 3.998688097009366e-06, "loss": 0.92872989, "num_input_tokens_seen": 7276590, "step": 343, "time_per_iteration": 2.5712718963623047 }, { "auxiliary_loss_clip": 0.01309182, "auxiliary_loss_mlp": 0.01043818, "balance_loss_clip": 1.0857029, "balance_loss_mlp": 1.03010273, "epoch": 0.04136355437984729, "flos": 25191375416160.0, "grad_norm": 2.2511418497785347, "language_loss": 0.79971904, "learning_rate": 3.998659735527636e-06, "loss": 0.82324898, "num_input_tokens_seen": 7295680, "step": 344, "time_per_iteration": 2.5778956413269043 }, { "auxiliary_loss_clip": 0.01310818, "auxiliary_loss_mlp": 0.01039569, "balance_loss_clip": 1.08549428, "balance_loss_mlp": 1.02731979, "epoch": 0.04148379727048638, "flos": 22966781265600.0, "grad_norm": 3.416399000456684, "language_loss": 0.77991277, "learning_rate": 3.998631070850813e-06, "loss": 0.80341661, "num_input_tokens_seen": 7316300, "step": 345, "time_per_iteration": 2.5545222759246826 }, { "auxiliary_loss_clip": 0.01311371, "auxiliary_loss_mlp": 0.01048371, "balance_loss_clip": 1.08708572, "balance_loss_mlp": 1.03598487, "epoch": 0.041604040161125476, "flos": 14063484039840.0, "grad_norm": 2.2775316134684234, "language_loss": 0.83747542, "learning_rate": 3.9986021029832455e-06, "loss": 0.86107284, "num_input_tokens_seen": 7333615, "step": 346, "time_per_iteration": 2.525789499282837 }, { "auxiliary_loss_clip": 0.01310398, "auxiliary_loss_mlp": 0.01047325, "balance_loss_clip": 1.08350599, "balance_loss_mlp": 1.03427184, "epoch": 0.041724283051764566, "flos": 12091880154240.0, "grad_norm": 2.6237023711618366, "language_loss": 0.91293937, "learning_rate": 3.9985728319293285e-06, "loss": 0.93651652, "num_input_tokens_seen": 7347590, "step": 347, "time_per_iteration": 2.5171167850494385 }, { "auxiliary_loss_clip": 0.01315558, "auxiliary_loss_mlp": 0.0104825, "balance_loss_clip": 1.08576465, "balance_loss_mlp": 1.03371227, "epoch": 0.041844525942403656, "flos": 12385306527840.0, "grad_norm": 2.783482232070701, "language_loss": 0.85179818, "learning_rate": 3.998543257693501e-06, "loss": 0.87543631, "num_input_tokens_seen": 7364345, "step": 348, "time_per_iteration": 2.5055763721466064 }, { "auxiliary_loss_clip": 0.01308773, "auxiliary_loss_mlp": 0.01043022, "balance_loss_clip": 1.08594298, "balance_loss_mlp": 1.02937901, "epoch": 0.041964768833042745, "flos": 23769348861600.0, "grad_norm": 3.418742147907398, "language_loss": 0.87676758, "learning_rate": 3.998513380280251e-06, "loss": 0.90028548, "num_input_tokens_seen": 7384625, "step": 349, "time_per_iteration": 2.5590643882751465 }, { "auxiliary_loss_clip": 0.01316962, "auxiliary_loss_mlp": 0.01059334, "balance_loss_clip": 1.08791924, "balance_loss_mlp": 1.0459404, "epoch": 0.042085011723681835, "flos": 11875339215360.0, "grad_norm": 2.526988353314892, "language_loss": 0.94909281, "learning_rate": 3.99848319969411e-06, "loss": 0.97285575, "num_input_tokens_seen": 7402225, "step": 350, "time_per_iteration": 2.5057101249694824 }, { "auxiliary_loss_clip": 0.01321562, "auxiliary_loss_mlp": 0.01059969, "balance_loss_clip": 1.09169674, "balance_loss_mlp": 1.04438829, "epoch": 0.042205254614320925, "flos": 16873961179680.0, "grad_norm": 2.7054629926027967, "language_loss": 0.79237533, "learning_rate": 3.9984527159396564e-06, "loss": 0.81619072, "num_input_tokens_seen": 7420865, "step": 351, "time_per_iteration": 2.5385665893554688 }, { "auxiliary_loss_clip": 0.01306405, "auxiliary_loss_mlp": 0.01043469, "balance_loss_clip": 1.08119965, "balance_loss_mlp": 1.02966511, "epoch": 0.04232549750496002, "flos": 25118512680960.0, "grad_norm": 2.0381468021623275, "language_loss": 0.84610844, "learning_rate": 3.9984219290215154e-06, "loss": 0.86960721, "num_input_tokens_seen": 7441040, "step": 352, "time_per_iteration": 2.605191230773926 }, { "auxiliary_loss_clip": 0.01306956, "auxiliary_loss_mlp": 0.01050507, "balance_loss_clip": 1.08560908, "balance_loss_mlp": 1.03731668, "epoch": 0.04244574039559911, "flos": 26724545796960.0, "grad_norm": 1.7845520242651662, "language_loss": 0.88867658, "learning_rate": 3.998390838944356e-06, "loss": 0.91225123, "num_input_tokens_seen": 7462545, "step": 353, "time_per_iteration": 2.6544814109802246 }, { "auxiliary_loss_clip": 0.01310826, "auxiliary_loss_mlp": 0.01044126, "balance_loss_clip": 1.08416581, "balance_loss_mlp": 1.03000546, "epoch": 0.0425659832862382, "flos": 20923248485760.0, "grad_norm": 10.614679522186451, "language_loss": 0.90588069, "learning_rate": 3.998359445712895e-06, "loss": 0.92943025, "num_input_tokens_seen": 7481650, "step": 354, "time_per_iteration": 2.567107915878296 }, { "auxiliary_loss_clip": 0.01307437, "auxiliary_loss_mlp": 0.01045487, "balance_loss_clip": 1.0821352, "balance_loss_mlp": 1.03168821, "epoch": 0.04268622617687729, "flos": 23331130858560.0, "grad_norm": 2.605391020634303, "language_loss": 0.81164491, "learning_rate": 3.9983277493318955e-06, "loss": 0.8351742, "num_input_tokens_seen": 7500945, "step": 355, "time_per_iteration": 3.373425006866455 }, { "auxiliary_loss_clip": 0.01309711, "auxiliary_loss_mlp": 0.01046909, "balance_loss_clip": 1.08399045, "balance_loss_mlp": 1.03217447, "epoch": 0.04280646906751638, "flos": 25994014846080.0, "grad_norm": 1.794362048282773, "language_loss": 0.81143683, "learning_rate": 3.998295749806165e-06, "loss": 0.83500302, "num_input_tokens_seen": 7522170, "step": 356, "time_per_iteration": 2.5894060134887695 }, { "auxiliary_loss_clip": 0.01314897, "auxiliary_loss_mlp": 0.01054673, "balance_loss_clip": 1.09016728, "balance_loss_mlp": 1.04088092, "epoch": 0.04292671195815547, "flos": 26906828344320.0, "grad_norm": 2.3269083202508822, "language_loss": 0.83410025, "learning_rate": 3.998263447140558e-06, "loss": 0.85779595, "num_input_tokens_seen": 7542370, "step": 357, "time_per_iteration": 2.6243414878845215 }, { "auxiliary_loss_clip": 0.01307204, "auxiliary_loss_mlp": 0.01057876, "balance_loss_clip": 1.0824765, "balance_loss_mlp": 1.04444718, "epoch": 0.04304695484879457, "flos": 39457320946560.0, "grad_norm": 1.8823530313136236, "language_loss": 0.81603611, "learning_rate": 3.998230841339976e-06, "loss": 0.83968699, "num_input_tokens_seen": 7564380, "step": 358, "time_per_iteration": 2.701960802078247 }, { "auxiliary_loss_clip": 0.01312151, "auxiliary_loss_mlp": 0.01058038, "balance_loss_clip": 1.0888561, "balance_loss_mlp": 1.04407823, "epoch": 0.04316719773943366, "flos": 19646300896320.0, "grad_norm": 2.156341839686477, "language_loss": 0.84801865, "learning_rate": 3.998197932409363e-06, "loss": 0.87172049, "num_input_tokens_seen": 7582390, "step": 359, "time_per_iteration": 3.4730687141418457 }, { "auxiliary_loss_clip": 0.01305126, "auxiliary_loss_mlp": 0.01048505, "balance_loss_clip": 1.0815748, "balance_loss_mlp": 1.03498054, "epoch": 0.04328744063007275, "flos": 22452324333120.0, "grad_norm": 1.985807025315814, "language_loss": 0.86238396, "learning_rate": 3.9981647203537125e-06, "loss": 0.88592029, "num_input_tokens_seen": 7599890, "step": 360, "time_per_iteration": 3.373750686645508 }, { "auxiliary_loss_clip": 0.01308607, "auxiliary_loss_mlp": 0.01043931, "balance_loss_clip": 1.08346605, "balance_loss_mlp": 1.0312705, "epoch": 0.04340768352071184, "flos": 21283036624800.0, "grad_norm": 1.956481947360966, "language_loss": 0.9594211, "learning_rate": 3.998131205178063e-06, "loss": 0.98294646, "num_input_tokens_seen": 7618360, "step": 361, "time_per_iteration": 2.621349334716797 }, { "auxiliary_loss_clip": 0.01307639, "auxiliary_loss_mlp": 0.01050485, "balance_loss_clip": 1.08154678, "balance_loss_mlp": 1.03619742, "epoch": 0.04352792641135093, "flos": 11583708689760.0, "grad_norm": 2.3534366035445986, "language_loss": 0.7679354, "learning_rate": 3.998097386887498e-06, "loss": 0.79151666, "num_input_tokens_seen": 7635435, "step": 362, "time_per_iteration": 2.5298831462860107 }, { "auxiliary_loss_clip": 0.01305019, "auxiliary_loss_mlp": 0.01048899, "balance_loss_clip": 1.08507037, "balance_loss_mlp": 1.03532076, "epoch": 0.04364816930199002, "flos": 23623551557280.0, "grad_norm": 1.8116697491510005, "language_loss": 0.84879363, "learning_rate": 3.998063265487148e-06, "loss": 0.87233281, "num_input_tokens_seen": 7656485, "step": 363, "time_per_iteration": 2.6145389080047607 }, { "auxiliary_loss_clip": 0.0131098, "auxiliary_loss_mlp": 0.01056942, "balance_loss_clip": 1.08677173, "balance_loss_mlp": 1.04285157, "epoch": 0.043768412192629114, "flos": 14429737231680.0, "grad_norm": 1.808184874370858, "language_loss": 0.8081733, "learning_rate": 3.99802884098219e-06, "loss": 0.83185256, "num_input_tokens_seen": 7674595, "step": 364, "time_per_iteration": 2.571897268295288 }, { "auxiliary_loss_clip": 0.0130843, "auxiliary_loss_mlp": 0.01043576, "balance_loss_clip": 1.08281171, "balance_loss_mlp": 1.02989721, "epoch": 0.043888655083268203, "flos": 26468897676960.0, "grad_norm": 3.9955707765727957, "language_loss": 0.82145381, "learning_rate": 3.997994113377845e-06, "loss": 0.84497386, "num_input_tokens_seen": 7693495, "step": 365, "time_per_iteration": 2.6005773544311523 }, { "auxiliary_loss_clip": 0.01312648, "auxiliary_loss_mlp": 0.01065705, "balance_loss_clip": 1.08549738, "balance_loss_mlp": 1.05211496, "epoch": 0.04400889797390729, "flos": 27235267365600.0, "grad_norm": 2.8005331157665245, "language_loss": 0.8325296, "learning_rate": 3.9979590826793815e-06, "loss": 0.85631311, "num_input_tokens_seen": 7714685, "step": 366, "time_per_iteration": 2.637108564376831 }, { "auxiliary_loss_clip": 0.01309447, "auxiliary_loss_mlp": 0.01051135, "balance_loss_clip": 1.08533502, "balance_loss_mlp": 1.03680611, "epoch": 0.04412914086454638, "flos": 20119531547040.0, "grad_norm": 2.142134498789322, "language_loss": 0.80970418, "learning_rate": 3.997923748892113e-06, "loss": 0.83331001, "num_input_tokens_seen": 7734005, "step": 367, "time_per_iteration": 2.5623958110809326 }, { "auxiliary_loss_clip": 0.0130297, "auxiliary_loss_mlp": 0.01053438, "balance_loss_clip": 1.08307862, "balance_loss_mlp": 1.03913927, "epoch": 0.04424938375518547, "flos": 22604622607200.0, "grad_norm": 1.6831086582604355, "language_loss": 0.88550818, "learning_rate": 3.9978881120214015e-06, "loss": 0.90907228, "num_input_tokens_seen": 7755525, "step": 368, "time_per_iteration": 2.59368896484375 }, { "auxiliary_loss_clip": 0.01307512, "auxiliary_loss_mlp": 0.01046764, "balance_loss_clip": 1.0831672, "balance_loss_mlp": 1.03205407, "epoch": 0.04436962664582456, "flos": 24132369527040.0, "grad_norm": 2.1023402179350743, "language_loss": 0.79199326, "learning_rate": 3.997852172072652e-06, "loss": 0.81553602, "num_input_tokens_seen": 7776740, "step": 369, "time_per_iteration": 2.587366819381714 }, { "auxiliary_loss_clip": 0.0131053, "auxiliary_loss_mlp": 0.01060482, "balance_loss_clip": 1.08576524, "balance_loss_mlp": 1.04623663, "epoch": 0.04448986953646366, "flos": 18222909497280.0, "grad_norm": 2.5170231704871555, "language_loss": 0.89174318, "learning_rate": 3.9978159290513155e-06, "loss": 0.91545326, "num_input_tokens_seen": 7794820, "step": 370, "time_per_iteration": 2.5547616481781006 }, { "auxiliary_loss_clip": 0.01310066, "auxiliary_loss_mlp": 0.01048983, "balance_loss_clip": 1.08406746, "balance_loss_mlp": 1.03542876, "epoch": 0.04461011242710275, "flos": 30117924818400.0, "grad_norm": 1.661998803651437, "language_loss": 0.80153412, "learning_rate": 3.997779382962892e-06, "loss": 0.82512462, "num_input_tokens_seen": 7817705, "step": 371, "time_per_iteration": 2.642487049102783 }, { "auxiliary_loss_clip": 0.01301988, "auxiliary_loss_mlp": 0.01050736, "balance_loss_clip": 1.07968473, "balance_loss_mlp": 1.03677619, "epoch": 0.04473035531774184, "flos": 29752533633600.0, "grad_norm": 1.9885441584478638, "language_loss": 0.73996818, "learning_rate": 3.997742533812924e-06, "loss": 0.76349545, "num_input_tokens_seen": 7840970, "step": 372, "time_per_iteration": 2.6447064876556396 }, { "auxiliary_loss_clip": 0.0131031, "auxiliary_loss_mlp": 0.01058972, "balance_loss_clip": 1.08655238, "balance_loss_mlp": 1.04525137, "epoch": 0.04485059820838093, "flos": 13151568465600.0, "grad_norm": 2.462632701588078, "language_loss": 0.92573333, "learning_rate": 3.997705381607001e-06, "loss": 0.94942617, "num_input_tokens_seen": 7857785, "step": 373, "time_per_iteration": 2.5282862186431885 }, { "auxiliary_loss_clip": 0.01229813, "auxiliary_loss_mlp": 0.01023246, "balance_loss_clip": 1.07545877, "balance_loss_mlp": 1.01769102, "epoch": 0.04497084109902002, "flos": 68094220421760.0, "grad_norm": 0.9770895256064118, "language_loss": 0.6030488, "learning_rate": 3.997667926350761e-06, "loss": 0.62557936, "num_input_tokens_seen": 7916115, "step": 374, "time_per_iteration": 3.073038101196289 }, { "auxiliary_loss_clip": 0.01229294, "auxiliary_loss_mlp": 0.01015344, "balance_loss_clip": 1.07519388, "balance_loss_mlp": 1.00959861, "epoch": 0.04509108398965911, "flos": 64342274437920.0, "grad_norm": 0.902457461687955, "language_loss": 0.57781553, "learning_rate": 3.997630168049886e-06, "loss": 0.60026193, "num_input_tokens_seen": 7974480, "step": 375, "time_per_iteration": 3.160196542739868 }, { "auxiliary_loss_clip": 0.01312831, "auxiliary_loss_mlp": 0.01059656, "balance_loss_clip": 1.08785415, "balance_loss_mlp": 1.04477286, "epoch": 0.045211326880298205, "flos": 22271119294560.0, "grad_norm": 1.9392639451014895, "language_loss": 0.7755484, "learning_rate": 3.997592106710101e-06, "loss": 0.79927325, "num_input_tokens_seen": 7993940, "step": 376, "time_per_iteration": 2.577322006225586 }, { "auxiliary_loss_clip": 0.01306246, "auxiliary_loss_mlp": 0.01061907, "balance_loss_clip": 1.08269191, "balance_loss_mlp": 1.04859173, "epoch": 0.045331569770937295, "flos": 32159446248480.0, "grad_norm": 2.5389989369345094, "language_loss": 0.65813398, "learning_rate": 3.997553742337182e-06, "loss": 0.68181545, "num_input_tokens_seen": 8013365, "step": 377, "time_per_iteration": 2.6671454906463623 }, { "auxiliary_loss_clip": 0.01306351, "auxiliary_loss_mlp": 0.01047792, "balance_loss_clip": 1.08486593, "balance_loss_mlp": 1.03448296, "epoch": 0.045451812661576385, "flos": 22163459413440.0, "grad_norm": 1.858102660255961, "language_loss": 0.91499126, "learning_rate": 3.997515074936949e-06, "loss": 0.93853265, "num_input_tokens_seen": 8034240, "step": 378, "time_per_iteration": 2.5798540115356445 }, { "auxiliary_loss_clip": 0.01307967, "auxiliary_loss_mlp": 0.01064279, "balance_loss_clip": 1.08559382, "balance_loss_mlp": 1.04981887, "epoch": 0.045572055552215475, "flos": 16581971484480.0, "grad_norm": 2.4399131361371977, "language_loss": 0.86867177, "learning_rate": 3.997476104515268e-06, "loss": 0.89239419, "num_input_tokens_seen": 8052430, "step": 379, "time_per_iteration": 2.597163200378418 }, { "auxiliary_loss_clip": 0.01301076, "auxiliary_loss_mlp": 0.01054117, "balance_loss_clip": 1.08286202, "balance_loss_mlp": 1.04104018, "epoch": 0.045692298442854565, "flos": 17603378704800.0, "grad_norm": 2.0142627688351515, "language_loss": 0.77443969, "learning_rate": 3.9974368310780485e-06, "loss": 0.79799163, "num_input_tokens_seen": 8069605, "step": 380, "time_per_iteration": 2.5262906551361084 }, { "auxiliary_loss_clip": 0.01314692, "auxiliary_loss_mlp": 0.01050419, "balance_loss_clip": 1.08547497, "balance_loss_mlp": 1.03555417, "epoch": 0.045812541333493655, "flos": 26761102873920.0, "grad_norm": 2.436143310680266, "language_loss": 0.74608797, "learning_rate": 3.997397254631251e-06, "loss": 0.76973909, "num_input_tokens_seen": 8090225, "step": 381, "time_per_iteration": 2.656644821166992 }, { "auxiliary_loss_clip": 0.01216115, "auxiliary_loss_mlp": 0.01018056, "balance_loss_clip": 1.06469321, "balance_loss_mlp": 1.01211977, "epoch": 0.04593278422413275, "flos": 60250360088640.0, "grad_norm": 0.8244037131050095, "language_loss": 0.60087097, "learning_rate": 3.997357375180878e-06, "loss": 0.6232127, "num_input_tokens_seen": 8154505, "step": 382, "time_per_iteration": 3.9769110679626465 }, { "auxiliary_loss_clip": 0.01303241, "auxiliary_loss_mlp": 0.01049385, "balance_loss_clip": 1.08205891, "balance_loss_mlp": 1.03578973, "epoch": 0.04605302711477184, "flos": 21799253488320.0, "grad_norm": 3.79220282685853, "language_loss": 0.75281084, "learning_rate": 3.997317192732979e-06, "loss": 0.77633715, "num_input_tokens_seen": 8173285, "step": 383, "time_per_iteration": 2.587550401687622 }, { "auxiliary_loss_clip": 0.01307878, "auxiliary_loss_mlp": 0.01039516, "balance_loss_clip": 1.08457923, "balance_loss_mlp": 1.02544332, "epoch": 0.04617327000541093, "flos": 19459708313760.0, "grad_norm": 1.8178468394184457, "language_loss": 0.82328844, "learning_rate": 3.99727670729365e-06, "loss": 0.84676236, "num_input_tokens_seen": 8191845, "step": 384, "time_per_iteration": 2.5319418907165527 }, { "auxiliary_loss_clip": 0.01303143, "auxiliary_loss_mlp": 0.01051422, "balance_loss_clip": 1.08447385, "balance_loss_mlp": 1.03806448, "epoch": 0.04629351289605002, "flos": 25411472134080.0, "grad_norm": 1.7279335518790389, "language_loss": 0.77972144, "learning_rate": 3.997235918869033e-06, "loss": 0.80326712, "num_input_tokens_seen": 8212880, "step": 385, "time_per_iteration": 4.174520492553711 }, { "auxiliary_loss_clip": 0.01303968, "auxiliary_loss_mlp": 0.01049062, "balance_loss_clip": 1.08421111, "balance_loss_mlp": 1.03535867, "epoch": 0.04641375578668911, "flos": 20558288304480.0, "grad_norm": 1.9188840162672722, "language_loss": 0.82830149, "learning_rate": 3.997194827465315e-06, "loss": 0.85183179, "num_input_tokens_seen": 8231475, "step": 386, "time_per_iteration": 3.4338295459747314 }, { "auxiliary_loss_clip": 0.0130407, "auxiliary_loss_mlp": 0.01059221, "balance_loss_clip": 1.08122194, "balance_loss_mlp": 1.04590535, "epoch": 0.0465339986773282, "flos": 13188664296960.0, "grad_norm": 2.582187734281517, "language_loss": 0.91559887, "learning_rate": 3.997153433088728e-06, "loss": 0.93923181, "num_input_tokens_seen": 8248600, "step": 387, "time_per_iteration": 2.627547264099121 }, { "auxiliary_loss_clip": 0.01303712, "auxiliary_loss_mlp": 0.01045787, "balance_loss_clip": 1.08213544, "balance_loss_mlp": 1.03031385, "epoch": 0.0466542415679673, "flos": 25556551099200.0, "grad_norm": 2.2017357706483422, "language_loss": 0.81176484, "learning_rate": 3.997111735745554e-06, "loss": 0.8352598, "num_input_tokens_seen": 8271570, "step": 388, "time_per_iteration": 2.695521116256714 }, { "auxiliary_loss_clip": 0.01301714, "auxiliary_loss_mlp": 0.01046431, "balance_loss_clip": 1.08180118, "balance_loss_mlp": 1.03206599, "epoch": 0.04677448445860639, "flos": 22236753152160.0, "grad_norm": 2.0701884948165765, "language_loss": 0.82684261, "learning_rate": 3.997069735442118e-06, "loss": 0.85032403, "num_input_tokens_seen": 8291265, "step": 389, "time_per_iteration": 2.6124491691589355 }, { "auxiliary_loss_clip": 0.01305528, "auxiliary_loss_mlp": 0.01047751, "balance_loss_clip": 1.08371043, "balance_loss_mlp": 1.03489447, "epoch": 0.04689472734924548, "flos": 28147829445120.0, "grad_norm": 1.5997773542218496, "language_loss": 0.80375218, "learning_rate": 3.997027432184792e-06, "loss": 0.82728499, "num_input_tokens_seen": 8315925, "step": 390, "time_per_iteration": 2.6710731983184814 }, { "auxiliary_loss_clip": 0.01301816, "auxiliary_loss_mlp": 0.01046393, "balance_loss_clip": 1.08336973, "balance_loss_mlp": 1.03199291, "epoch": 0.04701497023988457, "flos": 23148956062080.0, "grad_norm": 2.4041293240923016, "language_loss": 0.89489174, "learning_rate": 3.99698482597999e-06, "loss": 0.91837394, "num_input_tokens_seen": 8333605, "step": 391, "time_per_iteration": 2.5791847705841064 }, { "auxiliary_loss_clip": 0.01211862, "auxiliary_loss_mlp": 0.01008565, "balance_loss_clip": 1.05978298, "balance_loss_mlp": 1.00284314, "epoch": 0.04713521313052366, "flos": 64827680938080.0, "grad_norm": 0.8738595102275255, "language_loss": 0.64038897, "learning_rate": 3.99694191683418e-06, "loss": 0.66259325, "num_input_tokens_seen": 8394405, "step": 392, "time_per_iteration": 3.161806106567383 }, { "auxiliary_loss_clip": 0.01305849, "auxiliary_loss_mlp": 0.0105126, "balance_loss_clip": 1.08557391, "balance_loss_mlp": 1.03641868, "epoch": 0.047255456021162746, "flos": 18771588904320.0, "grad_norm": 2.209606731377881, "language_loss": 0.81708813, "learning_rate": 3.996898704753867e-06, "loss": 0.8406592, "num_input_tokens_seen": 8412355, "step": 393, "time_per_iteration": 2.572014808654785 }, { "auxiliary_loss_clip": 0.01301946, "auxiliary_loss_mlp": 0.01058137, "balance_loss_clip": 1.08149171, "balance_loss_mlp": 1.04289007, "epoch": 0.04737569891180184, "flos": 22053824099520.0, "grad_norm": 2.4062698545235373, "language_loss": 0.87588406, "learning_rate": 3.996855189745609e-06, "loss": 0.89948487, "num_input_tokens_seen": 8431620, "step": 394, "time_per_iteration": 2.5449414253234863 }, { "auxiliary_loss_clip": 0.01301247, "auxiliary_loss_mlp": 0.01058645, "balance_loss_clip": 1.07981682, "balance_loss_mlp": 1.04584765, "epoch": 0.04749594180244093, "flos": 29057374500000.0, "grad_norm": 1.9306924501566234, "language_loss": 0.92561179, "learning_rate": 3.996811371816007e-06, "loss": 0.94921076, "num_input_tokens_seen": 8454045, "step": 395, "time_per_iteration": 2.6335582733154297 }, { "auxiliary_loss_clip": 0.01300755, "auxiliary_loss_mlp": 0.01050427, "balance_loss_clip": 1.08349407, "balance_loss_mlp": 1.03714132, "epoch": 0.04761618469308002, "flos": 35112272664480.0, "grad_norm": 2.447880901782719, "language_loss": 0.77951711, "learning_rate": 3.996767250971707e-06, "loss": 0.803029, "num_input_tokens_seen": 8476785, "step": 396, "time_per_iteration": 2.6619760990142822 }, { "auxiliary_loss_clip": 0.0130315, "auxiliary_loss_mlp": 0.01048609, "balance_loss_clip": 1.0837189, "balance_loss_mlp": 1.0344764, "epoch": 0.04773642758371911, "flos": 25630706844960.0, "grad_norm": 2.4826493587250074, "language_loss": 0.86722517, "learning_rate": 3.996722827219403e-06, "loss": 0.89074272, "num_input_tokens_seen": 8498400, "step": 397, "time_per_iteration": 2.615945816040039 }, { "auxiliary_loss_clip": 0.0130309, "auxiliary_loss_mlp": 0.01059086, "balance_loss_clip": 1.08390081, "balance_loss_mlp": 1.04453695, "epoch": 0.0478566704743582, "flos": 20631510209280.0, "grad_norm": 2.7050425987871582, "language_loss": 0.82551664, "learning_rate": 3.996678100565833e-06, "loss": 0.84913838, "num_input_tokens_seen": 8517455, "step": 398, "time_per_iteration": 2.535600423812866 }, { "auxiliary_loss_clip": 0.01301087, "auxiliary_loss_mlp": 0.01053691, "balance_loss_clip": 1.08259273, "balance_loss_mlp": 1.04007149, "epoch": 0.04797691336499729, "flos": 18835723818240.0, "grad_norm": 2.6444378415740144, "language_loss": 0.88666236, "learning_rate": 3.996633071017783e-06, "loss": 0.91021013, "num_input_tokens_seen": 8534085, "step": 399, "time_per_iteration": 2.55423903465271 }, { "auxiliary_loss_clip": 0.01296558, "auxiliary_loss_mlp": 0.01051096, "balance_loss_clip": 1.08106732, "balance_loss_mlp": 1.03713644, "epoch": 0.04809715625563638, "flos": 21099676568640.0, "grad_norm": 2.393809295189033, "language_loss": 0.81729567, "learning_rate": 3.996587738582084e-06, "loss": 0.84077221, "num_input_tokens_seen": 8550885, "step": 400, "time_per_iteration": 2.533284902572632 }, { "auxiliary_loss_clip": 0.01295729, "auxiliary_loss_mlp": 0.01045166, "balance_loss_clip": 1.07926702, "balance_loss_mlp": 1.03213632, "epoch": 0.04821739914627548, "flos": 23805654519840.0, "grad_norm": 2.2622518465203907, "language_loss": 0.85924876, "learning_rate": 3.9965421032656115e-06, "loss": 0.88265771, "num_input_tokens_seen": 8570815, "step": 401, "time_per_iteration": 2.5787980556488037 }, { "auxiliary_loss_clip": 0.01302078, "auxiliary_loss_mlp": 0.01049118, "balance_loss_clip": 1.08283305, "balance_loss_mlp": 1.03578436, "epoch": 0.04833764203691457, "flos": 22200591161760.0, "grad_norm": 2.34371165599114, "language_loss": 0.94307858, "learning_rate": 3.99649616507529e-06, "loss": 0.96659052, "num_input_tokens_seen": 8589910, "step": 402, "time_per_iteration": 2.5339012145996094 }, { "auxiliary_loss_clip": 0.01209783, "auxiliary_loss_mlp": 0.01005194, "balance_loss_clip": 1.05899644, "balance_loss_mlp": 0.99949569, "epoch": 0.04845788492755366, "flos": 65904387500160.0, "grad_norm": 0.887475069143546, "language_loss": 0.63140571, "learning_rate": 3.996449924018088e-06, "loss": 0.65355551, "num_input_tokens_seen": 8650370, "step": 403, "time_per_iteration": 3.113971710205078 }, { "auxiliary_loss_clip": 0.01298324, "auxiliary_loss_mlp": 0.01044811, "balance_loss_clip": 1.08090603, "balance_loss_mlp": 1.03235376, "epoch": 0.04857812781819275, "flos": 19281304798080.0, "grad_norm": 1.9951156127676841, "language_loss": 0.79381573, "learning_rate": 3.99640338010102e-06, "loss": 0.81724709, "num_input_tokens_seen": 8669475, "step": 404, "time_per_iteration": 2.564493179321289 }, { "auxiliary_loss_clip": 0.01294284, "auxiliary_loss_mlp": 0.01042885, "balance_loss_clip": 1.0782994, "balance_loss_mlp": 1.0294559, "epoch": 0.04869837070883184, "flos": 24062380148640.0, "grad_norm": 1.9678792280884478, "language_loss": 0.78333509, "learning_rate": 3.996356533331146e-06, "loss": 0.80670679, "num_input_tokens_seen": 8691345, "step": 405, "time_per_iteration": 2.5771560668945312 }, { "auxiliary_loss_clip": 0.01301947, "auxiliary_loss_mlp": 0.01046195, "balance_loss_clip": 1.07888103, "balance_loss_mlp": 1.03126979, "epoch": 0.04881861359947093, "flos": 25187173131840.0, "grad_norm": 2.2738339593628094, "language_loss": 0.61929774, "learning_rate": 3.996309383715573e-06, "loss": 0.64277911, "num_input_tokens_seen": 8710125, "step": 406, "time_per_iteration": 2.617001533508301 }, { "auxiliary_loss_clip": 0.01297076, "auxiliary_loss_mlp": 0.01047503, "balance_loss_clip": 1.07971454, "balance_loss_mlp": 1.03356791, "epoch": 0.048938856490110025, "flos": 16362916358400.0, "grad_norm": 2.2650979985068727, "language_loss": 0.74054825, "learning_rate": 3.996261931261454e-06, "loss": 0.76399404, "num_input_tokens_seen": 8728705, "step": 407, "time_per_iteration": 2.544095516204834 }, { "auxiliary_loss_clip": 0.0130071, "auxiliary_loss_mlp": 0.010433, "balance_loss_clip": 1.08405709, "balance_loss_mlp": 1.02996612, "epoch": 0.049059099380749115, "flos": 29895098411520.0, "grad_norm": 1.7152888231775527, "language_loss": 0.86371613, "learning_rate": 3.996214175975987e-06, "loss": 0.88715625, "num_input_tokens_seen": 8749225, "step": 408, "time_per_iteration": 2.6488089561462402 }, { "auxiliary_loss_clip": 0.0130411, "auxiliary_loss_mlp": 0.01045289, "balance_loss_clip": 1.0857203, "balance_loss_mlp": 1.03200352, "epoch": 0.049179342271388204, "flos": 35918862960000.0, "grad_norm": 3.0328962566272293, "language_loss": 0.78719836, "learning_rate": 3.996166117866417e-06, "loss": 0.81069231, "num_input_tokens_seen": 8771160, "step": 409, "time_per_iteration": 3.437150239944458 }, { "auxiliary_loss_clip": 0.01292661, "auxiliary_loss_mlp": 0.01053706, "balance_loss_clip": 1.07632053, "balance_loss_mlp": 1.04124284, "epoch": 0.049299585162027294, "flos": 14611229605920.0, "grad_norm": 3.5246306089708312, "language_loss": 0.86625671, "learning_rate": 3.996117756940035e-06, "loss": 0.88972038, "num_input_tokens_seen": 8787845, "step": 410, "time_per_iteration": 2.563054323196411 }, { "auxiliary_loss_clip": 0.01294377, "auxiliary_loss_mlp": 0.01042601, "balance_loss_clip": 1.07897663, "balance_loss_mlp": 1.0290705, "epoch": 0.049419828052666384, "flos": 19567942866240.0, "grad_norm": 4.443720102494964, "language_loss": 0.97738159, "learning_rate": 3.996069093204175e-06, "loss": 1.00075126, "num_input_tokens_seen": 8803805, "step": 411, "time_per_iteration": 3.276668071746826 }, { "auxiliary_loss_clip": 0.01302953, "auxiliary_loss_mlp": 0.01058635, "balance_loss_clip": 1.08529258, "balance_loss_mlp": 1.04421043, "epoch": 0.049540070943305474, "flos": 13659919514880.0, "grad_norm": 2.6606015758612407, "language_loss": 0.88141394, "learning_rate": 3.996020126666221e-06, "loss": 0.90502977, "num_input_tokens_seen": 8820785, "step": 412, "time_per_iteration": 4.349470376968384 }, { "auxiliary_loss_clip": 0.01298027, "auxiliary_loss_mlp": 0.01049824, "balance_loss_clip": 1.08112431, "balance_loss_mlp": 1.03708625, "epoch": 0.04966031383394457, "flos": 21832039284480.0, "grad_norm": 1.9445035219028501, "language_loss": 0.81935567, "learning_rate": 3.995970857333601e-06, "loss": 0.84283417, "num_input_tokens_seen": 8841195, "step": 413, "time_per_iteration": 2.6244382858276367 }, { "auxiliary_loss_clip": 0.01298615, "auxiliary_loss_mlp": 0.01047522, "balance_loss_clip": 1.08070183, "balance_loss_mlp": 1.03301394, "epoch": 0.04978055672458366, "flos": 28618797327360.0, "grad_norm": 1.8843002781590064, "language_loss": 0.79673159, "learning_rate": 3.995921285213789e-06, "loss": 0.82019305, "num_input_tokens_seen": 8861455, "step": 414, "time_per_iteration": 2.6211471557617188 }, { "auxiliary_loss_clip": 0.01295158, "auxiliary_loss_mlp": 0.01049371, "balance_loss_clip": 1.0800643, "balance_loss_mlp": 1.03607929, "epoch": 0.04990079961522275, "flos": 19828224274080.0, "grad_norm": 2.2899760662356257, "language_loss": 0.80784726, "learning_rate": 3.995871410314305e-06, "loss": 0.83129251, "num_input_tokens_seen": 8880015, "step": 415, "time_per_iteration": 2.576106548309326 }, { "auxiliary_loss_clip": 0.01186082, "auxiliary_loss_mlp": 0.01004322, "balance_loss_clip": 1.05416155, "balance_loss_mlp": 0.99933857, "epoch": 0.05002104250586184, "flos": 62735056062240.0, "grad_norm": 0.9177902995596864, "language_loss": 0.5975548, "learning_rate": 3.995821232642714e-06, "loss": 0.61945885, "num_input_tokens_seen": 8938420, "step": 416, "time_per_iteration": 3.258617877960205 }, { "auxiliary_loss_clip": 0.0128208, "auxiliary_loss_mlp": 0.01047107, "balance_loss_clip": 1.08251369, "balance_loss_mlp": 1.03379118, "epoch": 0.05014128539650093, "flos": 27928522900320.0, "grad_norm": 2.382949273187456, "language_loss": 0.82167792, "learning_rate": 3.995770752206629e-06, "loss": 0.84496975, "num_input_tokens_seen": 8959495, "step": 417, "time_per_iteration": 2.6585676670074463 }, { "auxiliary_loss_clip": 0.01297624, "auxiliary_loss_mlp": 0.010495, "balance_loss_clip": 1.08168399, "balance_loss_mlp": 1.03567183, "epoch": 0.05026152828714002, "flos": 17705579208000.0, "grad_norm": 1.9102096636915462, "language_loss": 0.97123182, "learning_rate": 3.995719969013709e-06, "loss": 0.99470311, "num_input_tokens_seen": 8976675, "step": 418, "time_per_iteration": 2.574110269546509 }, { "auxiliary_loss_clip": 0.01261582, "auxiliary_loss_mlp": 0.01045049, "balance_loss_clip": 1.0773977, "balance_loss_mlp": 1.03164983, "epoch": 0.05038177117777912, "flos": 19133280642240.0, "grad_norm": 2.3937757536635087, "language_loss": 0.8560577, "learning_rate": 3.995668883071655e-06, "loss": 0.87912399, "num_input_tokens_seen": 8992900, "step": 419, "time_per_iteration": 2.602393627166748 }, { "auxiliary_loss_clip": 0.01298374, "auxiliary_loss_mlp": 0.01046278, "balance_loss_clip": 1.08215094, "balance_loss_mlp": 1.03247356, "epoch": 0.050502014068418206, "flos": 20667708116640.0, "grad_norm": 2.77901707599246, "language_loss": 0.90923619, "learning_rate": 3.995617494388219e-06, "loss": 0.93268269, "num_input_tokens_seen": 9011020, "step": 420, "time_per_iteration": 2.623922348022461 }, { "auxiliary_loss_clip": 0.01261526, "auxiliary_loss_mlp": 0.01052351, "balance_loss_clip": 1.07374036, "balance_loss_mlp": 1.03825438, "epoch": 0.050622256959057296, "flos": 21361107319200.0, "grad_norm": 2.3116061948600444, "language_loss": 0.80392432, "learning_rate": 3.995565802971196e-06, "loss": 0.82706314, "num_input_tokens_seen": 9030995, "step": 421, "time_per_iteration": 2.6328372955322266 }, { "auxiliary_loss_clip": 0.01257567, "auxiliary_loss_mlp": 0.01046115, "balance_loss_clip": 1.07381892, "balance_loss_mlp": 1.03353834, "epoch": 0.050742499849696386, "flos": 27673593119520.0, "grad_norm": 1.807192346640251, "language_loss": 0.67492747, "learning_rate": 3.995513808828427e-06, "loss": 0.69796431, "num_input_tokens_seen": 9053790, "step": 422, "time_per_iteration": 2.7012948989868164 }, { "auxiliary_loss_clip": 0.01262855, "auxiliary_loss_mlp": 0.01054741, "balance_loss_clip": 1.07558239, "balance_loss_mlp": 1.04122281, "epoch": 0.050862742740335476, "flos": 19866002527680.0, "grad_norm": 2.953806645685485, "language_loss": 0.76618624, "learning_rate": 3.9954615119678e-06, "loss": 0.78936213, "num_input_tokens_seen": 9072345, "step": 423, "time_per_iteration": 2.6251423358917236 }, { "auxiliary_loss_clip": 0.01262487, "auxiliary_loss_mlp": 0.01048713, "balance_loss_clip": 1.07454777, "balance_loss_mlp": 1.035689, "epoch": 0.050982985630974566, "flos": 22085101383360.0, "grad_norm": 2.0990339551009507, "language_loss": 0.80583048, "learning_rate": 3.995408912397248e-06, "loss": 0.82894242, "num_input_tokens_seen": 9090240, "step": 424, "time_per_iteration": 2.6152801513671875 }, { "auxiliary_loss_clip": 0.01262455, "auxiliary_loss_mlp": 0.01054013, "balance_loss_clip": 1.07735693, "balance_loss_mlp": 1.04029155, "epoch": 0.05110322852161366, "flos": 20740965938400.0, "grad_norm": 2.326584677030527, "language_loss": 0.92891383, "learning_rate": 3.99535601012475e-06, "loss": 0.95207846, "num_input_tokens_seen": 9105570, "step": 425, "time_per_iteration": 2.5979502201080322 }, { "auxiliary_loss_clip": 0.01243182, "auxiliary_loss_mlp": 0.0076757, "balance_loss_clip": 1.07494879, "balance_loss_mlp": 1.00048184, "epoch": 0.05122347141225275, "flos": 28547299436640.0, "grad_norm": 1.5994301168999916, "language_loss": 0.75556558, "learning_rate": 3.995302805158333e-06, "loss": 0.77567309, "num_input_tokens_seen": 9128225, "step": 426, "time_per_iteration": 2.7415690422058105 }, { "auxiliary_loss_clip": 0.01253892, "auxiliary_loss_mlp": 0.01044453, "balance_loss_clip": 1.07496178, "balance_loss_mlp": 1.03039837, "epoch": 0.05134371430289184, "flos": 19722683493600.0, "grad_norm": 1.8286766143583897, "language_loss": 0.83496916, "learning_rate": 3.9952492975060665e-06, "loss": 0.85795254, "num_input_tokens_seen": 9148295, "step": 427, "time_per_iteration": 2.634516716003418 }, { "auxiliary_loss_clip": 0.01275256, "auxiliary_loss_mlp": 0.01050441, "balance_loss_clip": 1.07732677, "balance_loss_mlp": 1.03678513, "epoch": 0.05146395719353093, "flos": 34458950400960.0, "grad_norm": 2.7541023056412466, "language_loss": 0.84804952, "learning_rate": 3.995195487176067e-06, "loss": 0.87130642, "num_input_tokens_seen": 9168525, "step": 428, "time_per_iteration": 2.6858997344970703 }, { "auxiliary_loss_clip": 0.01295594, "auxiliary_loss_mlp": 0.01057989, "balance_loss_clip": 1.0798161, "balance_loss_mlp": 1.04485226, "epoch": 0.05158420008417002, "flos": 21760289975040.0, "grad_norm": 1.8356162033430714, "language_loss": 0.8545596, "learning_rate": 3.995141374176499e-06, "loss": 0.87809539, "num_input_tokens_seen": 9186920, "step": 429, "time_per_iteration": 2.5675694942474365 }, { "auxiliary_loss_clip": 0.01150619, "auxiliary_loss_mlp": 0.00759199, "balance_loss_clip": 1.04467487, "balance_loss_mlp": 1.00023448, "epoch": 0.05170444297480911, "flos": 72553968309120.0, "grad_norm": 0.8748586401815318, "language_loss": 0.63074261, "learning_rate": 3.995086958515572e-06, "loss": 0.64984083, "num_input_tokens_seen": 9244940, "step": 430, "time_per_iteration": 3.22687029838562 }, { "auxiliary_loss_clip": 0.01189242, "auxiliary_loss_mlp": 0.00759032, "balance_loss_clip": 1.04852152, "balance_loss_mlp": 1.00008285, "epoch": 0.05182468586544821, "flos": 62416170952320.0, "grad_norm": 0.867180214497582, "language_loss": 0.60003483, "learning_rate": 3.995032240201538e-06, "loss": 0.61951756, "num_input_tokens_seen": 9307335, "step": 431, "time_per_iteration": 3.099756956100464 }, { "auxiliary_loss_clip": 0.0116621, "auxiliary_loss_mlp": 0.01008772, "balance_loss_clip": 1.04456353, "balance_loss_mlp": 1.00400352, "epoch": 0.0519449287560873, "flos": 41225996421120.0, "grad_norm": 0.9380680813784061, "language_loss": 0.63172704, "learning_rate": 3.9949772192427e-06, "loss": 0.65347689, "num_input_tokens_seen": 9353960, "step": 432, "time_per_iteration": 2.8797335624694824 }, { "auxiliary_loss_clip": 0.01260269, "auxiliary_loss_mlp": 0.01046539, "balance_loss_clip": 1.07546115, "balance_loss_mlp": 1.03437376, "epoch": 0.05206517164672639, "flos": 17494533564000.0, "grad_norm": 2.36761283360385, "language_loss": 0.79521215, "learning_rate": 3.994921895647405e-06, "loss": 0.81828022, "num_input_tokens_seen": 9372130, "step": 433, "time_per_iteration": 2.6303887367248535 }, { "auxiliary_loss_clip": 0.01185327, "auxiliary_loss_mlp": 0.01008633, "balance_loss_clip": 1.04646111, "balance_loss_mlp": 1.00381744, "epoch": 0.05218541453736548, "flos": 64002772992960.0, "grad_norm": 0.8336910141001878, "language_loss": 0.55328047, "learning_rate": 3.994866269424043e-06, "loss": 0.57522005, "num_input_tokens_seen": 9428500, "step": 434, "time_per_iteration": 3.0524895191192627 }, { "auxiliary_loss_clip": 0.01210891, "auxiliary_loss_mlp": 0.01046587, "balance_loss_clip": 1.06337953, "balance_loss_mlp": 1.03373075, "epoch": 0.05230565742800457, "flos": 19317323120640.0, "grad_norm": 2.674806081750402, "language_loss": 0.78503931, "learning_rate": 3.9948103405810545e-06, "loss": 0.80761409, "num_input_tokens_seen": 9447450, "step": 435, "time_per_iteration": 3.4178388118743896 }, { "auxiliary_loss_clip": 0.01223575, "auxiliary_loss_mlp": 0.01052874, "balance_loss_clip": 1.06860566, "balance_loss_mlp": 1.03916514, "epoch": 0.05242590031864366, "flos": 25298640210720.0, "grad_norm": 2.188055357971606, "language_loss": 0.85903442, "learning_rate": 3.994754109126923e-06, "loss": 0.88179892, "num_input_tokens_seen": 9468945, "step": 436, "time_per_iteration": 2.6789627075195312 }, { "auxiliary_loss_clip": 0.01201853, "auxiliary_loss_mlp": 0.01054644, "balance_loss_clip": 1.06667638, "balance_loss_mlp": 1.04124475, "epoch": 0.052546143209282754, "flos": 26211633293760.0, "grad_norm": 1.6650974112145152, "language_loss": 0.93489486, "learning_rate": 3.994697575070181e-06, "loss": 0.95745981, "num_input_tokens_seen": 9488405, "step": 437, "time_per_iteration": 3.4537301063537598 }, { "auxiliary_loss_clip": 0.01260496, "auxiliary_loss_mlp": 0.01048176, "balance_loss_clip": 1.0804472, "balance_loss_mlp": 1.03529572, "epoch": 0.052666386099921844, "flos": 22158143703360.0, "grad_norm": 2.0701935125167803, "language_loss": 0.91454577, "learning_rate": 3.994640738419402e-06, "loss": 0.93763244, "num_input_tokens_seen": 9507780, "step": 438, "time_per_iteration": 3.516934633255005 }, { "auxiliary_loss_clip": 0.01277497, "auxiliary_loss_mlp": 0.01050243, "balance_loss_clip": 1.07881784, "balance_loss_mlp": 1.03844738, "epoch": 0.052786628990560934, "flos": 23881821615360.0, "grad_norm": 1.9248209736898931, "language_loss": 0.80788183, "learning_rate": 3.9945835991832075e-06, "loss": 0.83115923, "num_input_tokens_seen": 9529665, "step": 439, "time_per_iteration": 2.6526801586151123 }, { "auxiliary_loss_clip": 0.0129526, "auxiliary_loss_mlp": 0.01053328, "balance_loss_clip": 1.08334947, "balance_loss_mlp": 1.04181874, "epoch": 0.052906871881200024, "flos": 24605025506400.0, "grad_norm": 2.0720114563350713, "language_loss": 0.92826605, "learning_rate": 3.994526157370268e-06, "loss": 0.95175195, "num_input_tokens_seen": 9548280, "step": 440, "time_per_iteration": 2.591693639755249 }, { "auxiliary_loss_clip": 0.01156166, "auxiliary_loss_mlp": 0.01004472, "balance_loss_clip": 1.03741753, "balance_loss_mlp": 0.99998951, "epoch": 0.053027114771839114, "flos": 56461641526080.0, "grad_norm": 0.8875074134308786, "language_loss": 0.59297705, "learning_rate": 3.994468412989296e-06, "loss": 0.61458337, "num_input_tokens_seen": 9609690, "step": 441, "time_per_iteration": 3.300382137298584 }, { "auxiliary_loss_clip": 0.01231266, "auxiliary_loss_mlp": 0.01040922, "balance_loss_clip": 1.06750357, "balance_loss_mlp": 1.0284462, "epoch": 0.053147357662478203, "flos": 17311101673920.0, "grad_norm": 1.902341320524787, "language_loss": 0.92536026, "learning_rate": 3.994410366049052e-06, "loss": 0.94808209, "num_input_tokens_seen": 9627550, "step": 442, "time_per_iteration": 2.5947396755218506 }, { "auxiliary_loss_clip": 0.01275074, "auxiliary_loss_mlp": 0.01042867, "balance_loss_clip": 1.07796741, "balance_loss_mlp": 1.03012335, "epoch": 0.0532676005531173, "flos": 17164981116960.0, "grad_norm": 2.5266209660532435, "language_loss": 0.83045781, "learning_rate": 3.994352016558341e-06, "loss": 0.85363716, "num_input_tokens_seen": 9644855, "step": 443, "time_per_iteration": 2.5908749103546143 }, { "auxiliary_loss_clip": 0.01272818, "auxiliary_loss_mlp": 0.01048986, "balance_loss_clip": 1.07855058, "balance_loss_mlp": 1.036183, "epoch": 0.05338784344375639, "flos": 27819965095200.0, "grad_norm": 2.057491425728281, "language_loss": 0.73811871, "learning_rate": 3.994293364526014e-06, "loss": 0.76133674, "num_input_tokens_seen": 9665740, "step": 444, "time_per_iteration": 2.6255626678466797 }, { "auxiliary_loss_clip": 0.01242844, "auxiliary_loss_mlp": 0.01047846, "balance_loss_clip": 1.07216549, "balance_loss_mlp": 1.03453684, "epoch": 0.05350808633439548, "flos": 21507694796640.0, "grad_norm": 2.1704684803872527, "language_loss": 0.84408754, "learning_rate": 3.99423440996097e-06, "loss": 0.86699444, "num_input_tokens_seen": 9685280, "step": 445, "time_per_iteration": 2.6616437435150146 }, { "auxiliary_loss_clip": 0.01265264, "auxiliary_loss_mlp": 0.01046429, "balance_loss_clip": 1.08135486, "balance_loss_mlp": 1.03366148, "epoch": 0.05362832922503457, "flos": 20084303397600.0, "grad_norm": 2.163750055538747, "language_loss": 0.81479597, "learning_rate": 3.994175152872152e-06, "loss": 0.8379128, "num_input_tokens_seen": 9704365, "step": 446, "time_per_iteration": 2.609626531600952 }, { "auxiliary_loss_clip": 0.01279013, "auxiliary_loss_mlp": 0.0104777, "balance_loss_clip": 1.07698798, "balance_loss_mlp": 1.03581309, "epoch": 0.05374857211567366, "flos": 26137226129280.0, "grad_norm": 2.9788508171502026, "language_loss": 0.78404462, "learning_rate": 3.994115593268548e-06, "loss": 0.80731249, "num_input_tokens_seen": 9724145, "step": 447, "time_per_iteration": 2.6313936710357666 }, { "auxiliary_loss_clip": 0.01290764, "auxiliary_loss_mlp": 0.01050543, "balance_loss_clip": 1.08029175, "balance_loss_mlp": 1.03776932, "epoch": 0.05386881500631275, "flos": 27486820952160.0, "grad_norm": 2.2000243905427816, "language_loss": 0.82207954, "learning_rate": 3.994055731159195e-06, "loss": 0.8454926, "num_input_tokens_seen": 9741615, "step": 448, "time_per_iteration": 2.59897780418396 }, { "auxiliary_loss_clip": 0.01276584, "auxiliary_loss_mlp": 0.01033615, "balance_loss_clip": 1.08096361, "balance_loss_mlp": 1.02208769, "epoch": 0.053989057896951846, "flos": 23585090881440.0, "grad_norm": 1.9943649514927893, "language_loss": 0.870121, "learning_rate": 3.993995566553172e-06, "loss": 0.89322305, "num_input_tokens_seen": 9760580, "step": 449, "time_per_iteration": 2.6130459308624268 }, { "auxiliary_loss_clip": 0.01242387, "auxiliary_loss_mlp": 0.01045866, "balance_loss_clip": 1.06764507, "balance_loss_mlp": 1.03292012, "epoch": 0.054109300787590936, "flos": 25228866334080.0, "grad_norm": 1.6648560807439245, "language_loss": 0.77276707, "learning_rate": 3.993935099459607e-06, "loss": 0.79564965, "num_input_tokens_seen": 9782195, "step": 450, "time_per_iteration": 2.640259027481079 }, { "auxiliary_loss_clip": 0.01282151, "auxiliary_loss_mlp": 0.01045949, "balance_loss_clip": 1.07568574, "balance_loss_mlp": 1.03461766, "epoch": 0.054229543678230026, "flos": 23841529174560.0, "grad_norm": 2.0844292162046463, "language_loss": 0.74058068, "learning_rate": 3.993874329887673e-06, "loss": 0.76386166, "num_input_tokens_seen": 9800850, "step": 451, "time_per_iteration": 2.58235502243042 }, { "auxiliary_loss_clip": 0.01273041, "auxiliary_loss_mlp": 0.01036638, "balance_loss_clip": 1.07684588, "balance_loss_mlp": 1.02443051, "epoch": 0.054349786568869116, "flos": 16320935820480.0, "grad_norm": 2.7330169678088203, "language_loss": 0.86163145, "learning_rate": 3.993813257846589e-06, "loss": 0.88472831, "num_input_tokens_seen": 9817605, "step": 452, "time_per_iteration": 2.5729880332946777 }, { "auxiliary_loss_clip": 0.01278899, "auxiliary_loss_mlp": 0.01063295, "balance_loss_clip": 1.07989311, "balance_loss_mlp": 1.05070627, "epoch": 0.054470029459508205, "flos": 18660732413760.0, "grad_norm": 2.1902032358531063, "language_loss": 0.92832971, "learning_rate": 3.993751883345619e-06, "loss": 0.95175171, "num_input_tokens_seen": 9835965, "step": 453, "time_per_iteration": 2.5996387004852295 }, { "auxiliary_loss_clip": 0.01255967, "auxiliary_loss_mlp": 0.01041474, "balance_loss_clip": 1.077299, "balance_loss_mlp": 1.0285641, "epoch": 0.054590272350147295, "flos": 17785302082560.0, "grad_norm": 2.42134679641757, "language_loss": 0.8775847, "learning_rate": 3.993690206394073e-06, "loss": 0.90055919, "num_input_tokens_seen": 9852265, "step": 454, "time_per_iteration": 2.614063262939453 }, { "auxiliary_loss_clip": 0.01266244, "auxiliary_loss_mlp": 0.0104229, "balance_loss_clip": 1.07712829, "balance_loss_mlp": 1.03057182, "epoch": 0.054710515240786385, "flos": 17785948587840.0, "grad_norm": 2.4558455105157977, "language_loss": 0.87601876, "learning_rate": 3.993628227001307e-06, "loss": 0.89910412, "num_input_tokens_seen": 9870465, "step": 455, "time_per_iteration": 2.6057381629943848 }, { "auxiliary_loss_clip": 0.01254389, "auxiliary_loss_mlp": 0.01051094, "balance_loss_clip": 1.07242656, "balance_loss_mlp": 1.03852904, "epoch": 0.05483075813142548, "flos": 48210905919360.0, "grad_norm": 1.9070410339562516, "language_loss": 0.71492517, "learning_rate": 3.993565945176726e-06, "loss": 0.73798001, "num_input_tokens_seen": 9891490, "step": 456, "time_per_iteration": 2.867328643798828 }, { "auxiliary_loss_clip": 0.01255583, "auxiliary_loss_mlp": 0.01050987, "balance_loss_clip": 1.07616472, "balance_loss_mlp": 1.03889287, "epoch": 0.05495100102206457, "flos": 19682247384960.0, "grad_norm": 2.6725283493098786, "language_loss": 0.84036756, "learning_rate": 3.993503360929776e-06, "loss": 0.86343336, "num_input_tokens_seen": 9910375, "step": 457, "time_per_iteration": 2.6067237854003906 }, { "auxiliary_loss_clip": 0.01190176, "auxiliary_loss_mlp": 0.01050754, "balance_loss_clip": 1.06630278, "balance_loss_mlp": 1.03777838, "epoch": 0.05507124391270366, "flos": 26360052536160.0, "grad_norm": 1.6162119954393033, "language_loss": 0.80897367, "learning_rate": 3.99344047426995e-06, "loss": 0.83138299, "num_input_tokens_seen": 9931635, "step": 458, "time_per_iteration": 3.014309883117676 }, { "auxiliary_loss_clip": 0.01230692, "auxiliary_loss_mlp": 0.01050821, "balance_loss_clip": 1.07169056, "balance_loss_mlp": 1.03837585, "epoch": 0.05519148680334275, "flos": 22601318246880.0, "grad_norm": 2.176036058041596, "language_loss": 0.93462992, "learning_rate": 3.993377285206789e-06, "loss": 0.95744514, "num_input_tokens_seen": 9951420, "step": 459, "time_per_iteration": 2.965991735458374 }, { "auxiliary_loss_clip": 0.01213269, "auxiliary_loss_mlp": 0.01043542, "balance_loss_clip": 1.06698024, "balance_loss_mlp": 1.03053617, "epoch": 0.05531172969398184, "flos": 40552524743040.0, "grad_norm": 1.7983641644299064, "language_loss": 0.86425519, "learning_rate": 3.99331379374988e-06, "loss": 0.8868233, "num_input_tokens_seen": 9975025, "step": 460, "time_per_iteration": 2.8075363636016846 }, { "auxiliary_loss_clip": 0.01256184, "auxiliary_loss_mlp": 0.01048093, "balance_loss_clip": 1.06839037, "balance_loss_mlp": 1.03524208, "epoch": 0.05543197258462093, "flos": 23477897920800.0, "grad_norm": 2.2989042596186264, "language_loss": 0.8003667, "learning_rate": 3.993249999908852e-06, "loss": 0.82340944, "num_input_tokens_seen": 9995175, "step": 461, "time_per_iteration": 3.4198412895202637 }, { "auxiliary_loss_clip": 0.01286275, "auxiliary_loss_mlp": 0.01044, "balance_loss_clip": 1.07723594, "balance_loss_mlp": 1.03168619, "epoch": 0.05555221547526003, "flos": 18624606340320.0, "grad_norm": 2.0105730801417625, "language_loss": 0.87005025, "learning_rate": 3.993185903693384e-06, "loss": 0.89335299, "num_input_tokens_seen": 10011975, "step": 462, "time_per_iteration": 2.5316007137298584 }, { "auxiliary_loss_clip": 0.01256336, "auxiliary_loss_mlp": 0.01042244, "balance_loss_clip": 1.07459259, "balance_loss_mlp": 1.0299952, "epoch": 0.05567245836589912, "flos": 23587102231200.0, "grad_norm": 2.3589413490426, "language_loss": 0.8226428, "learning_rate": 3.9931215051131995e-06, "loss": 0.84562862, "num_input_tokens_seen": 10032620, "step": 463, "time_per_iteration": 3.455369472503662 }, { "auxiliary_loss_clip": 0.01261908, "auxiliary_loss_mlp": 0.01047491, "balance_loss_clip": 1.07286239, "balance_loss_mlp": 1.03503323, "epoch": 0.05579270125653821, "flos": 27746671356480.0, "grad_norm": 1.6007291579757563, "language_loss": 0.7994352, "learning_rate": 3.993056804178068e-06, "loss": 0.8225292, "num_input_tokens_seen": 10054165, "step": 464, "time_per_iteration": 3.568784713745117 }, { "auxiliary_loss_clip": 0.01217229, "auxiliary_loss_mlp": 0.0105331, "balance_loss_clip": 1.07016408, "balance_loss_mlp": 1.04097748, "epoch": 0.0559129441471773, "flos": 27014164972800.0, "grad_norm": 2.1228413466603575, "language_loss": 0.84262449, "learning_rate": 3.992991800897803e-06, "loss": 0.86532986, "num_input_tokens_seen": 10073970, "step": 465, "time_per_iteration": 3.611203193664551 }, { "auxiliary_loss_clip": 0.01287133, "auxiliary_loss_mlp": 0.01039707, "balance_loss_clip": 1.07863057, "balance_loss_mlp": 1.02778053, "epoch": 0.05603318703781639, "flos": 15229790640480.0, "grad_norm": 2.1736701673079826, "language_loss": 0.89833701, "learning_rate": 3.9929264952822665e-06, "loss": 0.92160541, "num_input_tokens_seen": 10091505, "step": 466, "time_per_iteration": 2.5399117469787598 }, { "auxiliary_loss_clip": 0.01277706, "auxiliary_loss_mlp": 0.01056955, "balance_loss_clip": 1.07692504, "balance_loss_mlp": 1.04393125, "epoch": 0.05615342992845548, "flos": 22266486006720.0, "grad_norm": 1.8796468502940893, "language_loss": 0.8810128, "learning_rate": 3.992860887341366e-06, "loss": 0.90435934, "num_input_tokens_seen": 10109675, "step": 467, "time_per_iteration": 2.625434160232544 }, { "auxiliary_loss_clip": 0.0122288, "auxiliary_loss_mlp": 0.01039348, "balance_loss_clip": 1.06618226, "balance_loss_mlp": 1.02801692, "epoch": 0.056273672819094574, "flos": 23584983130560.0, "grad_norm": 3.8626105894208442, "language_loss": 0.80966735, "learning_rate": 3.992794977085052e-06, "loss": 0.83228958, "num_input_tokens_seen": 10127675, "step": 468, "time_per_iteration": 2.6855573654174805 }, { "auxiliary_loss_clip": 0.01239879, "auxiliary_loss_mlp": 0.01054615, "balance_loss_clip": 1.07216907, "balance_loss_mlp": 1.04276001, "epoch": 0.056393915709733664, "flos": 19858711384800.0, "grad_norm": 1.9721220444869143, "language_loss": 0.84841096, "learning_rate": 3.992728764523326e-06, "loss": 0.87135589, "num_input_tokens_seen": 10146620, "step": 469, "time_per_iteration": 2.6589713096618652 }, { "auxiliary_loss_clip": 0.01249976, "auxiliary_loss_mlp": 0.01039376, "balance_loss_clip": 1.07413507, "balance_loss_mlp": 1.02682328, "epoch": 0.05651415860037275, "flos": 22163782666080.0, "grad_norm": 1.8052378869861958, "language_loss": 0.80787486, "learning_rate": 3.99266224966623e-06, "loss": 0.83076841, "num_input_tokens_seen": 10167535, "step": 470, "time_per_iteration": 2.6548335552215576 }, { "auxiliary_loss_clip": 0.01239221, "auxiliary_loss_mlp": 0.01048763, "balance_loss_clip": 1.07148147, "balance_loss_mlp": 1.03684258, "epoch": 0.05663440149101184, "flos": 19463551428480.0, "grad_norm": 1.9425708252669942, "language_loss": 0.87761873, "learning_rate": 3.992595432523855e-06, "loss": 0.90049857, "num_input_tokens_seen": 10184825, "step": 471, "time_per_iteration": 2.6107513904571533 }, { "auxiliary_loss_clip": 0.01229178, "auxiliary_loss_mlp": 0.01039415, "balance_loss_clip": 1.07236755, "balance_loss_mlp": 1.02733898, "epoch": 0.05675464438165093, "flos": 22670230116480.0, "grad_norm": 2.0886090021949713, "language_loss": 0.85979187, "learning_rate": 3.992528313106338e-06, "loss": 0.88247776, "num_input_tokens_seen": 10203025, "step": 472, "time_per_iteration": 2.652224540710449 }, { "auxiliary_loss_clip": 0.01286199, "auxiliary_loss_mlp": 0.00766488, "balance_loss_clip": 1.07933462, "balance_loss_mlp": 1.00024796, "epoch": 0.05687488727229002, "flos": 16901179847040.0, "grad_norm": 2.456680786333007, "language_loss": 0.82190287, "learning_rate": 3.9924608914238595e-06, "loss": 0.8424297, "num_input_tokens_seen": 10218020, "step": 473, "time_per_iteration": 2.5480945110321045 }, { "auxiliary_loss_clip": 0.01267763, "auxiliary_loss_mlp": 0.0104101, "balance_loss_clip": 1.07572269, "balance_loss_mlp": 1.02788496, "epoch": 0.05699513016292912, "flos": 29168985246720.0, "grad_norm": 3.2759729093786207, "language_loss": 0.84004712, "learning_rate": 3.992393167486648e-06, "loss": 0.86313486, "num_input_tokens_seen": 10237170, "step": 474, "time_per_iteration": 2.646667957305908 }, { "auxiliary_loss_clip": 0.01288257, "auxiliary_loss_mlp": 0.01048516, "balance_loss_clip": 1.07939029, "balance_loss_mlp": 1.03603482, "epoch": 0.05711537305356821, "flos": 18916200948960.0, "grad_norm": 2.4285517757323984, "language_loss": 0.8088665, "learning_rate": 3.992325141304977e-06, "loss": 0.83223432, "num_input_tokens_seen": 10255125, "step": 475, "time_per_iteration": 2.5652239322662354 }, { "auxiliary_loss_clip": 0.01222093, "auxiliary_loss_mlp": 0.01045897, "balance_loss_clip": 1.07146955, "balance_loss_mlp": 1.03439355, "epoch": 0.0572356159442073, "flos": 26758983773280.0, "grad_norm": 2.3021020165478343, "language_loss": 0.86587787, "learning_rate": 3.992256812889166e-06, "loss": 0.88855779, "num_input_tokens_seen": 10271230, "step": 476, "time_per_iteration": 2.6643128395080566 }, { "auxiliary_loss_clip": 0.01286302, "auxiliary_loss_mlp": 0.01037687, "balance_loss_clip": 1.08082771, "balance_loss_mlp": 1.0252893, "epoch": 0.05735585883484639, "flos": 35116151696160.0, "grad_norm": 2.073984542061311, "language_loss": 0.76888299, "learning_rate": 3.992188182249582e-06, "loss": 0.79212296, "num_input_tokens_seen": 10293125, "step": 477, "time_per_iteration": 2.7032032012939453 }, { "auxiliary_loss_clip": 0.01253891, "auxiliary_loss_mlp": 0.01044603, "balance_loss_clip": 1.07535052, "balance_loss_mlp": 1.03045297, "epoch": 0.05747610172548548, "flos": 18734385322080.0, "grad_norm": 2.5703474059273432, "language_loss": 0.90787625, "learning_rate": 3.992119249396633e-06, "loss": 0.93086123, "num_input_tokens_seen": 10311810, "step": 478, "time_per_iteration": 2.6784303188323975 }, { "auxiliary_loss_clip": 0.01247305, "auxiliary_loss_mlp": 0.0076729, "balance_loss_clip": 1.07124293, "balance_loss_mlp": 1.00029778, "epoch": 0.05759634461612457, "flos": 27964756724640.0, "grad_norm": 1.8902492194583171, "language_loss": 0.8205713, "learning_rate": 3.992050014340778e-06, "loss": 0.84071732, "num_input_tokens_seen": 10332165, "step": 479, "time_per_iteration": 2.7640492916107178 }, { "auxiliary_loss_clip": 0.01155143, "auxiliary_loss_mlp": 0.01013326, "balance_loss_clip": 1.03432178, "balance_loss_mlp": 1.00924957, "epoch": 0.057716587506763666, "flos": 69292019808000.0, "grad_norm": 0.8431773509686935, "language_loss": 0.55033898, "learning_rate": 3.99198047709252e-06, "loss": 0.57202375, "num_input_tokens_seen": 10393685, "step": 480, "time_per_iteration": 3.2574448585510254 }, { "auxiliary_loss_clip": 0.0123392, "auxiliary_loss_mlp": 0.01049033, "balance_loss_clip": 1.07015586, "balance_loss_mlp": 1.03592038, "epoch": 0.057836830397402755, "flos": 25009200619680.0, "grad_norm": 1.9653005241103354, "language_loss": 0.78541994, "learning_rate": 3.991910637662408e-06, "loss": 0.80824947, "num_input_tokens_seen": 10413975, "step": 481, "time_per_iteration": 2.7204277515411377 }, { "auxiliary_loss_clip": 0.01283942, "auxiliary_loss_mlp": 0.01043363, "balance_loss_clip": 1.07853341, "balance_loss_mlp": 1.03109622, "epoch": 0.057957073288041845, "flos": 25593898349280.0, "grad_norm": 2.2939004569840606, "language_loss": 0.80730021, "learning_rate": 3.9918404960610355e-06, "loss": 0.8305732, "num_input_tokens_seen": 10433005, "step": 482, "time_per_iteration": 2.60778546333313 }, { "auxiliary_loss_clip": 0.01270412, "auxiliary_loss_mlp": 0.01058946, "balance_loss_clip": 1.07569885, "balance_loss_mlp": 1.04686415, "epoch": 0.058077316178680935, "flos": 20777415264480.0, "grad_norm": 2.3038613220084345, "language_loss": 0.77602172, "learning_rate": 3.991770052299043e-06, "loss": 0.79931533, "num_input_tokens_seen": 10451235, "step": 483, "time_per_iteration": 2.627777338027954 }, { "auxiliary_loss_clip": 0.01253276, "auxiliary_loss_mlp": 0.01042432, "balance_loss_clip": 1.07228959, "balance_loss_mlp": 1.03034997, "epoch": 0.058197559069320025, "flos": 18916488284640.0, "grad_norm": 2.5446459469222913, "language_loss": 0.87944889, "learning_rate": 3.991699306387118e-06, "loss": 0.90240598, "num_input_tokens_seen": 10469705, "step": 484, "time_per_iteration": 2.616971015930176 }, { "auxiliary_loss_clip": 0.01273592, "auxiliary_loss_mlp": 0.01049848, "balance_loss_clip": 1.07651782, "balance_loss_mlp": 1.03856492, "epoch": 0.058317801959959115, "flos": 24863331481440.0, "grad_norm": 2.223990201774848, "language_loss": 0.77922273, "learning_rate": 3.991628258335991e-06, "loss": 0.80245715, "num_input_tokens_seen": 10491910, "step": 485, "time_per_iteration": 2.678868532180786 }, { "auxiliary_loss_clip": 0.0122798, "auxiliary_loss_mlp": 0.01047887, "balance_loss_clip": 1.06936097, "balance_loss_mlp": 1.03575754, "epoch": 0.05843804485059821, "flos": 23257980787680.0, "grad_norm": 4.3045522339423234, "language_loss": 0.87714565, "learning_rate": 3.991556908156442e-06, "loss": 0.89990431, "num_input_tokens_seen": 10508435, "step": 486, "time_per_iteration": 2.6515965461730957 }, { "auxiliary_loss_clip": 0.01253995, "auxiliary_loss_mlp": 0.01047315, "balance_loss_clip": 1.07163048, "balance_loss_mlp": 1.03522706, "epoch": 0.0585582877412373, "flos": 23150536408320.0, "grad_norm": 2.284811672948009, "language_loss": 0.877949, "learning_rate": 3.9914852558592914e-06, "loss": 0.90096211, "num_input_tokens_seen": 10529485, "step": 487, "time_per_iteration": 3.4077024459838867 }, { "auxiliary_loss_clip": 0.01269811, "auxiliary_loss_mlp": 0.01042247, "balance_loss_clip": 1.07841659, "balance_loss_mlp": 1.02981353, "epoch": 0.05867853063187639, "flos": 23506409598720.0, "grad_norm": 3.0423428435912845, "language_loss": 0.81314242, "learning_rate": 3.991413301455413e-06, "loss": 0.83626306, "num_input_tokens_seen": 10545935, "step": 488, "time_per_iteration": 2.6381471157073975 }, { "auxiliary_loss_clip": 0.01236167, "auxiliary_loss_mlp": 0.01046929, "balance_loss_clip": 1.06848025, "balance_loss_mlp": 1.03579497, "epoch": 0.05879877352251548, "flos": 29495808004800.0, "grad_norm": 2.26013485551387, "language_loss": 0.77765954, "learning_rate": 3.991341044955719e-06, "loss": 0.8004905, "num_input_tokens_seen": 10565690, "step": 489, "time_per_iteration": 3.396557092666626 }, { "auxiliary_loss_clip": 0.01266455, "auxiliary_loss_mlp": 0.00767356, "balance_loss_clip": 1.07246375, "balance_loss_mlp": 1.0004698, "epoch": 0.05891901641315457, "flos": 20157489385440.0, "grad_norm": 2.14758371239443, "language_loss": 0.81978488, "learning_rate": 3.991268486371172e-06, "loss": 0.840123, "num_input_tokens_seen": 10584245, "step": 490, "time_per_iteration": 2.625828981399536 }, { "auxiliary_loss_clip": 0.01251006, "auxiliary_loss_mlp": 0.01048735, "balance_loss_clip": 1.07076287, "balance_loss_mlp": 1.03571165, "epoch": 0.05903925930379366, "flos": 24644204521440.0, "grad_norm": 2.9637434720490674, "language_loss": 0.87810493, "learning_rate": 3.991195625712779e-06, "loss": 0.9011023, "num_input_tokens_seen": 10601210, "step": 491, "time_per_iteration": 4.404514789581299 }, { "auxiliary_loss_clip": 0.01282524, "auxiliary_loss_mlp": 0.01046501, "balance_loss_clip": 1.07706368, "balance_loss_mlp": 1.03418016, "epoch": 0.05915950219443276, "flos": 21250394496480.0, "grad_norm": 2.2306776099500376, "language_loss": 0.81285107, "learning_rate": 3.991122462991592e-06, "loss": 0.83614129, "num_input_tokens_seen": 10620730, "step": 492, "time_per_iteration": 2.5794947147369385 }, { "auxiliary_loss_clip": 0.01289023, "auxiliary_loss_mlp": 0.0104733, "balance_loss_clip": 1.07875562, "balance_loss_mlp": 1.0347774, "epoch": 0.05927974508507185, "flos": 9902729654880.0, "grad_norm": 3.9504392817159117, "language_loss": 0.80772537, "learning_rate": 3.991048998218712e-06, "loss": 0.8310889, "num_input_tokens_seen": 10634035, "step": 493, "time_per_iteration": 2.5431132316589355 }, { "auxiliary_loss_clip": 0.01268674, "auxiliary_loss_mlp": 0.01045002, "balance_loss_clip": 1.07202458, "balance_loss_mlp": 1.03211522, "epoch": 0.05939998797571094, "flos": 18259502491200.0, "grad_norm": 3.745840387120359, "language_loss": 0.76814902, "learning_rate": 3.990975231405281e-06, "loss": 0.79128575, "num_input_tokens_seen": 10652485, "step": 494, "time_per_iteration": 2.587059259414673 }, { "auxiliary_loss_clip": 0.01259774, "auxiliary_loss_mlp": 0.01054233, "balance_loss_clip": 1.07315326, "balance_loss_mlp": 1.04191828, "epoch": 0.05952023086635003, "flos": 28256602752000.0, "grad_norm": 1.8280327656497886, "language_loss": 0.78918779, "learning_rate": 3.990901162562491e-06, "loss": 0.81232792, "num_input_tokens_seen": 10673175, "step": 495, "time_per_iteration": 2.6537671089172363 }, { "auxiliary_loss_clip": 0.0122442, "auxiliary_loss_mlp": 0.00767129, "balance_loss_clip": 1.06356549, "balance_loss_mlp": 1.00035369, "epoch": 0.05964047375698912, "flos": 14902500961920.0, "grad_norm": 72.46925237378531, "language_loss": 0.90303814, "learning_rate": 3.9908267917015765e-06, "loss": 0.92295372, "num_input_tokens_seen": 10691235, "step": 496, "time_per_iteration": 2.6486809253692627 }, { "auxiliary_loss_clip": 0.01253598, "auxiliary_loss_mlp": 0.01045041, "balance_loss_clip": 1.07039762, "balance_loss_mlp": 1.03307796, "epoch": 0.059760716647628206, "flos": 23185584972960.0, "grad_norm": 2.0138487307069393, "language_loss": 0.93066454, "learning_rate": 3.990752118833821e-06, "loss": 0.95365095, "num_input_tokens_seen": 10708675, "step": 497, "time_per_iteration": 2.6204047203063965 }, { "auxiliary_loss_clip": 0.01282614, "auxiliary_loss_mlp": 0.01042014, "balance_loss_clip": 1.07594705, "balance_loss_mlp": 1.03073716, "epoch": 0.0598809595382673, "flos": 22746971883360.0, "grad_norm": 1.9542580053063556, "language_loss": 0.78042912, "learning_rate": 3.990677143970553e-06, "loss": 0.80367541, "num_input_tokens_seen": 10729485, "step": 498, "time_per_iteration": 2.628593921661377 }, { "auxiliary_loss_clip": 0.01228755, "auxiliary_loss_mlp": 0.01044208, "balance_loss_clip": 1.07192159, "balance_loss_mlp": 1.03108287, "epoch": 0.06000120242890639, "flos": 22127225589120.0, "grad_norm": 2.472990255783631, "language_loss": 0.80931646, "learning_rate": 3.990601867123144e-06, "loss": 0.83204615, "num_input_tokens_seen": 10749210, "step": 499, "time_per_iteration": 2.696723699569702 }, { "auxiliary_loss_clip": 0.01215243, "auxiliary_loss_mlp": 0.01057545, "balance_loss_clip": 1.06707871, "balance_loss_mlp": 1.04472971, "epoch": 0.06012144531954548, "flos": 19171777235040.0, "grad_norm": 2.1146703508317213, "language_loss": 0.8494575, "learning_rate": 3.990526288303014e-06, "loss": 0.87218541, "num_input_tokens_seen": 10768000, "step": 500, "time_per_iteration": 2.6681313514709473 }, { "auxiliary_loss_clip": 0.01246228, "auxiliary_loss_mlp": 0.00767138, "balance_loss_clip": 1.07206249, "balance_loss_mlp": 1.00034308, "epoch": 0.06024168821018457, "flos": 22783349375520.0, "grad_norm": 1.9983327059454443, "language_loss": 0.90671611, "learning_rate": 3.9904504075216295e-06, "loss": 0.92684972, "num_input_tokens_seen": 10788760, "step": 501, "time_per_iteration": 2.6621286869049072 }, { "auxiliary_loss_clip": 0.01235222, "auxiliary_loss_mlp": 0.01042604, "balance_loss_clip": 1.06639469, "balance_loss_mlp": 1.02972949, "epoch": 0.06036193110082366, "flos": 18770690980320.0, "grad_norm": 2.3462788728396125, "language_loss": 0.93712664, "learning_rate": 3.990374224790501e-06, "loss": 0.95990497, "num_input_tokens_seen": 10806965, "step": 502, "time_per_iteration": 2.6327080726623535 }, { "auxiliary_loss_clip": 0.01250769, "auxiliary_loss_mlp": 0.01048233, "balance_loss_clip": 1.07407498, "balance_loss_mlp": 1.03649116, "epoch": 0.06048217399146275, "flos": 17201574110880.0, "grad_norm": 2.193078168348032, "language_loss": 0.70839876, "learning_rate": 3.990297740121185e-06, "loss": 0.73138881, "num_input_tokens_seen": 10824900, "step": 503, "time_per_iteration": 2.6122827529907227 }, { "auxiliary_loss_clip": 0.01265227, "auxiliary_loss_mlp": 0.00766452, "balance_loss_clip": 1.07316351, "balance_loss_mlp": 1.00038004, "epoch": 0.06060241688210185, "flos": 24024134974560.0, "grad_norm": 1.9810146228006038, "language_loss": 0.78152394, "learning_rate": 3.990220953525284e-06, "loss": 0.80184066, "num_input_tokens_seen": 10842010, "step": 504, "time_per_iteration": 2.6226556301116943 }, { "auxiliary_loss_clip": 0.01232603, "auxiliary_loss_mlp": 0.01040579, "balance_loss_clip": 1.06634295, "balance_loss_mlp": 1.02772236, "epoch": 0.06072265977274094, "flos": 14611193688960.0, "grad_norm": 2.5866602680791884, "language_loss": 0.74336803, "learning_rate": 3.9901438650144465e-06, "loss": 0.76609987, "num_input_tokens_seen": 10858260, "step": 505, "time_per_iteration": 2.608776092529297 }, { "auxiliary_loss_clip": 0.01255267, "auxiliary_loss_mlp": 0.01044897, "balance_loss_clip": 1.07191324, "balance_loss_mlp": 1.03250492, "epoch": 0.06084290266338003, "flos": 20558288304480.0, "grad_norm": 3.6230953482919266, "language_loss": 0.9162513, "learning_rate": 3.990066474600367e-06, "loss": 0.93925297, "num_input_tokens_seen": 10876230, "step": 506, "time_per_iteration": 2.598280668258667 }, { "auxiliary_loss_clip": 0.01247828, "auxiliary_loss_mlp": 0.01049512, "balance_loss_clip": 1.06788707, "balance_loss_mlp": 1.03815091, "epoch": 0.06096314555401912, "flos": 22309220800800.0, "grad_norm": 1.9215708244666698, "language_loss": 0.67852569, "learning_rate": 3.989988782294786e-06, "loss": 0.70149904, "num_input_tokens_seen": 10896320, "step": 507, "time_per_iteration": 2.605743885040283 }, { "auxiliary_loss_clip": 0.01213331, "auxiliary_loss_mlp": 0.01040673, "balance_loss_clip": 1.06611037, "balance_loss_mlp": 1.02869225, "epoch": 0.06108338844465821, "flos": 19131377043360.0, "grad_norm": 1.8112441337361407, "language_loss": 0.94653815, "learning_rate": 3.989910788109489e-06, "loss": 0.96907818, "num_input_tokens_seen": 10912970, "step": 508, "time_per_iteration": 2.646843433380127 }, { "auxiliary_loss_clip": 0.01230828, "auxiliary_loss_mlp": 0.01056958, "balance_loss_clip": 1.07158804, "balance_loss_mlp": 1.0443697, "epoch": 0.0612036313352973, "flos": 33584022907200.0, "grad_norm": 3.0423694226360425, "language_loss": 0.75289893, "learning_rate": 3.989832492056307e-06, "loss": 0.77577686, "num_input_tokens_seen": 10933995, "step": 509, "time_per_iteration": 2.7537338733673096 }, { "auxiliary_loss_clip": 0.012697, "auxiliary_loss_mlp": 0.01055257, "balance_loss_clip": 1.07854033, "balance_loss_mlp": 1.04204869, "epoch": 0.06132387422593639, "flos": 27490556316000.0, "grad_norm": 3.5405966602478154, "language_loss": 0.81050199, "learning_rate": 3.989753894147119e-06, "loss": 0.83375156, "num_input_tokens_seen": 10954120, "step": 510, "time_per_iteration": 2.654505729675293 }, { "auxiliary_loss_clip": 0.01264261, "auxiliary_loss_mlp": 0.01050652, "balance_loss_clip": 1.07965326, "balance_loss_mlp": 1.0390588, "epoch": 0.061444117116575485, "flos": 25885061954400.0, "grad_norm": 1.7721228256143007, "language_loss": 0.79771286, "learning_rate": 3.989674994393846e-06, "loss": 0.82086205, "num_input_tokens_seen": 10973595, "step": 511, "time_per_iteration": 2.62699556350708 }, { "auxiliary_loss_clip": 0.01257799, "auxiliary_loss_mlp": 0.01041521, "balance_loss_clip": 1.07394981, "balance_loss_mlp": 1.02856898, "epoch": 0.061564360007214575, "flos": 28512035370240.0, "grad_norm": 2.1063657968009313, "language_loss": 0.93665415, "learning_rate": 3.98959579280846e-06, "loss": 0.95964736, "num_input_tokens_seen": 10991995, "step": 512, "time_per_iteration": 2.658146619796753 }, { "auxiliary_loss_clip": 0.01194688, "auxiliary_loss_mlp": 0.01045381, "balance_loss_clip": 1.06618392, "balance_loss_mlp": 1.03300118, "epoch": 0.061684602897853665, "flos": 12094358424480.0, "grad_norm": 2.0304019747428237, "language_loss": 0.83093524, "learning_rate": 3.989516289402973e-06, "loss": 0.85333598, "num_input_tokens_seen": 11007625, "step": 513, "time_per_iteration": 3.452840805053711 }, { "auxiliary_loss_clip": 0.01175998, "auxiliary_loss_mlp": 0.0104086, "balance_loss_clip": 1.05699301, "balance_loss_mlp": 1.02830124, "epoch": 0.061804845788492754, "flos": 19532642882880.0, "grad_norm": 2.105874677452928, "language_loss": 0.80690742, "learning_rate": 3.989436484189447e-06, "loss": 0.82907593, "num_input_tokens_seen": 11025570, "step": 514, "time_per_iteration": 2.7382962703704834 }, { "auxiliary_loss_clip": 0.01263165, "auxiliary_loss_mlp": 0.01038856, "balance_loss_clip": 1.06991744, "balance_loss_mlp": 1.02533746, "epoch": 0.061925088679131844, "flos": 15341114051520.0, "grad_norm": 3.750763650869491, "language_loss": 0.80902261, "learning_rate": 3.9893563771799885e-06, "loss": 0.83204281, "num_input_tokens_seen": 11042045, "step": 515, "time_per_iteration": 3.2847843170166016 }, { "auxiliary_loss_clip": 0.01281914, "auxiliary_loss_mlp": 0.01048038, "balance_loss_clip": 1.07800198, "balance_loss_mlp": 1.0353719, "epoch": 0.062045331569770934, "flos": 25919930934240.0, "grad_norm": 3.4854235949896495, "language_loss": 0.86212206, "learning_rate": 3.989275968386749e-06, "loss": 0.88542157, "num_input_tokens_seen": 11059955, "step": 516, "time_per_iteration": 3.4421348571777344 }, { "auxiliary_loss_clip": 0.01234445, "auxiliary_loss_mlp": 0.01038921, "balance_loss_clip": 1.06709766, "balance_loss_mlp": 1.02604651, "epoch": 0.06216557446041003, "flos": 28110625862880.0, "grad_norm": 2.5824567880180402, "language_loss": 0.76916742, "learning_rate": 3.989195257821926e-06, "loss": 0.79190111, "num_input_tokens_seen": 11078440, "step": 517, "time_per_iteration": 3.5222556591033936 }, { "auxiliary_loss_clip": 0.01247649, "auxiliary_loss_mlp": 0.0104384, "balance_loss_clip": 1.07623899, "balance_loss_mlp": 1.03099525, "epoch": 0.06228581735104912, "flos": 23478185256480.0, "grad_norm": 2.126638437848156, "language_loss": 0.84353209, "learning_rate": 3.989114245497765e-06, "loss": 0.86644709, "num_input_tokens_seen": 11098240, "step": 518, "time_per_iteration": 2.6679184436798096 }, { "auxiliary_loss_clip": 0.0126338, "auxiliary_loss_mlp": 0.01043892, "balance_loss_clip": 1.07000327, "balance_loss_mlp": 1.03235877, "epoch": 0.06240606024168821, "flos": 15195208996320.0, "grad_norm": 2.173385269893734, "language_loss": 0.94662273, "learning_rate": 3.989032931426554e-06, "loss": 0.96969545, "num_input_tokens_seen": 11115395, "step": 519, "time_per_iteration": 2.6206774711608887 }, { "auxiliary_loss_clip": 0.01247616, "auxiliary_loss_mlp": 0.01038408, "balance_loss_clip": 1.07192683, "balance_loss_mlp": 1.02584386, "epoch": 0.06252630313232731, "flos": 20631833461920.0, "grad_norm": 2.6227310084511255, "language_loss": 0.86653495, "learning_rate": 3.9889513156206295e-06, "loss": 0.88939524, "num_input_tokens_seen": 11134835, "step": 520, "time_per_iteration": 2.6192123889923096 }, { "auxiliary_loss_clip": 0.01231492, "auxiliary_loss_mlp": 0.01037119, "balance_loss_clip": 1.06914186, "balance_loss_mlp": 1.02476311, "epoch": 0.06264654602296639, "flos": 20778061769760.0, "grad_norm": 2.9962817276145084, "language_loss": 0.73573476, "learning_rate": 3.988869398092371e-06, "loss": 0.75842088, "num_input_tokens_seen": 11154745, "step": 521, "time_per_iteration": 2.6377201080322266 }, { "auxiliary_loss_clip": 0.01247147, "auxiliary_loss_mlp": 0.01042508, "balance_loss_clip": 1.07303882, "balance_loss_mlp": 1.03021145, "epoch": 0.06276678891360549, "flos": 29605802488320.0, "grad_norm": 2.5132390963017848, "language_loss": 0.78766751, "learning_rate": 3.988787178854206e-06, "loss": 0.81056404, "num_input_tokens_seen": 11174280, "step": 522, "time_per_iteration": 2.704514265060425 }, { "auxiliary_loss_clip": 0.01281511, "auxiliary_loss_mlp": 0.01042821, "balance_loss_clip": 1.07751095, "balance_loss_mlp": 1.03075099, "epoch": 0.06288703180424457, "flos": 22126291748160.0, "grad_norm": 2.3635090767889015, "language_loss": 0.87407517, "learning_rate": 3.988704657918608e-06, "loss": 0.89731848, "num_input_tokens_seen": 11193340, "step": 523, "time_per_iteration": 2.5837016105651855 }, { "auxiliary_loss_clip": 0.01266643, "auxiliary_loss_mlp": 0.01048496, "balance_loss_clip": 1.07558036, "balance_loss_mlp": 1.03640783, "epoch": 0.06300727469488367, "flos": 14976692624640.0, "grad_norm": 2.474930269979785, "language_loss": 0.79308242, "learning_rate": 3.988621835298094e-06, "loss": 0.81623381, "num_input_tokens_seen": 11210555, "step": 524, "time_per_iteration": 2.6032567024230957 }, { "auxiliary_loss_clip": 0.01276628, "auxiliary_loss_mlp": 0.01053303, "balance_loss_clip": 1.0763123, "balance_loss_mlp": 1.04142416, "epoch": 0.06312751758552275, "flos": 24535395297600.0, "grad_norm": 2.2184555766053706, "language_loss": 0.91778976, "learning_rate": 3.988538711005229e-06, "loss": 0.94108915, "num_input_tokens_seen": 11230010, "step": 525, "time_per_iteration": 2.6020708084106445 }, { "auxiliary_loss_clip": 0.0125838, "auxiliary_loss_mlp": 0.01041255, "balance_loss_clip": 1.07358742, "balance_loss_mlp": 1.02953076, "epoch": 0.06324776047616185, "flos": 21507012374400.0, "grad_norm": 2.2960108099861745, "language_loss": 0.88111079, "learning_rate": 3.988455285052622e-06, "loss": 0.90410709, "num_input_tokens_seen": 11246190, "step": 526, "time_per_iteration": 2.5942394733428955 }, { "auxiliary_loss_clip": 0.01261453, "auxiliary_loss_mlp": 0.01046569, "balance_loss_clip": 1.07494581, "balance_loss_mlp": 1.03343821, "epoch": 0.06336800336680094, "flos": 21688037828160.0, "grad_norm": 1.879854696345437, "language_loss": 0.83830833, "learning_rate": 3.98837155745293e-06, "loss": 0.86138856, "num_input_tokens_seen": 11264230, "step": 527, "time_per_iteration": 2.6034908294677734 }, { "auxiliary_loss_clip": 0.01263006, "auxiliary_loss_mlp": 0.01039458, "balance_loss_clip": 1.07703173, "balance_loss_mlp": 1.02699482, "epoch": 0.06348824625744003, "flos": 19500898678560.0, "grad_norm": 2.430876609411782, "language_loss": 0.76489341, "learning_rate": 3.988287528218854e-06, "loss": 0.78791803, "num_input_tokens_seen": 11283015, "step": 528, "time_per_iteration": 2.6131937503814697 }, { "auxiliary_loss_clip": 0.01264944, "auxiliary_loss_mlp": 0.01040518, "balance_loss_clip": 1.07910514, "balance_loss_mlp": 1.02881193, "epoch": 0.06360848914807912, "flos": 15481236476160.0, "grad_norm": 1.964619001212542, "language_loss": 0.90356755, "learning_rate": 3.98820319736314e-06, "loss": 0.92662215, "num_input_tokens_seen": 11299630, "step": 529, "time_per_iteration": 2.5832860469818115 }, { "auxiliary_loss_clip": 0.01234235, "auxiliary_loss_mlp": 0.01039389, "balance_loss_clip": 1.06915665, "balance_loss_mlp": 1.02728891, "epoch": 0.0637287320387182, "flos": 20593372786080.0, "grad_norm": 1.8072577415997249, "language_loss": 0.85228992, "learning_rate": 3.988118564898582e-06, "loss": 0.87502623, "num_input_tokens_seen": 11319170, "step": 530, "time_per_iteration": 2.661220073699951 }, { "auxiliary_loss_clip": 0.01217384, "auxiliary_loss_mlp": 0.00767391, "balance_loss_clip": 1.06872797, "balance_loss_mlp": 1.00064874, "epoch": 0.0638489749293573, "flos": 17412224668320.0, "grad_norm": 2.473050018120754, "language_loss": 0.88969135, "learning_rate": 3.988033630838019e-06, "loss": 0.90953904, "num_input_tokens_seen": 11333210, "step": 531, "time_per_iteration": 2.638777017593384 }, { "auxiliary_loss_clip": 0.01269, "auxiliary_loss_mlp": 0.01042282, "balance_loss_clip": 1.07765663, "balance_loss_mlp": 1.03050399, "epoch": 0.0639692178199964, "flos": 23807665869600.0, "grad_norm": 1.6505427097274645, "language_loss": 0.88005787, "learning_rate": 3.987948395194334e-06, "loss": 0.90317076, "num_input_tokens_seen": 11355590, "step": 532, "time_per_iteration": 2.648777961730957 }, { "auxiliary_loss_clip": 0.01256649, "auxiliary_loss_mlp": 0.0104903, "balance_loss_clip": 1.07223678, "balance_loss_mlp": 1.03777647, "epoch": 0.06408946071063548, "flos": 18477228689760.0, "grad_norm": 2.012596818293186, "language_loss": 0.76622981, "learning_rate": 3.987862857980458e-06, "loss": 0.78928661, "num_input_tokens_seen": 11371535, "step": 533, "time_per_iteration": 2.603414297103882 }, { "auxiliary_loss_clip": 0.01236891, "auxiliary_loss_mlp": 0.01042926, "balance_loss_clip": 1.0719471, "balance_loss_mlp": 1.03126168, "epoch": 0.06420970360127458, "flos": 27162225045600.0, "grad_norm": 5.271961621264388, "language_loss": 0.7634871, "learning_rate": 3.987777019209368e-06, "loss": 0.78628528, "num_input_tokens_seen": 11392050, "step": 534, "time_per_iteration": 2.691378355026245 }, { "auxiliary_loss_clip": 0.01277876, "auxiliary_loss_mlp": 0.01044588, "balance_loss_clip": 1.07825184, "balance_loss_mlp": 1.03265524, "epoch": 0.06432994649191366, "flos": 23659677630720.0, "grad_norm": 6.349263572967466, "language_loss": 0.81344473, "learning_rate": 3.987690878894084e-06, "loss": 0.83666945, "num_input_tokens_seen": 11411765, "step": 535, "time_per_iteration": 2.6124727725982666 }, { "auxiliary_loss_clip": 0.0125063, "auxiliary_loss_mlp": 0.01042185, "balance_loss_clip": 1.07333112, "balance_loss_mlp": 1.03028226, "epoch": 0.06445018938255276, "flos": 23403957676800.0, "grad_norm": 15.574333751971201, "language_loss": 0.85192049, "learning_rate": 3.987604437047673e-06, "loss": 0.8748486, "num_input_tokens_seen": 11431565, "step": 536, "time_per_iteration": 2.6085753440856934 }, { "auxiliary_loss_clip": 0.012587, "auxiliary_loss_mlp": 0.01044834, "balance_loss_clip": 1.07485652, "balance_loss_mlp": 1.03343725, "epoch": 0.06457043227319184, "flos": 19646695982880.0, "grad_norm": 2.9958151851615713, "language_loss": 0.77336395, "learning_rate": 3.987517693683251e-06, "loss": 0.79639924, "num_input_tokens_seen": 11450140, "step": 537, "time_per_iteration": 2.632753610610962 }, { "auxiliary_loss_clip": 0.01240269, "auxiliary_loss_mlp": 0.01047805, "balance_loss_clip": 1.07331324, "balance_loss_mlp": 1.03650987, "epoch": 0.06469067516383094, "flos": 16978747704000.0, "grad_norm": 3.921020935362285, "language_loss": 0.96128237, "learning_rate": 3.9874306488139745e-06, "loss": 0.98416317, "num_input_tokens_seen": 11465400, "step": 538, "time_per_iteration": 2.5682549476623535 }, { "auxiliary_loss_clip": 0.0122567, "auxiliary_loss_mlp": 0.01042852, "balance_loss_clip": 1.07111955, "balance_loss_mlp": 1.03149128, "epoch": 0.06481091805447003, "flos": 23296405546560.0, "grad_norm": 2.144902754612172, "language_loss": 0.87871253, "learning_rate": 3.987343302453049e-06, "loss": 0.90139782, "num_input_tokens_seen": 11486675, "step": 539, "time_per_iteration": 3.5163910388946533 }, { "auxiliary_loss_clip": 0.01240548, "auxiliary_loss_mlp": 0.01040126, "balance_loss_clip": 1.07250905, "balance_loss_mlp": 1.02967167, "epoch": 0.06493116094510912, "flos": 29172361440960.0, "grad_norm": 1.648143717541895, "language_loss": 0.82823324, "learning_rate": 3.987255654613724e-06, "loss": 0.85104001, "num_input_tokens_seen": 11510440, "step": 540, "time_per_iteration": 2.691441774368286 }, { "auxiliary_loss_clip": 0.01228994, "auxiliary_loss_mlp": 0.01041852, "balance_loss_clip": 1.07021391, "balance_loss_mlp": 1.03037834, "epoch": 0.06505140383574821, "flos": 19865068686720.0, "grad_norm": 2.2580023819094084, "language_loss": 0.70627177, "learning_rate": 3.987167705309296e-06, "loss": 0.72898018, "num_input_tokens_seen": 11529715, "step": 541, "time_per_iteration": 3.3875763416290283 }, { "auxiliary_loss_clip": 0.01266271, "auxiliary_loss_mlp": 0.00766298, "balance_loss_clip": 1.07671857, "balance_loss_mlp": 1.00052381, "epoch": 0.0651716467263873, "flos": 17924706168000.0, "grad_norm": 3.3076575693408072, "language_loss": 0.95810544, "learning_rate": 3.987079454553108e-06, "loss": 0.97843111, "num_input_tokens_seen": 11547665, "step": 542, "time_per_iteration": 3.461528778076172 }, { "auxiliary_loss_clip": 0.01222467, "auxiliary_loss_mlp": 0.01042195, "balance_loss_clip": 1.07220101, "balance_loss_mlp": 1.0303154, "epoch": 0.0652918896170264, "flos": 20842843188960.0, "grad_norm": 1.797775422031687, "language_loss": 0.91017354, "learning_rate": 3.986990902358546e-06, "loss": 0.9328202, "num_input_tokens_seen": 11564605, "step": 543, "time_per_iteration": 3.5131430625915527 }, { "auxiliary_loss_clip": 0.01259513, "auxiliary_loss_mlp": 0.01038355, "balance_loss_clip": 1.07115316, "balance_loss_mlp": 1.02747095, "epoch": 0.06541213250766549, "flos": 21872511310080.0, "grad_norm": 2.087209615528387, "language_loss": 0.93565738, "learning_rate": 3.986902048739045e-06, "loss": 0.95863605, "num_input_tokens_seen": 11584550, "step": 544, "time_per_iteration": 2.60103178024292 }, { "auxiliary_loss_clip": 0.01245124, "auxiliary_loss_mlp": 0.01050811, "balance_loss_clip": 1.07108116, "balance_loss_mlp": 1.03773367, "epoch": 0.06553237539830457, "flos": 23110746804960.0, "grad_norm": 3.5374138398841826, "language_loss": 0.79722893, "learning_rate": 3.986812893708082e-06, "loss": 0.82018828, "num_input_tokens_seen": 11600740, "step": 545, "time_per_iteration": 2.638688564300537 }, { "auxiliary_loss_clip": 0.0124444, "auxiliary_loss_mlp": 0.01043149, "balance_loss_clip": 1.06961882, "balance_loss_mlp": 1.03026819, "epoch": 0.06565261828894367, "flos": 17923772327040.0, "grad_norm": 1.8486216000579483, "language_loss": 0.81444234, "learning_rate": 3.9867234372791826e-06, "loss": 0.83731818, "num_input_tokens_seen": 11618695, "step": 546, "time_per_iteration": 2.601147174835205 }, { "auxiliary_loss_clip": 0.01258276, "auxiliary_loss_mlp": 0.01046349, "balance_loss_clip": 1.07339191, "balance_loss_mlp": 1.03461862, "epoch": 0.06577286117958275, "flos": 22783062039840.0, "grad_norm": 1.5211763174649864, "language_loss": 0.87432218, "learning_rate": 3.986633679465918e-06, "loss": 0.89736843, "num_input_tokens_seen": 11638850, "step": 547, "time_per_iteration": 2.63434100151062 }, { "auxiliary_loss_clip": 0.01214577, "auxiliary_loss_mlp": 0.01037802, "balance_loss_clip": 1.06776071, "balance_loss_mlp": 1.02688241, "epoch": 0.06589310407022185, "flos": 23696198790720.0, "grad_norm": 2.2256969234029835, "language_loss": 0.80509615, "learning_rate": 3.986543620281904e-06, "loss": 0.82761991, "num_input_tokens_seen": 11658500, "step": 548, "time_per_iteration": 2.6820871829986572 }, { "auxiliary_loss_clip": 0.01222729, "auxiliary_loss_mlp": 0.01047278, "balance_loss_clip": 1.06557536, "balance_loss_mlp": 1.0358516, "epoch": 0.06601334696086093, "flos": 26864775972480.0, "grad_norm": 4.39636057602269, "language_loss": 0.90967548, "learning_rate": 3.986453259740802e-06, "loss": 0.93237555, "num_input_tokens_seen": 11676670, "step": 549, "time_per_iteration": 2.7082741260528564 }, { "auxiliary_loss_clip": 0.0124517, "auxiliary_loss_mlp": 0.01038615, "balance_loss_clip": 1.07370996, "balance_loss_mlp": 1.02705741, "epoch": 0.06613358985150003, "flos": 12567696826080.0, "grad_norm": 2.497198801542605, "language_loss": 0.78939259, "learning_rate": 3.986362597856319e-06, "loss": 0.81223041, "num_input_tokens_seen": 11693170, "step": 550, "time_per_iteration": 2.5872538089752197 }, { "auxiliary_loss_clip": 0.01241153, "auxiliary_loss_mlp": 0.00766659, "balance_loss_clip": 1.07175207, "balance_loss_mlp": 1.00060725, "epoch": 0.06625383274213913, "flos": 18332508894240.0, "grad_norm": 3.2299992611504003, "language_loss": 0.81832755, "learning_rate": 3.986271634642211e-06, "loss": 0.83840567, "num_input_tokens_seen": 11710150, "step": 551, "time_per_iteration": 2.6357100009918213 }, { "auxiliary_loss_clip": 0.01275591, "auxiliary_loss_mlp": 0.01048306, "balance_loss_clip": 1.07653582, "balance_loss_mlp": 1.03624821, "epoch": 0.06637407563277821, "flos": 15375587944800.0, "grad_norm": 1.987600289752203, "language_loss": 0.81754237, "learning_rate": 3.986180370112274e-06, "loss": 0.84078133, "num_input_tokens_seen": 11726670, "step": 552, "time_per_iteration": 2.5234875679016113 }, { "auxiliary_loss_clip": 0.01254353, "auxiliary_loss_mlp": 0.00766433, "balance_loss_clip": 1.07112265, "balance_loss_mlp": 1.00056136, "epoch": 0.0664943185234173, "flos": 24025248400320.0, "grad_norm": 2.287987001016888, "language_loss": 0.7487157, "learning_rate": 3.986088804280354e-06, "loss": 0.76892358, "num_input_tokens_seen": 11746400, "step": 553, "time_per_iteration": 2.650268077850342 }, { "auxiliary_loss_clip": 0.01243552, "auxiliary_loss_mlp": 0.01045532, "balance_loss_clip": 1.07044888, "balance_loss_mlp": 1.03359938, "epoch": 0.06661456141405639, "flos": 20957506877280.0, "grad_norm": 2.0807219066918705, "language_loss": 0.94144571, "learning_rate": 3.985996937160342e-06, "loss": 0.96433651, "num_input_tokens_seen": 11765590, "step": 554, "time_per_iteration": 2.6144120693206787 }, { "auxiliary_loss_clip": 0.01258168, "auxiliary_loss_mlp": 0.01044222, "balance_loss_clip": 1.07637322, "balance_loss_mlp": 1.03329015, "epoch": 0.06673480430469549, "flos": 52223959401120.0, "grad_norm": 1.862373500224051, "language_loss": 0.68834054, "learning_rate": 3.985904768766173e-06, "loss": 0.71136439, "num_input_tokens_seen": 11788365, "step": 555, "time_per_iteration": 2.884136438369751 }, { "auxiliary_loss_clip": 0.01233089, "auxiliary_loss_mlp": 0.01043901, "balance_loss_clip": 1.07056642, "balance_loss_mlp": 1.03145576, "epoch": 0.06685504719533458, "flos": 16217083137120.0, "grad_norm": 4.145663938287344, "language_loss": 0.76200104, "learning_rate": 3.98581229911183e-06, "loss": 0.78477097, "num_input_tokens_seen": 11807285, "step": 556, "time_per_iteration": 2.6376049518585205 }, { "auxiliary_loss_clip": 0.01261685, "auxiliary_loss_mlp": 0.01041954, "balance_loss_clip": 1.07052135, "balance_loss_mlp": 1.0303787, "epoch": 0.06697529008597367, "flos": 22491539265120.0, "grad_norm": 3.3447137058320773, "language_loss": 0.92120886, "learning_rate": 3.985719528211341e-06, "loss": 0.94424522, "num_input_tokens_seen": 11826655, "step": 557, "time_per_iteration": 2.6473801136016846 }, { "auxiliary_loss_clip": 0.01158865, "auxiliary_loss_mlp": 0.01004784, "balance_loss_clip": 1.04340529, "balance_loss_mlp": 1.00125504, "epoch": 0.06709553297661276, "flos": 62688226734720.0, "grad_norm": 0.8460179249225163, "language_loss": 0.63065577, "learning_rate": 3.985626456078777e-06, "loss": 0.65229225, "num_input_tokens_seen": 11891310, "step": 558, "time_per_iteration": 3.312495470046997 }, { "auxiliary_loss_clip": 0.0122867, "auxiliary_loss_mlp": 0.01051554, "balance_loss_clip": 1.07138944, "balance_loss_mlp": 1.03975272, "epoch": 0.06721577586725185, "flos": 11216593490880.0, "grad_norm": 2.1288282420967626, "language_loss": 0.86331809, "learning_rate": 3.985533082728259e-06, "loss": 0.88612038, "num_input_tokens_seen": 11906965, "step": 559, "time_per_iteration": 2.645555019378662 }, { "auxiliary_loss_clip": 0.01273022, "auxiliary_loss_mlp": 0.01047037, "balance_loss_clip": 1.07200539, "balance_loss_mlp": 1.03536034, "epoch": 0.06733601875789094, "flos": 25922193702720.0, "grad_norm": 1.8280216540105279, "language_loss": 0.74935484, "learning_rate": 3.985439408173951e-06, "loss": 0.77255535, "num_input_tokens_seen": 11927190, "step": 560, "time_per_iteration": 2.6196157932281494 }, { "auxiliary_loss_clip": 0.01274186, "auxiliary_loss_mlp": 0.01043745, "balance_loss_clip": 1.07343972, "balance_loss_mlp": 1.03259945, "epoch": 0.06745626164853002, "flos": 20813649088800.0, "grad_norm": 2.1252715931650457, "language_loss": 0.70815402, "learning_rate": 3.9853454324300634e-06, "loss": 0.73133337, "num_input_tokens_seen": 11946400, "step": 561, "time_per_iteration": 2.5672569274902344 }, { "auxiliary_loss_clip": 0.01202481, "auxiliary_loss_mlp": 0.0104505, "balance_loss_clip": 1.06323445, "balance_loss_mlp": 1.03271222, "epoch": 0.06757650453916912, "flos": 19829265865920.0, "grad_norm": 1.8413129934729122, "language_loss": 0.7774139, "learning_rate": 3.985251155510852e-06, "loss": 0.79988921, "num_input_tokens_seen": 11965430, "step": 562, "time_per_iteration": 2.745044469833374 }, { "auxiliary_loss_clip": 0.01210652, "auxiliary_loss_mlp": 0.01043553, "balance_loss_clip": 1.06847405, "balance_loss_mlp": 1.03197193, "epoch": 0.06769674742980822, "flos": 25739228733120.0, "grad_norm": 1.7915209977557978, "language_loss": 0.80092442, "learning_rate": 3.98515657743062e-06, "loss": 0.82346654, "num_input_tokens_seen": 11984895, "step": 563, "time_per_iteration": 2.713737726211548 }, { "auxiliary_loss_clip": 0.01240895, "auxiliary_loss_mlp": 0.01034541, "balance_loss_clip": 1.06820798, "balance_loss_mlp": 1.02285874, "epoch": 0.0678169903204473, "flos": 13074790781760.0, "grad_norm": 2.1929181875773653, "language_loss": 0.77665359, "learning_rate": 3.985061698203711e-06, "loss": 0.79940802, "num_input_tokens_seen": 12002010, "step": 564, "time_per_iteration": 2.6082117557525635 }, { "auxiliary_loss_clip": 0.01172163, "auxiliary_loss_mlp": 0.01004857, "balance_loss_clip": 1.03848886, "balance_loss_mlp": 1.00149488, "epoch": 0.0679372332110864, "flos": 70865841799200.0, "grad_norm": 0.9163568780179705, "language_loss": 0.63849294, "learning_rate": 3.984966517844523e-06, "loss": 0.66026312, "num_input_tokens_seen": 12057255, "step": 565, "time_per_iteration": 3.831334352493286 }, { "auxiliary_loss_clip": 0.01279635, "auxiliary_loss_mlp": 0.0104546, "balance_loss_clip": 1.07697904, "balance_loss_mlp": 1.03225183, "epoch": 0.06805747610172548, "flos": 28256423167200.0, "grad_norm": 2.406723178086417, "language_loss": 0.80822849, "learning_rate": 3.984871036367492e-06, "loss": 0.83147943, "num_input_tokens_seen": 12077280, "step": 566, "time_per_iteration": 3.34836745262146 }, { "auxiliary_loss_clip": 0.01256244, "auxiliary_loss_mlp": 0.00766777, "balance_loss_clip": 1.07474613, "balance_loss_mlp": 1.00052655, "epoch": 0.06817771899236458, "flos": 20120537221920.0, "grad_norm": 2.0063756625647344, "language_loss": 0.82760596, "learning_rate": 3.984775253787102e-06, "loss": 0.8478362, "num_input_tokens_seen": 12095570, "step": 567, "time_per_iteration": 2.6284937858581543 }, { "auxiliary_loss_clip": 0.01261407, "auxiliary_loss_mlp": 0.01046851, "balance_loss_clip": 1.07009208, "balance_loss_mlp": 1.0355444, "epoch": 0.06829796188300366, "flos": 17930632466400.0, "grad_norm": 3.404740081062419, "language_loss": 0.87857902, "learning_rate": 3.984679170117885e-06, "loss": 0.90166163, "num_input_tokens_seen": 12111775, "step": 568, "time_per_iteration": 3.452848196029663 }, { "auxiliary_loss_clip": 0.01255343, "auxiliary_loss_mlp": 0.01037385, "balance_loss_clip": 1.07092023, "balance_loss_mlp": 1.02662647, "epoch": 0.06841820477364276, "flos": 14501630208960.0, "grad_norm": 2.3394401603200565, "language_loss": 0.78431016, "learning_rate": 3.984582785374415e-06, "loss": 0.80723745, "num_input_tokens_seen": 12129215, "step": 569, "time_per_iteration": 3.468801498413086 }, { "auxiliary_loss_clip": 0.01242771, "auxiliary_loss_mlp": 0.00766571, "balance_loss_clip": 1.07200837, "balance_loss_mlp": 1.00047302, "epoch": 0.06853844766428185, "flos": 21938477988960.0, "grad_norm": 2.573822801548471, "language_loss": 0.80527854, "learning_rate": 3.9844860995713155e-06, "loss": 0.82537198, "num_input_tokens_seen": 12148755, "step": 570, "time_per_iteration": 2.6744894981384277 }, { "auxiliary_loss_clip": 0.01257179, "auxiliary_loss_mlp": 0.01045196, "balance_loss_clip": 1.07727468, "balance_loss_mlp": 1.03412724, "epoch": 0.06865869055492094, "flos": 16800631524000.0, "grad_norm": 2.703899829605937, "language_loss": 0.82572979, "learning_rate": 3.9843891127232524e-06, "loss": 0.84875357, "num_input_tokens_seen": 12166290, "step": 571, "time_per_iteration": 2.615769624710083 }, { "auxiliary_loss_clip": 0.01196818, "auxiliary_loss_mlp": 0.01043989, "balance_loss_clip": 1.06415272, "balance_loss_mlp": 1.03256869, "epoch": 0.06877893344556003, "flos": 19937284916640.0, "grad_norm": 2.4955148894429833, "language_loss": 0.67172605, "learning_rate": 3.984291824844938e-06, "loss": 0.69413412, "num_input_tokens_seen": 12181385, "step": 572, "time_per_iteration": 2.682562828063965 }, { "auxiliary_loss_clip": 0.01275027, "auxiliary_loss_mlp": 0.01046853, "balance_loss_clip": 1.07395554, "balance_loss_mlp": 1.03471184, "epoch": 0.06889917633619912, "flos": 23039392582080.0, "grad_norm": 3.130428582573969, "language_loss": 0.84755957, "learning_rate": 3.984194235951132e-06, "loss": 0.87077844, "num_input_tokens_seen": 12197530, "step": 573, "time_per_iteration": 2.5697073936462402 }, { "auxiliary_loss_clip": 0.01276884, "auxiliary_loss_mlp": 0.01050101, "balance_loss_clip": 1.07739496, "balance_loss_mlp": 1.03863311, "epoch": 0.06901941922683821, "flos": 20960559818880.0, "grad_norm": 3.0661261027831754, "language_loss": 0.8482818, "learning_rate": 3.9840963460566375e-06, "loss": 0.87155163, "num_input_tokens_seen": 12216310, "step": 574, "time_per_iteration": 2.569500207901001 }, { "auxiliary_loss_clip": 0.01172413, "auxiliary_loss_mlp": 0.01045182, "balance_loss_clip": 1.06006837, "balance_loss_mlp": 1.03380382, "epoch": 0.06913966211747731, "flos": 24821853780960.0, "grad_norm": 1.575410204636505, "language_loss": 0.89572191, "learning_rate": 3.983998155176305e-06, "loss": 0.91789788, "num_input_tokens_seen": 12236670, "step": 575, "time_per_iteration": 2.847196102142334 }, { "auxiliary_loss_clip": 0.01164516, "auxiliary_loss_mlp": 0.01007326, "balance_loss_clip": 1.0313623, "balance_loss_mlp": 1.00374973, "epoch": 0.06925990500811639, "flos": 58367458317600.0, "grad_norm": 0.8177166117964569, "language_loss": 0.57069856, "learning_rate": 3.9838996633250305e-06, "loss": 0.59241694, "num_input_tokens_seen": 12297185, "step": 576, "time_per_iteration": 3.2252142429351807 }, { "auxiliary_loss_clip": 0.01257912, "auxiliary_loss_mlp": 0.01047727, "balance_loss_clip": 1.07084012, "balance_loss_mlp": 1.03617585, "epoch": 0.06938014789875549, "flos": 12749943456480.0, "grad_norm": 2.0807565526088805, "language_loss": 0.8800863, "learning_rate": 3.983800870517753e-06, "loss": 0.90314269, "num_input_tokens_seen": 12313975, "step": 577, "time_per_iteration": 2.576641321182251 }, { "auxiliary_loss_clip": 0.01252407, "auxiliary_loss_mlp": 0.01037698, "balance_loss_clip": 1.07337427, "balance_loss_mlp": 1.02653408, "epoch": 0.06950039078939457, "flos": 22820229705120.0, "grad_norm": 3.0162618459428034, "language_loss": 0.78265083, "learning_rate": 3.983701776769463e-06, "loss": 0.80555189, "num_input_tokens_seen": 12331385, "step": 578, "time_per_iteration": 2.6130175590515137 }, { "auxiliary_loss_clip": 0.01247824, "auxiliary_loss_mlp": 0.01051563, "balance_loss_clip": 1.07340205, "balance_loss_mlp": 1.03968358, "epoch": 0.06962063368003367, "flos": 21941351345760.0, "grad_norm": 2.2384284699394104, "language_loss": 0.85614419, "learning_rate": 3.9836023820951885e-06, "loss": 0.87913805, "num_input_tokens_seen": 12350600, "step": 579, "time_per_iteration": 2.6327028274536133 }, { "auxiliary_loss_clip": 0.01220052, "auxiliary_loss_mlp": 0.01041621, "balance_loss_clip": 1.06610656, "balance_loss_mlp": 1.03004587, "epoch": 0.06974087657067275, "flos": 20706025124640.0, "grad_norm": 2.243229348299106, "language_loss": 0.68708652, "learning_rate": 3.983502686510011e-06, "loss": 0.70970327, "num_input_tokens_seen": 12371430, "step": 580, "time_per_iteration": 2.641106128692627 }, { "auxiliary_loss_clip": 0.01259792, "auxiliary_loss_mlp": 0.00766391, "balance_loss_clip": 1.07124734, "balance_loss_mlp": 1.00042534, "epoch": 0.06986111946131185, "flos": 22638234493440.0, "grad_norm": 2.0735265145152675, "language_loss": 0.73587453, "learning_rate": 3.9834026900290525e-06, "loss": 0.7561363, "num_input_tokens_seen": 12390825, "step": 581, "time_per_iteration": 2.619147300720215 }, { "auxiliary_loss_clip": 0.0127379, "auxiliary_loss_mlp": 0.01042245, "balance_loss_clip": 1.07538986, "balance_loss_mlp": 1.03139091, "epoch": 0.06998136235195095, "flos": 26943457255200.0, "grad_norm": 1.9951904847931166, "language_loss": 1.00210655, "learning_rate": 3.983302392667482e-06, "loss": 1.02526689, "num_input_tokens_seen": 12411670, "step": 582, "time_per_iteration": 2.5945401191711426 }, { "auxiliary_loss_clip": 0.01255858, "auxiliary_loss_mlp": 0.01043028, "balance_loss_clip": 1.075086, "balance_loss_mlp": 1.03199506, "epoch": 0.07010160524259003, "flos": 22492509023040.0, "grad_norm": 1.798462671227382, "language_loss": 0.93607128, "learning_rate": 3.983201794440517e-06, "loss": 0.95906013, "num_input_tokens_seen": 12431245, "step": 583, "time_per_iteration": 2.6009421348571777 }, { "auxiliary_loss_clip": 0.01227924, "auxiliary_loss_mlp": 0.01049715, "balance_loss_clip": 1.06844616, "balance_loss_mlp": 1.03812802, "epoch": 0.07022184813322913, "flos": 18332544811200.0, "grad_norm": 1.9222169083865988, "language_loss": 0.67725009, "learning_rate": 3.9831008953634165e-06, "loss": 0.70002651, "num_input_tokens_seen": 12450535, "step": 584, "time_per_iteration": 2.598215341567993 }, { "auxiliary_loss_clip": 0.01185112, "auxiliary_loss_mlp": 0.01045844, "balance_loss_clip": 1.06013572, "balance_loss_mlp": 1.03414333, "epoch": 0.07034209102386821, "flos": 24675553639200.0, "grad_norm": 2.079991298750965, "language_loss": 0.81267107, "learning_rate": 3.9829996954514864e-06, "loss": 0.83498067, "num_input_tokens_seen": 12469675, "step": 585, "time_per_iteration": 2.7205541133880615 }, { "auxiliary_loss_clip": 0.01244212, "auxiliary_loss_mlp": 0.01042344, "balance_loss_clip": 1.07116222, "balance_loss_mlp": 1.0303334, "epoch": 0.0704623339145073, "flos": 25995882528000.0, "grad_norm": 2.3732419459961607, "language_loss": 0.8421675, "learning_rate": 3.982898194720079e-06, "loss": 0.86503303, "num_input_tokens_seen": 12490405, "step": 586, "time_per_iteration": 2.6235885620117188 }, { "auxiliary_loss_clip": 0.01229293, "auxiliary_loss_mlp": 0.00766931, "balance_loss_clip": 1.06947291, "balance_loss_mlp": 1.00031996, "epoch": 0.0705825768051464, "flos": 25338322063200.0, "grad_norm": 2.1286402587278515, "language_loss": 0.82731128, "learning_rate": 3.982796393184592e-06, "loss": 0.84727359, "num_input_tokens_seen": 12509485, "step": 587, "time_per_iteration": 2.676130771636963 }, { "auxiliary_loss_clip": 0.01148736, "auxiliary_loss_mlp": 0.01004064, "balance_loss_clip": 1.03008711, "balance_loss_mlp": 1.00044012, "epoch": 0.07070281969578548, "flos": 66047562866400.0, "grad_norm": 0.7957339933329464, "language_loss": 0.62714362, "learning_rate": 3.98269429086047e-06, "loss": 0.64867163, "num_input_tokens_seen": 12567325, "step": 588, "time_per_iteration": 3.047417640686035 }, { "auxiliary_loss_clip": 0.01224362, "auxiliary_loss_mlp": 0.01039374, "balance_loss_clip": 1.06522918, "balance_loss_mlp": 1.02643991, "epoch": 0.07082306258642458, "flos": 23653571747520.0, "grad_norm": 2.556151636541883, "language_loss": 0.86371213, "learning_rate": 3.982591887763199e-06, "loss": 0.88634944, "num_input_tokens_seen": 12584785, "step": 589, "time_per_iteration": 2.6377665996551514 }, { "auxiliary_loss_clip": 0.01194078, "auxiliary_loss_mlp": 0.01037569, "balance_loss_clip": 1.05766964, "balance_loss_mlp": 1.02536166, "epoch": 0.07094330547706366, "flos": 13880052149760.0, "grad_norm": 2.1904997842138014, "language_loss": 0.81163275, "learning_rate": 3.982489183908316e-06, "loss": 0.83394921, "num_input_tokens_seen": 12601205, "step": 590, "time_per_iteration": 2.6214044094085693 }, { "auxiliary_loss_clip": 0.01160383, "auxiliary_loss_mlp": 0.01043583, "balance_loss_clip": 1.05040979, "balance_loss_mlp": 1.03268766, "epoch": 0.07106354836770276, "flos": 24645102445440.0, "grad_norm": 1.6910298640939474, "language_loss": 0.84447891, "learning_rate": 3.982386179311399e-06, "loss": 0.86651856, "num_input_tokens_seen": 12621725, "step": 591, "time_per_iteration": 3.519251823425293 }, { "auxiliary_loss_clip": 0.01263921, "auxiliary_loss_mlp": 0.01050454, "balance_loss_clip": 1.07531285, "balance_loss_mlp": 1.03779948, "epoch": 0.07118379125834184, "flos": 16217226804960.0, "grad_norm": 2.1729963250597097, "language_loss": 0.8760376, "learning_rate": 3.982282873988075e-06, "loss": 0.89918137, "num_input_tokens_seen": 12639600, "step": 592, "time_per_iteration": 3.3523497581481934 }, { "auxiliary_loss_clip": 0.01237484, "auxiliary_loss_mlp": 0.01043021, "balance_loss_clip": 1.07078445, "balance_loss_mlp": 1.03239346, "epoch": 0.07130403414898094, "flos": 19719989721600.0, "grad_norm": 1.6484050107751673, "language_loss": 0.87079084, "learning_rate": 3.982179267954016e-06, "loss": 0.89359587, "num_input_tokens_seen": 12660030, "step": 593, "time_per_iteration": 2.6404666900634766 }, { "auxiliary_loss_clip": 0.01268699, "auxiliary_loss_mlp": 0.01038767, "balance_loss_clip": 1.07183623, "balance_loss_mlp": 1.02738297, "epoch": 0.07142427703962004, "flos": 21871936638720.0, "grad_norm": 2.2412394309783505, "language_loss": 0.96082783, "learning_rate": 3.982075361224937e-06, "loss": 0.98390245, "num_input_tokens_seen": 12678395, "step": 594, "time_per_iteration": 3.465794324874878 }, { "auxiliary_loss_clip": 0.01254969, "auxiliary_loss_mlp": 0.00766975, "balance_loss_clip": 1.07309103, "balance_loss_mlp": 1.00048304, "epoch": 0.07154451993025912, "flos": 18296598322560.0, "grad_norm": 1.8152828018358496, "language_loss": 0.87938648, "learning_rate": 3.981971153816602e-06, "loss": 0.89960599, "num_input_tokens_seen": 12696000, "step": 595, "time_per_iteration": 3.5089657306671143 }, { "auxiliary_loss_clip": 0.0127201, "auxiliary_loss_mlp": 0.01038567, "balance_loss_clip": 1.07636726, "balance_loss_mlp": 1.02754033, "epoch": 0.07166476282089822, "flos": 22160693807520.0, "grad_norm": 1.9142068659968758, "language_loss": 0.96133554, "learning_rate": 3.981866645744819e-06, "loss": 0.98444134, "num_input_tokens_seen": 12716715, "step": 596, "time_per_iteration": 2.5763323307037354 }, { "auxiliary_loss_clip": 0.01269697, "auxiliary_loss_mlp": 0.00766837, "balance_loss_clip": 1.0713613, "balance_loss_mlp": 1.00046659, "epoch": 0.0717850057115373, "flos": 14136346775040.0, "grad_norm": 2.291078619581766, "language_loss": 0.81322044, "learning_rate": 3.9817618370254416e-06, "loss": 0.8335858, "num_input_tokens_seen": 12733370, "step": 597, "time_per_iteration": 2.536823272705078 }, { "auxiliary_loss_clip": 0.01272564, "auxiliary_loss_mlp": 0.0104171, "balance_loss_clip": 1.07405734, "balance_loss_mlp": 1.03121924, "epoch": 0.0719052486021764, "flos": 30917798642400.0, "grad_norm": 2.0501907099703414, "language_loss": 0.87242806, "learning_rate": 3.9816567276743684e-06, "loss": 0.89557087, "num_input_tokens_seen": 12753235, "step": 598, "time_per_iteration": 2.618614912033081 }, { "auxiliary_loss_clip": 0.01231834, "auxiliary_loss_mlp": 0.01050139, "balance_loss_clip": 1.06754768, "balance_loss_mlp": 1.03841496, "epoch": 0.0720254914928155, "flos": 21287023407360.0, "grad_norm": 1.8623268933096775, "language_loss": 0.77156997, "learning_rate": 3.9815513177075466e-06, "loss": 0.79438972, "num_input_tokens_seen": 12772020, "step": 599, "time_per_iteration": 2.6576366424560547 }, { "auxiliary_loss_clip": 0.0124727, "auxiliary_loss_mlp": 0.01047093, "balance_loss_clip": 1.07085121, "balance_loss_mlp": 1.03600037, "epoch": 0.07214573438345458, "flos": 27819174922080.0, "grad_norm": 1.5665975058722403, "language_loss": 0.70276237, "learning_rate": 3.9814456071409646e-06, "loss": 0.72570604, "num_input_tokens_seen": 12792555, "step": 600, "time_per_iteration": 2.630927801132202 }, { "auxiliary_loss_clip": 0.01213492, "auxiliary_loss_mlp": 0.01043704, "balance_loss_clip": 1.06508613, "balance_loss_mlp": 1.03127611, "epoch": 0.07226597727409367, "flos": 25483580613120.0, "grad_norm": 2.510900448069448, "language_loss": 0.85465777, "learning_rate": 3.981339595990659e-06, "loss": 0.87722975, "num_input_tokens_seen": 12811085, "step": 601, "time_per_iteration": 2.727975368499756 }, { "auxiliary_loss_clip": 0.01253947, "auxiliary_loss_mlp": 0.01041157, "balance_loss_clip": 1.07208776, "balance_loss_mlp": 1.02830672, "epoch": 0.07238622016473276, "flos": 23513844409440.0, "grad_norm": 1.8488786092462823, "language_loss": 0.81260854, "learning_rate": 3.981233284272713e-06, "loss": 0.83555955, "num_input_tokens_seen": 12830830, "step": 602, "time_per_iteration": 2.5908780097961426 }, { "auxiliary_loss_clip": 0.01222508, "auxiliary_loss_mlp": 0.01055267, "balance_loss_clip": 1.06668091, "balance_loss_mlp": 1.04467499, "epoch": 0.07250646305537185, "flos": 25453524505920.0, "grad_norm": 1.651963304925799, "language_loss": 0.900397, "learning_rate": 3.981126672003253e-06, "loss": 0.92317474, "num_input_tokens_seen": 12853505, "step": 603, "time_per_iteration": 2.7312610149383545 }, { "auxiliary_loss_clip": 0.01240206, "auxiliary_loss_mlp": 0.01046277, "balance_loss_clip": 1.06408715, "balance_loss_mlp": 1.03465974, "epoch": 0.07262670594601094, "flos": 27155041653600.0, "grad_norm": 2.3621869449672555, "language_loss": 0.77782631, "learning_rate": 3.981019759198451e-06, "loss": 0.80069113, "num_input_tokens_seen": 12872455, "step": 604, "time_per_iteration": 2.6329872608184814 }, { "auxiliary_loss_clip": 0.01240826, "auxiliary_loss_mlp": 0.01039561, "balance_loss_clip": 1.067626, "balance_loss_mlp": 1.027843, "epoch": 0.07274694883665003, "flos": 26651611227840.0, "grad_norm": 2.758424912059481, "language_loss": 0.84570128, "learning_rate": 3.980912545874528e-06, "loss": 0.86850518, "num_input_tokens_seen": 12892620, "step": 605, "time_per_iteration": 2.644073009490967 }, { "auxiliary_loss_clip": 0.01245805, "auxiliary_loss_mlp": 0.00766925, "balance_loss_clip": 1.07038188, "balance_loss_mlp": 1.00049615, "epoch": 0.07286719172728913, "flos": 29862348532320.0, "grad_norm": 2.108441452305805, "language_loss": 0.85602117, "learning_rate": 3.980805032047746e-06, "loss": 0.8761484, "num_input_tokens_seen": 12914090, "step": 606, "time_per_iteration": 2.64504075050354 }, { "auxiliary_loss_clip": 0.01237478, "auxiliary_loss_mlp": 0.01045865, "balance_loss_clip": 1.06866026, "balance_loss_mlp": 1.03473127, "epoch": 0.07298743461792821, "flos": 17382060810240.0, "grad_norm": 3.602483253855261, "language_loss": 0.81239825, "learning_rate": 3.980697217734415e-06, "loss": 0.83523172, "num_input_tokens_seen": 12931830, "step": 607, "time_per_iteration": 2.604418992996216 }, { "auxiliary_loss_clip": 0.01210545, "auxiliary_loss_mlp": 0.00766453, "balance_loss_clip": 1.06448865, "balance_loss_mlp": 1.0005672, "epoch": 0.07310767750856731, "flos": 19498204906560.0, "grad_norm": 3.303099963199472, "language_loss": 0.91643333, "learning_rate": 3.980589102950891e-06, "loss": 0.93620336, "num_input_tokens_seen": 12949995, "step": 608, "time_per_iteration": 2.7069473266601562 }, { "auxiliary_loss_clip": 0.01238968, "auxiliary_loss_mlp": 0.01042139, "balance_loss_clip": 1.07269478, "balance_loss_mlp": 1.03010488, "epoch": 0.07322792039920639, "flos": 29168697911040.0, "grad_norm": 2.4621864940052585, "language_loss": 0.75577939, "learning_rate": 3.9804806877135755e-06, "loss": 0.7785905, "num_input_tokens_seen": 12968040, "step": 609, "time_per_iteration": 2.699467658996582 }, { "auxiliary_loss_clip": 0.01256493, "auxiliary_loss_mlp": 0.00766995, "balance_loss_clip": 1.06893408, "balance_loss_mlp": 1.00046229, "epoch": 0.07334816328984549, "flos": 23477826086880.0, "grad_norm": 2.278845818402841, "language_loss": 0.86305505, "learning_rate": 3.980371972038915e-06, "loss": 0.88328993, "num_input_tokens_seen": 12988530, "step": 610, "time_per_iteration": 2.6316542625427246 }, { "auxiliary_loss_clip": 0.01274244, "auxiliary_loss_mlp": 0.01038416, "balance_loss_clip": 1.07771897, "balance_loss_mlp": 1.02662587, "epoch": 0.07346840618048459, "flos": 22962471230400.0, "grad_norm": 1.7238269567632716, "language_loss": 0.84439743, "learning_rate": 3.980262955943399e-06, "loss": 0.86752403, "num_input_tokens_seen": 13008195, "step": 611, "time_per_iteration": 2.565328598022461 }, { "auxiliary_loss_clip": 0.01232124, "auxiliary_loss_mlp": 0.01042937, "balance_loss_clip": 1.07157099, "balance_loss_mlp": 1.03157616, "epoch": 0.07358864907112367, "flos": 17673906837600.0, "grad_norm": 3.462154944704319, "language_loss": 0.87017971, "learning_rate": 3.980153639443569e-06, "loss": 0.89293027, "num_input_tokens_seen": 13024180, "step": 612, "time_per_iteration": 2.6049892902374268 }, { "auxiliary_loss_clip": 0.01248904, "auxiliary_loss_mlp": 0.01038767, "balance_loss_clip": 1.07277608, "balance_loss_mlp": 1.02676845, "epoch": 0.07370889196176277, "flos": 24097033626720.0, "grad_norm": 2.0926807279025614, "language_loss": 0.80119753, "learning_rate": 3.980044022556005e-06, "loss": 0.82407427, "num_input_tokens_seen": 13043865, "step": 613, "time_per_iteration": 2.6534032821655273 }, { "auxiliary_loss_clip": 0.01252886, "auxiliary_loss_mlp": 0.01045887, "balance_loss_clip": 1.07237458, "balance_loss_mlp": 1.03423476, "epoch": 0.07382913485240185, "flos": 25885923961440.0, "grad_norm": 4.0249802598581, "language_loss": 0.7306332, "learning_rate": 3.9799341052973375e-06, "loss": 0.75362092, "num_input_tokens_seen": 13063700, "step": 614, "time_per_iteration": 2.6757562160491943 }, { "auxiliary_loss_clip": 0.01237769, "auxiliary_loss_mlp": 0.01043948, "balance_loss_clip": 1.07174897, "balance_loss_mlp": 1.03143704, "epoch": 0.07394937774304094, "flos": 16873853428800.0, "grad_norm": 2.4753577998257237, "language_loss": 0.75207967, "learning_rate": 3.979823887684241e-06, "loss": 0.77489686, "num_input_tokens_seen": 13082640, "step": 615, "time_per_iteration": 2.612959384918213 }, { "auxiliary_loss_clip": 0.01271804, "auxiliary_loss_mlp": 0.01050857, "balance_loss_clip": 1.0749706, "balance_loss_mlp": 1.03889441, "epoch": 0.07406962063368003, "flos": 20703475020480.0, "grad_norm": 2.6512767850798498, "language_loss": 0.84788322, "learning_rate": 3.979713369733434e-06, "loss": 0.87110984, "num_input_tokens_seen": 13100505, "step": 616, "time_per_iteration": 2.5922670364379883 }, { "auxiliary_loss_clip": 0.01248026, "auxiliary_loss_mlp": 0.01045244, "balance_loss_clip": 1.07124102, "balance_loss_mlp": 1.03351378, "epoch": 0.07418986352431912, "flos": 21430988946720.0, "grad_norm": 2.115925918596952, "language_loss": 0.84742725, "learning_rate": 3.979602551461683e-06, "loss": 0.87035996, "num_input_tokens_seen": 13121285, "step": 617, "time_per_iteration": 3.38090181350708 }, { "auxiliary_loss_clip": 0.0123519, "auxiliary_loss_mlp": 0.01042628, "balance_loss_clip": 1.07060289, "balance_loss_mlp": 1.03157091, "epoch": 0.07431010641495822, "flos": 12021136519680.0, "grad_norm": 3.219838664167317, "language_loss": 0.91911429, "learning_rate": 3.979491432885799e-06, "loss": 0.94189239, "num_input_tokens_seen": 13137550, "step": 618, "time_per_iteration": 3.320349931716919 }, { "auxiliary_loss_clip": 0.01202616, "auxiliary_loss_mlp": 0.00767048, "balance_loss_clip": 1.06308138, "balance_loss_mlp": 1.0004518, "epoch": 0.0744303493055973, "flos": 20957578711200.0, "grad_norm": 2.0509294396983413, "language_loss": 0.82951045, "learning_rate": 3.97938001402264e-06, "loss": 0.84920704, "num_input_tokens_seen": 13156675, "step": 619, "time_per_iteration": 2.679399013519287 }, { "auxiliary_loss_clip": 0.0121468, "auxiliary_loss_mlp": 0.0104485, "balance_loss_clip": 1.06729698, "balance_loss_mlp": 1.03356075, "epoch": 0.0745505921962364, "flos": 16253137376640.0, "grad_norm": 2.494707506373387, "language_loss": 0.79692435, "learning_rate": 3.979268294889105e-06, "loss": 0.81951964, "num_input_tokens_seen": 13172225, "step": 620, "time_per_iteration": 3.502690315246582 }, { "auxiliary_loss_clip": 0.01271305, "auxiliary_loss_mlp": 0.01057081, "balance_loss_clip": 1.07516992, "balance_loss_mlp": 1.04579794, "epoch": 0.07467083508687548, "flos": 50944641292320.0, "grad_norm": 2.4360570949141382, "language_loss": 0.73909807, "learning_rate": 3.979156275502143e-06, "loss": 0.76238191, "num_input_tokens_seen": 13195885, "step": 621, "time_per_iteration": 3.6679556369781494 }, { "auxiliary_loss_clip": 0.01223063, "auxiliary_loss_mlp": 0.01043175, "balance_loss_clip": 1.06734133, "balance_loss_mlp": 1.03125429, "epoch": 0.07479107797751458, "flos": 17529689879520.0, "grad_norm": 2.3113517028109385, "language_loss": 0.91799867, "learning_rate": 3.979043955878749e-06, "loss": 0.94066107, "num_input_tokens_seen": 13213730, "step": 622, "time_per_iteration": 2.6765923500061035 }, { "auxiliary_loss_clip": 0.01238273, "auxiliary_loss_mlp": 0.01050056, "balance_loss_clip": 1.07274997, "balance_loss_mlp": 1.03812957, "epoch": 0.07491132086815366, "flos": 23473947055200.0, "grad_norm": 1.9139703569692343, "language_loss": 0.83486986, "learning_rate": 3.978931336035959e-06, "loss": 0.8577531, "num_input_tokens_seen": 13232540, "step": 623, "time_per_iteration": 2.6598615646362305 }, { "auxiliary_loss_clip": 0.01258006, "auxiliary_loss_mlp": 0.01043941, "balance_loss_clip": 1.0750668, "balance_loss_mlp": 1.03275895, "epoch": 0.07503156375879276, "flos": 20157561219360.0, "grad_norm": 2.6241087086894845, "language_loss": 0.82315379, "learning_rate": 3.9788184159908595e-06, "loss": 0.84617329, "num_input_tokens_seen": 13249670, "step": 624, "time_per_iteration": 2.6051828861236572 }, { "auxiliary_loss_clip": 0.0123283, "auxiliary_loss_mlp": 0.01046334, "balance_loss_clip": 1.06921911, "balance_loss_mlp": 1.0356648, "epoch": 0.07515180664943186, "flos": 15115522038720.0, "grad_norm": 3.603378084550165, "language_loss": 0.82702327, "learning_rate": 3.97870519576058e-06, "loss": 0.84981489, "num_input_tokens_seen": 13266095, "step": 625, "time_per_iteration": 2.626652479171753 }, { "auxiliary_loss_clip": 0.01223919, "auxiliary_loss_mlp": 0.00767485, "balance_loss_clip": 1.06950819, "balance_loss_mlp": 1.00034857, "epoch": 0.07527204954007094, "flos": 21287705829600.0, "grad_norm": 2.3982191809155498, "language_loss": 0.80664772, "learning_rate": 3.978591675362295e-06, "loss": 0.82656175, "num_input_tokens_seen": 13284810, "step": 626, "time_per_iteration": 2.7207834720611572 }, { "auxiliary_loss_clip": 0.01205951, "auxiliary_loss_mlp": 0.01044576, "balance_loss_clip": 1.07103801, "balance_loss_mlp": 1.03359699, "epoch": 0.07539229243071004, "flos": 21324191072640.0, "grad_norm": 1.5943383898371077, "language_loss": 0.87481678, "learning_rate": 3.978477854813226e-06, "loss": 0.89732206, "num_input_tokens_seen": 13304150, "step": 627, "time_per_iteration": 2.693767786026001 }, { "auxiliary_loss_clip": 0.01254178, "auxiliary_loss_mlp": 0.01036032, "balance_loss_clip": 1.07154024, "balance_loss_mlp": 1.02473068, "epoch": 0.07551253532134912, "flos": 13042543740000.0, "grad_norm": 2.1934196939628543, "language_loss": 0.82180858, "learning_rate": 3.97836373413064e-06, "loss": 0.84471071, "num_input_tokens_seen": 13322205, "step": 628, "time_per_iteration": 2.6490347385406494 }, { "auxiliary_loss_clip": 0.01272221, "auxiliary_loss_mlp": 0.01046119, "balance_loss_clip": 1.07415462, "balance_loss_mlp": 1.03475785, "epoch": 0.07563277821198822, "flos": 19208765315520.0, "grad_norm": 5.072332733472573, "language_loss": 0.74982893, "learning_rate": 3.978249313331848e-06, "loss": 0.77301228, "num_input_tokens_seen": 13340435, "step": 629, "time_per_iteration": 2.5552055835723877 }, { "auxiliary_loss_clip": 0.01256618, "auxiliary_loss_mlp": 0.00766784, "balance_loss_clip": 1.06818998, "balance_loss_mlp": 1.00044024, "epoch": 0.07575302110262731, "flos": 19537204336800.0, "grad_norm": 2.860310976897451, "language_loss": 0.6226306, "learning_rate": 3.978134592434208e-06, "loss": 0.64286458, "num_input_tokens_seen": 13358185, "step": 630, "time_per_iteration": 2.6194236278533936 }, { "auxiliary_loss_clip": 0.01108666, "auxiliary_loss_mlp": 0.01007221, "balance_loss_clip": 1.03591824, "balance_loss_mlp": 1.00412202, "epoch": 0.0758732639932664, "flos": 67961820143520.0, "grad_norm": 1.0318628783985329, "language_loss": 0.59416747, "learning_rate": 3.978019571455123e-06, "loss": 0.6153264, "num_input_tokens_seen": 13410130, "step": 631, "time_per_iteration": 3.282238483428955 }, { "auxiliary_loss_clip": 0.01268701, "auxiliary_loss_mlp": 0.01053854, "balance_loss_clip": 1.07510173, "balance_loss_mlp": 1.04283357, "epoch": 0.07599350688390549, "flos": 18989207352000.0, "grad_norm": 2.3761069503953243, "language_loss": 0.83893037, "learning_rate": 3.977904250412042e-06, "loss": 0.86215591, "num_input_tokens_seen": 13429085, "step": 632, "time_per_iteration": 2.601353168487549 }, { "auxiliary_loss_clip": 0.01249875, "auxiliary_loss_mlp": 0.01047949, "balance_loss_clip": 1.07519698, "balance_loss_mlp": 1.03556359, "epoch": 0.07611374977454458, "flos": 21069009873120.0, "grad_norm": 2.227795276760294, "language_loss": 0.85926682, "learning_rate": 3.97778862932246e-06, "loss": 0.88224512, "num_input_tokens_seen": 13446250, "step": 633, "time_per_iteration": 2.6546571254730225 }, { "auxiliary_loss_clip": 0.01134019, "auxiliary_loss_mlp": 0.01054083, "balance_loss_clip": 1.04667711, "balance_loss_mlp": 1.04204905, "epoch": 0.07623399266518367, "flos": 18514540022880.0, "grad_norm": 2.172310038762509, "language_loss": 0.93797445, "learning_rate": 3.9776727082039144e-06, "loss": 0.95985556, "num_input_tokens_seen": 13463220, "step": 634, "time_per_iteration": 2.9450581073760986 }, { "auxiliary_loss_clip": 0.01173474, "auxiliary_loss_mlp": 0.01006855, "balance_loss_clip": 1.04377031, "balance_loss_mlp": 1.00381517, "epoch": 0.07635423555582276, "flos": 44663044077600.0, "grad_norm": 0.814540331349219, "language_loss": 0.554672, "learning_rate": 3.977556487073991e-06, "loss": 0.57647526, "num_input_tokens_seen": 13517775, "step": 635, "time_per_iteration": 3.280874013900757 }, { "auxiliary_loss_clip": 0.01228597, "auxiliary_loss_mlp": 0.01033883, "balance_loss_clip": 1.06643963, "balance_loss_mlp": 1.02345812, "epoch": 0.07647447844646185, "flos": 21761151982080.0, "grad_norm": 1.8141386558455068, "language_loss": 0.81486285, "learning_rate": 3.97743996595032e-06, "loss": 0.83748758, "num_input_tokens_seen": 13537815, "step": 636, "time_per_iteration": 2.638566255569458 }, { "auxiliary_loss_clip": 0.01272076, "auxiliary_loss_mlp": 0.01043812, "balance_loss_clip": 1.07502377, "balance_loss_mlp": 1.03212929, "epoch": 0.07659472133710095, "flos": 23806803862560.0, "grad_norm": 1.5154901050789094, "language_loss": 0.81537092, "learning_rate": 3.9773231448505804e-06, "loss": 0.83852977, "num_input_tokens_seen": 13559605, "step": 637, "time_per_iteration": 2.5924265384674072 }, { "auxiliary_loss_clip": 0.01237133, "auxiliary_loss_mlp": 0.00766941, "balance_loss_clip": 1.07298326, "balance_loss_mlp": 1.00042295, "epoch": 0.07671496422774003, "flos": 21469988376960.0, "grad_norm": 2.501979364912409, "language_loss": 0.78215724, "learning_rate": 3.977206023792491e-06, "loss": 0.80219799, "num_input_tokens_seen": 13579495, "step": 638, "time_per_iteration": 2.6257896423339844 }, { "auxiliary_loss_clip": 0.01251832, "auxiliary_loss_mlp": 0.01050383, "balance_loss_clip": 1.07402182, "balance_loss_mlp": 1.03958273, "epoch": 0.07683520711837913, "flos": 16980974555520.0, "grad_norm": 1.9159934587858032, "language_loss": 0.81001478, "learning_rate": 3.97708860279382e-06, "loss": 0.83303696, "num_input_tokens_seen": 13597605, "step": 639, "time_per_iteration": 2.581230878829956 }, { "auxiliary_loss_clip": 0.01214497, "auxiliary_loss_mlp": 0.01048951, "balance_loss_clip": 1.06622708, "balance_loss_mlp": 1.03725064, "epoch": 0.07695545000901821, "flos": 23476748578080.0, "grad_norm": 2.1301423634694543, "language_loss": 0.78260356, "learning_rate": 3.97697088187238e-06, "loss": 0.80523801, "num_input_tokens_seen": 13618120, "step": 640, "time_per_iteration": 2.6749932765960693 }, { "auxiliary_loss_clip": 0.01234531, "auxiliary_loss_mlp": 0.01043523, "balance_loss_clip": 1.07257175, "balance_loss_mlp": 1.03336668, "epoch": 0.07707569289965731, "flos": 17634260902080.0, "grad_norm": 3.2633278004786197, "language_loss": 0.92149574, "learning_rate": 3.976852861046029e-06, "loss": 0.94427627, "num_input_tokens_seen": 13634735, "step": 641, "time_per_iteration": 2.6025497913360596 }, { "auxiliary_loss_clip": 0.01202765, "auxiliary_loss_mlp": 0.01034621, "balance_loss_clip": 1.06804276, "balance_loss_mlp": 1.02427983, "epoch": 0.0771959357902964, "flos": 25775678059200.0, "grad_norm": 1.6355807703733432, "language_loss": 0.80024165, "learning_rate": 3.97673454033267e-06, "loss": 0.8226155, "num_input_tokens_seen": 13656835, "step": 642, "time_per_iteration": 2.755314588546753 }, { "auxiliary_loss_clip": 0.0123927, "auxiliary_loss_mlp": 0.01044837, "balance_loss_clip": 1.06923258, "balance_loss_mlp": 1.03409588, "epoch": 0.07731617868093549, "flos": 19828655277600.0, "grad_norm": 1.9529281931044158, "language_loss": 0.82681727, "learning_rate": 3.976615919750254e-06, "loss": 0.84965837, "num_input_tokens_seen": 13674535, "step": 643, "time_per_iteration": 4.047010660171509 }, { "auxiliary_loss_clip": 0.01255313, "auxiliary_loss_mlp": 0.01043533, "balance_loss_clip": 1.07480264, "balance_loss_mlp": 1.03199935, "epoch": 0.07743642157157458, "flos": 21324657993120.0, "grad_norm": 1.899902162372093, "language_loss": 0.87296027, "learning_rate": 3.976496999316775e-06, "loss": 0.89594877, "num_input_tokens_seen": 13693290, "step": 644, "time_per_iteration": 2.615112543106079 }, { "auxiliary_loss_clip": 0.012356, "auxiliary_loss_mlp": 0.01043411, "balance_loss_clip": 1.07267797, "balance_loss_mlp": 1.03209162, "epoch": 0.07755666446221367, "flos": 19969136871840.0, "grad_norm": 1.9289035013548004, "language_loss": 0.84396875, "learning_rate": 3.976377779050271e-06, "loss": 0.86675888, "num_input_tokens_seen": 13711420, "step": 645, "time_per_iteration": 2.6171135902404785 }, { "auxiliary_loss_clip": 0.01239976, "auxiliary_loss_mlp": 0.01046933, "balance_loss_clip": 1.06697738, "balance_loss_mlp": 1.03495276, "epoch": 0.07767690735285276, "flos": 23623228304640.0, "grad_norm": 4.39746002510127, "language_loss": 0.84358138, "learning_rate": 3.976258258968831e-06, "loss": 0.86645049, "num_input_tokens_seen": 13729965, "step": 646, "time_per_iteration": 3.513695001602173 }, { "auxiliary_loss_clip": 0.01221032, "auxiliary_loss_mlp": 0.01044622, "balance_loss_clip": 1.06923676, "balance_loss_mlp": 1.03427422, "epoch": 0.07779715024349185, "flos": 22236250314720.0, "grad_norm": 2.0639699952079327, "language_loss": 0.74398482, "learning_rate": 3.976138439090583e-06, "loss": 0.76664144, "num_input_tokens_seen": 13748045, "step": 647, "time_per_iteration": 3.549034833908081 }, { "auxiliary_loss_clip": 0.01224015, "auxiliary_loss_mlp": 0.01047899, "balance_loss_clip": 1.07086349, "balance_loss_mlp": 1.03542376, "epoch": 0.07791739313413094, "flos": 20955100440960.0, "grad_norm": 2.22265543602847, "language_loss": 0.84881407, "learning_rate": 3.976018319433706e-06, "loss": 0.87153322, "num_input_tokens_seen": 13765590, "step": 648, "time_per_iteration": 2.6584606170654297 }, { "auxiliary_loss_clip": 0.01255692, "auxiliary_loss_mlp": 0.01050315, "balance_loss_clip": 1.07511973, "balance_loss_mlp": 1.0377202, "epoch": 0.07803763602477004, "flos": 19312330663200.0, "grad_norm": 2.3113334675778634, "language_loss": 0.91551644, "learning_rate": 3.9758979000164205e-06, "loss": 0.93857652, "num_input_tokens_seen": 13782410, "step": 649, "time_per_iteration": 2.5739452838897705 }, { "auxiliary_loss_clip": 0.01226983, "auxiliary_loss_mlp": 0.01051691, "balance_loss_clip": 1.07012987, "balance_loss_mlp": 1.03945422, "epoch": 0.07815787891540912, "flos": 22710809892960.0, "grad_norm": 1.748979475468767, "language_loss": 0.72137284, "learning_rate": 3.975777180856995e-06, "loss": 0.74415958, "num_input_tokens_seen": 13801530, "step": 650, "time_per_iteration": 2.6558690071105957 }, { "auxiliary_loss_clip": 0.01272112, "auxiliary_loss_mlp": 0.0104778, "balance_loss_clip": 1.07408893, "balance_loss_mlp": 1.03616309, "epoch": 0.07827812180604822, "flos": 22711133145600.0, "grad_norm": 4.064738582809697, "language_loss": 0.8623333, "learning_rate": 3.975656161973742e-06, "loss": 0.8855322, "num_input_tokens_seen": 13820615, "step": 651, "time_per_iteration": 2.566561460494995 }, { "auxiliary_loss_clip": 0.01269123, "auxiliary_loss_mlp": 0.01044341, "balance_loss_clip": 1.07168925, "balance_loss_mlp": 1.03228343, "epoch": 0.0783983646966873, "flos": 21725600580000.0, "grad_norm": 3.4512447336649843, "language_loss": 0.89009053, "learning_rate": 3.9755348433850194e-06, "loss": 0.91322517, "num_input_tokens_seen": 13835955, "step": 652, "time_per_iteration": 2.5539796352386475 }, { "auxiliary_loss_clip": 0.01137692, "auxiliary_loss_mlp": 0.01003222, "balance_loss_clip": 1.03610396, "balance_loss_mlp": 1.00027776, "epoch": 0.0785186075873264, "flos": 60640886757120.0, "grad_norm": 0.9560486743902353, "language_loss": 0.63601327, "learning_rate": 3.975413225109232e-06, "loss": 0.65742242, "num_input_tokens_seen": 13896505, "step": 653, "time_per_iteration": 3.2324118614196777 }, { "auxiliary_loss_clip": 0.01253867, "auxiliary_loss_mlp": 0.01046193, "balance_loss_clip": 1.07366502, "balance_loss_mlp": 1.0338068, "epoch": 0.0786388504779655, "flos": 23877906666720.0, "grad_norm": 2.899805713875974, "language_loss": 0.93841457, "learning_rate": 3.975291307164829e-06, "loss": 0.96141517, "num_input_tokens_seen": 13915150, "step": 654, "time_per_iteration": 2.609147548675537 }, { "auxiliary_loss_clip": 0.01208062, "auxiliary_loss_mlp": 0.01040963, "balance_loss_clip": 1.06224978, "balance_loss_mlp": 1.02942991, "epoch": 0.07875909336860458, "flos": 15158687836320.0, "grad_norm": 3.087246181973293, "language_loss": 0.85190058, "learning_rate": 3.975169089570306e-06, "loss": 0.87439084, "num_input_tokens_seen": 13933525, "step": 655, "time_per_iteration": 2.6472885608673096 }, { "auxiliary_loss_clip": 0.01239019, "auxiliary_loss_mlp": 0.0103975, "balance_loss_clip": 1.06826758, "balance_loss_mlp": 1.02820468, "epoch": 0.07887933625924368, "flos": 22236860903040.0, "grad_norm": 1.8463111453986218, "language_loss": 0.91521645, "learning_rate": 3.975046572344202e-06, "loss": 0.93800414, "num_input_tokens_seen": 13949985, "step": 656, "time_per_iteration": 2.606722354888916 }, { "auxiliary_loss_clip": 0.01214431, "auxiliary_loss_mlp": 0.01044105, "balance_loss_clip": 1.06452572, "balance_loss_mlp": 1.03282189, "epoch": 0.07899957914988276, "flos": 20777738517120.0, "grad_norm": 2.0118523038510894, "language_loss": 0.71199465, "learning_rate": 3.974923755505103e-06, "loss": 0.73458004, "num_input_tokens_seen": 13969215, "step": 657, "time_per_iteration": 2.625723123550415 }, { "auxiliary_loss_clip": 0.01210831, "auxiliary_loss_mlp": 0.01041984, "balance_loss_clip": 1.06661606, "balance_loss_mlp": 1.03113019, "epoch": 0.07911982204052186, "flos": 23003051006880.0, "grad_norm": 1.8473291551461049, "language_loss": 0.91113043, "learning_rate": 3.974800639071641e-06, "loss": 0.9336586, "num_input_tokens_seen": 13989935, "step": 658, "time_per_iteration": 2.6875298023223877 }, { "auxiliary_loss_clip": 0.01170506, "auxiliary_loss_mlp": 0.00767073, "balance_loss_clip": 1.05922651, "balance_loss_mlp": 1.0004009, "epoch": 0.07924006493116094, "flos": 23111393310240.0, "grad_norm": 2.487386775861614, "language_loss": 1.00972247, "learning_rate": 3.974677223062492e-06, "loss": 1.02909827, "num_input_tokens_seen": 14007150, "step": 659, "time_per_iteration": 2.684455156326294 }, { "auxiliary_loss_clip": 0.01233908, "auxiliary_loss_mlp": 0.01043487, "balance_loss_clip": 1.07072544, "balance_loss_mlp": 1.03210902, "epoch": 0.07936030782180004, "flos": 16472156585760.0, "grad_norm": 2.07976271559783, "language_loss": 0.74502087, "learning_rate": 3.974553507496378e-06, "loss": 0.76779479, "num_input_tokens_seen": 14025725, "step": 660, "time_per_iteration": 2.6068992614746094 }, { "auxiliary_loss_clip": 0.01224388, "auxiliary_loss_mlp": 0.01049343, "balance_loss_clip": 1.06697166, "balance_loss_mlp": 1.03695726, "epoch": 0.07948055071243913, "flos": 23733294622080.0, "grad_norm": 1.8940007016820095, "language_loss": 0.88883907, "learning_rate": 3.974429492392068e-06, "loss": 0.91157639, "num_input_tokens_seen": 14045750, "step": 661, "time_per_iteration": 2.631363868713379 }, { "auxiliary_loss_clip": 0.01265364, "auxiliary_loss_mlp": 0.00767321, "balance_loss_clip": 1.0735333, "balance_loss_mlp": 1.0003159, "epoch": 0.07960079360307822, "flos": 19573330410240.0, "grad_norm": 1.9005625739221557, "language_loss": 0.90842605, "learning_rate": 3.974305177768373e-06, "loss": 0.9287529, "num_input_tokens_seen": 14063960, "step": 662, "time_per_iteration": 2.55228590965271 }, { "auxiliary_loss_clip": 0.01208719, "auxiliary_loss_mlp": 0.01040564, "balance_loss_clip": 1.06875288, "balance_loss_mlp": 1.02854788, "epoch": 0.07972103649371731, "flos": 23513413405920.0, "grad_norm": 1.9777769968826948, "language_loss": 0.86563408, "learning_rate": 3.974180563644152e-06, "loss": 0.88812697, "num_input_tokens_seen": 14082525, "step": 663, "time_per_iteration": 2.644479990005493 }, { "auxiliary_loss_clip": 0.01236571, "auxiliary_loss_mlp": 0.01043305, "balance_loss_clip": 1.06857979, "balance_loss_mlp": 1.03194416, "epoch": 0.0798412793843564, "flos": 16726870864800.0, "grad_norm": 2.692245963748258, "language_loss": 0.89277577, "learning_rate": 3.97405565003831e-06, "loss": 0.91557449, "num_input_tokens_seen": 14098610, "step": 664, "time_per_iteration": 2.5997250080108643 }, { "auxiliary_loss_clip": 0.01223989, "auxiliary_loss_mlp": 0.01038329, "balance_loss_clip": 1.0672617, "balance_loss_mlp": 1.02658749, "epoch": 0.07996152227499549, "flos": 18223340500800.0, "grad_norm": 2.1241909152335525, "language_loss": 0.78551757, "learning_rate": 3.973930436969794e-06, "loss": 0.80814075, "num_input_tokens_seen": 14117065, "step": 665, "time_per_iteration": 2.6303303241729736 }, { "auxiliary_loss_clip": 0.01226437, "auxiliary_loss_mlp": 0.01049763, "balance_loss_clip": 1.06574321, "balance_loss_mlp": 1.03741312, "epoch": 0.08008176516563459, "flos": 20594881298400.0, "grad_norm": 1.807549271606789, "language_loss": 0.85802567, "learning_rate": 3.973804924457602e-06, "loss": 0.88078773, "num_input_tokens_seen": 14135145, "step": 666, "time_per_iteration": 2.636333465576172 }, { "auxiliary_loss_clip": 0.01227089, "auxiliary_loss_mlp": 0.01047896, "balance_loss_clip": 1.06479335, "balance_loss_mlp": 1.03627276, "epoch": 0.08020200805627367, "flos": 31834311587520.0, "grad_norm": 1.6843480560798205, "language_loss": 0.85682768, "learning_rate": 3.973679112520771e-06, "loss": 0.87957752, "num_input_tokens_seen": 14156860, "step": 667, "time_per_iteration": 2.703010082244873 }, { "auxiliary_loss_clip": 0.01206901, "auxiliary_loss_mlp": 0.01046275, "balance_loss_clip": 1.06291318, "balance_loss_mlp": 1.03523624, "epoch": 0.08032225094691277, "flos": 17783506234560.0, "grad_norm": 1.9135056882573207, "language_loss": 0.98694986, "learning_rate": 3.973553001178389e-06, "loss": 1.00948155, "num_input_tokens_seen": 14174365, "step": 668, "time_per_iteration": 3.4046833515167236 }, { "auxiliary_loss_clip": 0.01218732, "auxiliary_loss_mlp": 0.01051299, "balance_loss_clip": 1.06921339, "balance_loss_mlp": 1.0400461, "epoch": 0.08044249383755185, "flos": 24061697726400.0, "grad_norm": 2.054044033331302, "language_loss": 0.75487179, "learning_rate": 3.973426590449585e-06, "loss": 0.77757215, "num_input_tokens_seen": 14192320, "step": 669, "time_per_iteration": 3.483915090560913 }, { "auxiliary_loss_clip": 0.01205089, "auxiliary_loss_mlp": 0.01054496, "balance_loss_clip": 1.06722045, "balance_loss_mlp": 1.04365396, "epoch": 0.08056273672819095, "flos": 18223627836480.0, "grad_norm": 1.9503410723639114, "language_loss": 0.75182366, "learning_rate": 3.9732998803535364e-06, "loss": 0.77441949, "num_input_tokens_seen": 14210380, "step": 670, "time_per_iteration": 2.6637461185455322 }, { "auxiliary_loss_clip": 0.01266523, "auxiliary_loss_mlp": 0.01047016, "balance_loss_clip": 1.07374823, "balance_loss_mlp": 1.03562021, "epoch": 0.08068297961883003, "flos": 19676859840960.0, "grad_norm": 2.534783188288381, "language_loss": 0.85175842, "learning_rate": 3.973172870909465e-06, "loss": 0.87489378, "num_input_tokens_seen": 14225145, "step": 671, "time_per_iteration": 2.5401675701141357 }, { "auxiliary_loss_clip": 0.01237868, "auxiliary_loss_mlp": 0.01039819, "balance_loss_clip": 1.06693482, "balance_loss_mlp": 1.02853012, "epoch": 0.08080322250946913, "flos": 23148740560320.0, "grad_norm": 2.394811430618414, "language_loss": 0.80499148, "learning_rate": 3.973045562136638e-06, "loss": 0.82776833, "num_input_tokens_seen": 14241960, "step": 672, "time_per_iteration": 3.5317881107330322 }, { "auxiliary_loss_clip": 0.01254752, "auxiliary_loss_mlp": 0.01038979, "balance_loss_clip": 1.07111192, "balance_loss_mlp": 1.02817309, "epoch": 0.08092346540010822, "flos": 21763630252320.0, "grad_norm": 1.922985598822951, "language_loss": 0.91393989, "learning_rate": 3.972917954054368e-06, "loss": 0.93687725, "num_input_tokens_seen": 14260515, "step": 673, "time_per_iteration": 2.587735176086426 }, { "auxiliary_loss_clip": 0.01235661, "auxiliary_loss_mlp": 0.01054887, "balance_loss_clip": 1.07348847, "balance_loss_mlp": 1.04178548, "epoch": 0.08104370829074731, "flos": 21032488713120.0, "grad_norm": 2.206928161078017, "language_loss": 0.81872004, "learning_rate": 3.972790046682013e-06, "loss": 0.84162551, "num_input_tokens_seen": 14279190, "step": 674, "time_per_iteration": 2.629478931427002 }, { "auxiliary_loss_clip": 0.01223248, "auxiliary_loss_mlp": 0.01041753, "balance_loss_clip": 1.06551862, "balance_loss_mlp": 1.03053498, "epoch": 0.0811639511813864, "flos": 20083189971840.0, "grad_norm": 2.0247510525389343, "language_loss": 0.79153764, "learning_rate": 3.972661840038977e-06, "loss": 0.81418765, "num_input_tokens_seen": 14299480, "step": 675, "time_per_iteration": 2.6529576778411865 }, { "auxiliary_loss_clip": 0.01250619, "auxiliary_loss_mlp": 0.01056901, "balance_loss_clip": 1.07286477, "balance_loss_mlp": 1.04502726, "epoch": 0.08128419407202549, "flos": 16836721680480.0, "grad_norm": 2.04646764326879, "language_loss": 0.83413124, "learning_rate": 3.972533334144707e-06, "loss": 0.85720646, "num_input_tokens_seen": 14316405, "step": 676, "time_per_iteration": 2.5777246952056885 }, { "auxiliary_loss_clip": 0.01257397, "auxiliary_loss_mlp": 0.0104269, "balance_loss_clip": 1.0715189, "balance_loss_mlp": 1.03098953, "epoch": 0.08140443696266458, "flos": 23769277027680.0, "grad_norm": 1.8850697466431474, "language_loss": 0.78578085, "learning_rate": 3.972404529018699e-06, "loss": 0.80878174, "num_input_tokens_seen": 14336265, "step": 677, "time_per_iteration": 2.594007968902588 }, { "auxiliary_loss_clip": 0.01233365, "auxiliary_loss_mlp": 0.01055042, "balance_loss_clip": 1.06531954, "balance_loss_mlp": 1.04350877, "epoch": 0.08152467985330367, "flos": 24390136747680.0, "grad_norm": 1.7157278784328458, "language_loss": 0.85358882, "learning_rate": 3.972275424680493e-06, "loss": 0.87647289, "num_input_tokens_seen": 14356375, "step": 678, "time_per_iteration": 2.6532905101776123 }, { "auxiliary_loss_clip": 0.01268312, "auxiliary_loss_mlp": 0.01037379, "balance_loss_clip": 1.07483411, "balance_loss_mlp": 1.02556503, "epoch": 0.08164492274394276, "flos": 19317754124160.0, "grad_norm": 1.9624161231821298, "language_loss": 0.91922826, "learning_rate": 3.972146021149673e-06, "loss": 0.94228518, "num_input_tokens_seen": 14374650, "step": 679, "time_per_iteration": 2.5253312587738037 }, { "auxiliary_loss_clip": 0.01220749, "auxiliary_loss_mlp": 0.01040341, "balance_loss_clip": 1.0671556, "balance_loss_mlp": 1.02949262, "epoch": 0.08176516563458186, "flos": 14830464316800.0, "grad_norm": 2.366851439464178, "language_loss": 0.78652883, "learning_rate": 3.972016318445868e-06, "loss": 0.80913973, "num_input_tokens_seen": 14392650, "step": 680, "time_per_iteration": 2.626128911972046 }, { "auxiliary_loss_clip": 0.01248808, "auxiliary_loss_mlp": 0.01048745, "balance_loss_clip": 1.06973958, "balance_loss_mlp": 1.03605533, "epoch": 0.08188540852522094, "flos": 22602323921760.0, "grad_norm": 2.065302033460663, "language_loss": 0.92330778, "learning_rate": 3.971886316588757e-06, "loss": 0.94628322, "num_input_tokens_seen": 14413155, "step": 681, "time_per_iteration": 2.599038600921631 }, { "auxiliary_loss_clip": 0.01208258, "auxiliary_loss_mlp": 0.01042671, "balance_loss_clip": 1.06547308, "balance_loss_mlp": 1.03042245, "epoch": 0.08200565141586004, "flos": 19463731013280.0, "grad_norm": 2.1811427640453434, "language_loss": 0.73602021, "learning_rate": 3.9717560155980595e-06, "loss": 0.75852948, "num_input_tokens_seen": 14428805, "step": 682, "time_per_iteration": 2.6308248043060303 }, { "auxiliary_loss_clip": 0.01248075, "auxiliary_loss_mlp": 0.0104374, "balance_loss_clip": 1.06952977, "balance_loss_mlp": 1.03240371, "epoch": 0.08212589430649912, "flos": 20594665796640.0, "grad_norm": 2.448013509228069, "language_loss": 0.91893244, "learning_rate": 3.971625415493542e-06, "loss": 0.94185054, "num_input_tokens_seen": 14447125, "step": 683, "time_per_iteration": 2.577993392944336 }, { "auxiliary_loss_clip": 0.01214753, "auxiliary_loss_mlp": 0.01046061, "balance_loss_clip": 1.06538224, "balance_loss_mlp": 1.0335623, "epoch": 0.08224613719713822, "flos": 25953614654400.0, "grad_norm": 3.665846383322527, "language_loss": 0.87658441, "learning_rate": 3.971494516295017e-06, "loss": 0.89919257, "num_input_tokens_seen": 14466575, "step": 684, "time_per_iteration": 2.697868824005127 }, { "auxiliary_loss_clip": 0.01221259, "auxiliary_loss_mlp": 0.01045835, "balance_loss_clip": 1.06511927, "balance_loss_mlp": 1.03347945, "epoch": 0.08236638008777732, "flos": 23768738273280.0, "grad_norm": 4.590327462501679, "language_loss": 0.85470939, "learning_rate": 3.971363318022341e-06, "loss": 0.87738037, "num_input_tokens_seen": 14487915, "step": 685, "time_per_iteration": 2.649477958679199 }, { "auxiliary_loss_clip": 0.01237232, "auxiliary_loss_mlp": 0.01042516, "balance_loss_clip": 1.0665133, "balance_loss_mlp": 1.03112578, "epoch": 0.0824866229784164, "flos": 38799149893440.0, "grad_norm": 2.0088377674615137, "language_loss": 0.68475688, "learning_rate": 3.971231820695417e-06, "loss": 0.70755434, "num_input_tokens_seen": 14511530, "step": 686, "time_per_iteration": 2.756894588470459 }, { "auxiliary_loss_clip": 0.01238003, "auxiliary_loss_mlp": 0.01047761, "balance_loss_clip": 1.06789088, "balance_loss_mlp": 1.03516054, "epoch": 0.0826068658690555, "flos": 23107765697280.0, "grad_norm": 4.964395827449639, "language_loss": 0.81477463, "learning_rate": 3.971100024334193e-06, "loss": 0.8376323, "num_input_tokens_seen": 14529050, "step": 687, "time_per_iteration": 2.614535331726074 }, { "auxiliary_loss_clip": 0.01200727, "auxiliary_loss_mlp": 0.01058487, "balance_loss_clip": 1.05879116, "balance_loss_mlp": 1.04795456, "epoch": 0.08272710875969458, "flos": 21136377313440.0, "grad_norm": 2.075356059702937, "language_loss": 0.86393219, "learning_rate": 3.970967928958663e-06, "loss": 0.88652432, "num_input_tokens_seen": 14546165, "step": 688, "time_per_iteration": 2.664027214050293 }, { "auxiliary_loss_clip": 0.01213124, "auxiliary_loss_mlp": 0.010451, "balance_loss_clip": 1.06792665, "balance_loss_mlp": 1.03407896, "epoch": 0.08284735165033368, "flos": 19063003928160.0, "grad_norm": 1.7039237018079467, "language_loss": 0.83587772, "learning_rate": 3.970835534588865e-06, "loss": 0.85845995, "num_input_tokens_seen": 14563660, "step": 689, "time_per_iteration": 2.728610038757324 }, { "auxiliary_loss_clip": 0.01244856, "auxiliary_loss_mlp": 0.01047818, "balance_loss_clip": 1.07709575, "balance_loss_mlp": 1.0356586, "epoch": 0.08296759454097276, "flos": 16727445536160.0, "grad_norm": 2.2017384311625263, "language_loss": 0.85753608, "learning_rate": 3.970702841244883e-06, "loss": 0.88046288, "num_input_tokens_seen": 14581980, "step": 690, "time_per_iteration": 2.6316030025482178 }, { "auxiliary_loss_clip": 0.01258292, "auxiliary_loss_mlp": 0.01044085, "balance_loss_clip": 1.07492149, "balance_loss_mlp": 1.0319736, "epoch": 0.08308783743161186, "flos": 18004931880000.0, "grad_norm": 1.7449024730933007, "language_loss": 0.82504135, "learning_rate": 3.970569848946847e-06, "loss": 0.84806514, "num_input_tokens_seen": 14601795, "step": 691, "time_per_iteration": 2.599276304244995 }, { "auxiliary_loss_clip": 0.01235469, "auxiliary_loss_mlp": 0.01042419, "balance_loss_clip": 1.06763077, "balance_loss_mlp": 1.03055215, "epoch": 0.08320808032225095, "flos": 15079791051840.0, "grad_norm": 2.2899876807529376, "language_loss": 0.82973105, "learning_rate": 3.970436557714932e-06, "loss": 0.85250992, "num_input_tokens_seen": 14618315, "step": 692, "time_per_iteration": 2.4932539463043213 }, { "auxiliary_loss_clip": 0.01232346, "auxiliary_loss_mlp": 0.01041988, "balance_loss_clip": 1.0692699, "balance_loss_mlp": 1.02981615, "epoch": 0.08332832321289003, "flos": 22383089210880.0, "grad_norm": 2.17857082108315, "language_loss": 0.8687014, "learning_rate": 3.970302967569358e-06, "loss": 0.89144474, "num_input_tokens_seen": 14636905, "step": 693, "time_per_iteration": 2.6036031246185303 }, { "auxiliary_loss_clip": 0.0124917, "auxiliary_loss_mlp": 0.01049419, "balance_loss_clip": 1.07132196, "balance_loss_mlp": 1.03745627, "epoch": 0.08344856610352913, "flos": 24717390509280.0, "grad_norm": 1.7975760995920118, "language_loss": 0.68342853, "learning_rate": 3.9701690785303896e-06, "loss": 0.7064144, "num_input_tokens_seen": 14656100, "step": 694, "time_per_iteration": 3.3052356243133545 }, { "auxiliary_loss_clip": 0.0125486, "auxiliary_loss_mlp": 0.01047786, "balance_loss_clip": 1.06996143, "balance_loss_mlp": 1.03591299, "epoch": 0.08356880899416821, "flos": 25370209935360.0, "grad_norm": 2.243191464352142, "language_loss": 0.87676752, "learning_rate": 3.970034890618339e-06, "loss": 0.89979398, "num_input_tokens_seen": 14675790, "step": 695, "time_per_iteration": 3.3555333614349365 }, { "auxiliary_loss_clip": 0.01230871, "auxiliary_loss_mlp": 0.01047939, "balance_loss_clip": 1.06351411, "balance_loss_mlp": 1.03625655, "epoch": 0.08368905188480731, "flos": 24353076833280.0, "grad_norm": 2.576162672876023, "language_loss": 0.88356799, "learning_rate": 3.969900403853562e-06, "loss": 0.90635604, "num_input_tokens_seen": 14694830, "step": 696, "time_per_iteration": 2.5880281925201416 }, { "auxiliary_loss_clip": 0.01270554, "auxiliary_loss_mlp": 0.01044435, "balance_loss_clip": 1.07592189, "balance_loss_mlp": 1.0325911, "epoch": 0.08380929477544641, "flos": 18037322589600.0, "grad_norm": 2.0138650684052113, "language_loss": 0.78062308, "learning_rate": 3.96976561825646e-06, "loss": 0.80377293, "num_input_tokens_seen": 14711920, "step": 697, "time_per_iteration": 2.528076410293579 }, { "auxiliary_loss_clip": 0.01210451, "auxiliary_loss_mlp": 0.01044216, "balance_loss_clip": 1.06595683, "balance_loss_mlp": 1.03271198, "epoch": 0.08392953766608549, "flos": 26286291876960.0, "grad_norm": 2.270291254016474, "language_loss": 0.86722386, "learning_rate": 3.969630533847479e-06, "loss": 0.88977051, "num_input_tokens_seen": 14730880, "step": 698, "time_per_iteration": 3.5605340003967285 }, { "auxiliary_loss_clip": 0.01249898, "auxiliary_loss_mlp": 0.01044603, "balance_loss_clip": 1.06947649, "balance_loss_mlp": 1.03302169, "epoch": 0.08404978055672459, "flos": 22492149853440.0, "grad_norm": 1.8658603942100953, "language_loss": 0.84343112, "learning_rate": 3.969495150647113e-06, "loss": 0.8663761, "num_input_tokens_seen": 14749050, "step": 699, "time_per_iteration": 3.4372196197509766 }, { "auxiliary_loss_clip": 0.01221245, "auxiliary_loss_mlp": 0.01043036, "balance_loss_clip": 1.07222223, "balance_loss_mlp": 1.03146684, "epoch": 0.08417002344736367, "flos": 24826882155360.0, "grad_norm": 1.8450217669927538, "language_loss": 0.76418275, "learning_rate": 3.969359468675899e-06, "loss": 0.78682554, "num_input_tokens_seen": 14769180, "step": 700, "time_per_iteration": 2.687025785446167 }, { "auxiliary_loss_clip": 0.0124868, "auxiliary_loss_mlp": 0.01043564, "balance_loss_clip": 1.07286096, "balance_loss_mlp": 1.03160691, "epoch": 0.08429026633800277, "flos": 16945925990880.0, "grad_norm": 1.9771437248128245, "language_loss": 0.89274311, "learning_rate": 3.969223487954418e-06, "loss": 0.91566563, "num_input_tokens_seen": 14786640, "step": 701, "time_per_iteration": 2.561228036880493 }, { "auxiliary_loss_clip": 0.01203125, "auxiliary_loss_mlp": 0.01048575, "balance_loss_clip": 1.06780124, "balance_loss_mlp": 1.03760159, "epoch": 0.08441050922864185, "flos": 23841924261120.0, "grad_norm": 2.20701243814835, "language_loss": 0.8258267, "learning_rate": 3.969087208503301e-06, "loss": 0.84834367, "num_input_tokens_seen": 14806720, "step": 702, "time_per_iteration": 2.726752758026123 }, { "auxiliary_loss_clip": 0.01202521, "auxiliary_loss_mlp": 0.01041669, "balance_loss_clip": 1.06536329, "balance_loss_mlp": 1.02989697, "epoch": 0.08453075211928095, "flos": 25520209524000.0, "grad_norm": 2.952209623237092, "language_loss": 0.84710026, "learning_rate": 3.968950630343219e-06, "loss": 0.86954212, "num_input_tokens_seen": 14823705, "step": 703, "time_per_iteration": 2.6986191272735596 }, { "auxiliary_loss_clip": 0.01228655, "auxiliary_loss_mlp": 0.010461, "balance_loss_clip": 1.06386876, "balance_loss_mlp": 1.03448331, "epoch": 0.08465099500992004, "flos": 19532499215040.0, "grad_norm": 1.9251053671920946, "language_loss": 0.93410408, "learning_rate": 3.968813753494892e-06, "loss": 0.95685166, "num_input_tokens_seen": 14841865, "step": 704, "time_per_iteration": 2.622133731842041 }, { "auxiliary_loss_clip": 0.01204788, "auxiliary_loss_mlp": 0.00766881, "balance_loss_clip": 1.06206989, "balance_loss_mlp": 1.00037646, "epoch": 0.08477123790055913, "flos": 29351303711040.0, "grad_norm": 2.1049528078682256, "language_loss": 0.75534314, "learning_rate": 3.968676577979084e-06, "loss": 0.77505982, "num_input_tokens_seen": 14861415, "step": 705, "time_per_iteration": 2.767460823059082 }, { "auxiliary_loss_clip": 0.0119911, "auxiliary_loss_mlp": 0.01051928, "balance_loss_clip": 1.06408668, "balance_loss_mlp": 1.03970277, "epoch": 0.08489148079119822, "flos": 18624498589440.0, "grad_norm": 2.214283895262732, "language_loss": 0.78453159, "learning_rate": 3.968539103816605e-06, "loss": 0.807042, "num_input_tokens_seen": 14879215, "step": 706, "time_per_iteration": 2.6421942710876465 }, { "auxiliary_loss_clip": 0.0122921, "auxiliary_loss_mlp": 0.00766888, "balance_loss_clip": 1.06710362, "balance_loss_mlp": 1.000283, "epoch": 0.0850117236818373, "flos": 23471396951040.0, "grad_norm": 1.8254291683462611, "language_loss": 0.89570355, "learning_rate": 3.9684013310283085e-06, "loss": 0.91566449, "num_input_tokens_seen": 14897900, "step": 707, "time_per_iteration": 2.695338726043701 }, { "auxiliary_loss_clip": 0.01234473, "auxiliary_loss_mlp": 0.01042621, "balance_loss_clip": 1.07389712, "balance_loss_mlp": 1.03123093, "epoch": 0.0851319665724764, "flos": 40625064225600.0, "grad_norm": 1.7255978248581467, "language_loss": 0.63904679, "learning_rate": 3.9682632596350956e-06, "loss": 0.66181773, "num_input_tokens_seen": 14919065, "step": 708, "time_per_iteration": 2.7661330699920654 }, { "auxiliary_loss_clip": 0.01244837, "auxiliary_loss_mlp": 0.01047751, "balance_loss_clip": 1.07192206, "balance_loss_mlp": 1.03629446, "epoch": 0.0852522094631155, "flos": 15879557124960.0, "grad_norm": 1.954084640640314, "language_loss": 0.78658426, "learning_rate": 3.968124889657911e-06, "loss": 0.80951011, "num_input_tokens_seen": 14934165, "step": 709, "time_per_iteration": 2.5955276489257812 }, { "auxiliary_loss_clip": 0.0120446, "auxiliary_loss_mlp": 0.01038542, "balance_loss_clip": 1.06516242, "balance_loss_mlp": 1.02705038, "epoch": 0.08537245235375458, "flos": 14567094050400.0, "grad_norm": 2.363535672857073, "language_loss": 0.90521699, "learning_rate": 3.967986221117746e-06, "loss": 0.92764693, "num_input_tokens_seen": 14950105, "step": 710, "time_per_iteration": 2.636207103729248 }, { "auxiliary_loss_clip": 0.01178989, "auxiliary_loss_mlp": 0.01040887, "balance_loss_clip": 1.06004632, "balance_loss_mlp": 1.02963948, "epoch": 0.08549269524439368, "flos": 26468933593920.0, "grad_norm": 1.8754474470097464, "language_loss": 0.86652744, "learning_rate": 3.967847254035635e-06, "loss": 0.88872617, "num_input_tokens_seen": 14969490, "step": 711, "time_per_iteration": 2.821092367172241 }, { "auxiliary_loss_clip": 0.01222064, "auxiliary_loss_mlp": 0.01036092, "balance_loss_clip": 1.06737089, "balance_loss_mlp": 1.0249517, "epoch": 0.08561293813503276, "flos": 13590217472160.0, "grad_norm": 2.2515563750856655, "language_loss": 0.86775267, "learning_rate": 3.967707988432661e-06, "loss": 0.89033425, "num_input_tokens_seen": 14987195, "step": 712, "time_per_iteration": 2.6429872512817383 }, { "auxiliary_loss_clip": 0.01263824, "auxiliary_loss_mlp": 0.01046552, "balance_loss_clip": 1.06942129, "balance_loss_mlp": 1.03517318, "epoch": 0.08573318102567186, "flos": 26943529089120.0, "grad_norm": 2.165625046157137, "language_loss": 0.87555778, "learning_rate": 3.967568424329949e-06, "loss": 0.89866149, "num_input_tokens_seen": 15007620, "step": 713, "time_per_iteration": 2.623579263687134 }, { "auxiliary_loss_clip": 0.01156626, "auxiliary_loss_mlp": 0.01026552, "balance_loss_clip": 1.04770672, "balance_loss_mlp": 1.02354813, "epoch": 0.08585342391631094, "flos": 67302751166400.0, "grad_norm": 0.8308937936635009, "language_loss": 0.55483234, "learning_rate": 3.967428561748671e-06, "loss": 0.57666421, "num_input_tokens_seen": 15075590, "step": 714, "time_per_iteration": 3.3379924297332764 }, { "auxiliary_loss_clip": 0.01195401, "auxiliary_loss_mlp": 0.01042467, "balance_loss_clip": 1.06225169, "balance_loss_mlp": 1.03057551, "epoch": 0.08597366680695004, "flos": 22456598451360.0, "grad_norm": 2.055977488296055, "language_loss": 0.87591159, "learning_rate": 3.967288400710045e-06, "loss": 0.89829028, "num_input_tokens_seen": 15095055, "step": 715, "time_per_iteration": 2.688385009765625 }, { "auxiliary_loss_clip": 0.01215491, "auxiliary_loss_mlp": 0.01047225, "balance_loss_clip": 1.07277369, "balance_loss_mlp": 1.03562045, "epoch": 0.08609390969758914, "flos": 23550509237280.0, "grad_norm": 1.803849853795399, "language_loss": 0.88433421, "learning_rate": 3.9671479412353335e-06, "loss": 0.90696132, "num_input_tokens_seen": 15113520, "step": 716, "time_per_iteration": 2.6633379459381104 }, { "auxiliary_loss_clip": 0.01250839, "auxiliary_loss_mlp": 0.01040685, "balance_loss_clip": 1.07290483, "balance_loss_mlp": 1.029634, "epoch": 0.08621415258822822, "flos": 25885852127520.0, "grad_norm": 1.9627090842920427, "language_loss": 0.74268174, "learning_rate": 3.967007183345843e-06, "loss": 0.76559699, "num_input_tokens_seen": 15133375, "step": 717, "time_per_iteration": 2.633574962615967 }, { "auxiliary_loss_clip": 0.0124533, "auxiliary_loss_mlp": 0.01043354, "balance_loss_clip": 1.07140827, "balance_loss_mlp": 1.03220248, "epoch": 0.08633439547886732, "flos": 13589570966880.0, "grad_norm": 2.092230228134688, "language_loss": 0.89188898, "learning_rate": 3.966866127062927e-06, "loss": 0.91477585, "num_input_tokens_seen": 15150500, "step": 718, "time_per_iteration": 2.5567915439605713 }, { "auxiliary_loss_clip": 0.01149626, "auxiliary_loss_mlp": 0.01005922, "balance_loss_clip": 1.04198813, "balance_loss_mlp": 1.00287044, "epoch": 0.0864546383695064, "flos": 57767352212160.0, "grad_norm": 0.869563298957875, "language_loss": 0.6269623, "learning_rate": 3.966724772407982e-06, "loss": 0.64851779, "num_input_tokens_seen": 15208015, "step": 719, "time_per_iteration": 3.046557664871216 }, { "auxiliary_loss_clip": 0.01209148, "auxiliary_loss_mlp": 0.01051118, "balance_loss_clip": 1.06668723, "balance_loss_mlp": 1.03942966, "epoch": 0.0865748812601455, "flos": 20046956147520.0, "grad_norm": 2.1227046055478733, "language_loss": 0.89088327, "learning_rate": 3.966583119402454e-06, "loss": 0.91348588, "num_input_tokens_seen": 15224780, "step": 720, "time_per_iteration": 3.3629355430603027 }, { "auxiliary_loss_clip": 0.01245707, "auxiliary_loss_mlp": 0.00766791, "balance_loss_clip": 1.07083511, "balance_loss_mlp": 1.00030077, "epoch": 0.08669512415078459, "flos": 35262451837920.0, "grad_norm": 1.6429075392216845, "language_loss": 0.81991196, "learning_rate": 3.9664411680678305e-06, "loss": 0.84003699, "num_input_tokens_seen": 15246535, "step": 721, "time_per_iteration": 3.471534013748169 }, { "auxiliary_loss_clip": 0.01121953, "auxiliary_loss_mlp": 0.01011945, "balance_loss_clip": 1.03731728, "balance_loss_mlp": 1.00894058, "epoch": 0.08681536704142367, "flos": 65654881180320.0, "grad_norm": 0.848395468094242, "language_loss": 0.61456805, "learning_rate": 3.966298918425644e-06, "loss": 0.63590699, "num_input_tokens_seen": 15304025, "step": 722, "time_per_iteration": 3.087395191192627 }, { "auxiliary_loss_clip": 0.0125175, "auxiliary_loss_mlp": 0.01049365, "balance_loss_clip": 1.06954288, "balance_loss_mlp": 1.03710413, "epoch": 0.08693560993206277, "flos": 34529945454240.0, "grad_norm": 1.8987107179709113, "language_loss": 0.82914507, "learning_rate": 3.966156370497476e-06, "loss": 0.85215628, "num_input_tokens_seen": 15327635, "step": 723, "time_per_iteration": 2.7731807231903076 }, { "auxiliary_loss_clip": 0.01249611, "auxiliary_loss_mlp": 0.01048111, "balance_loss_clip": 1.06901693, "balance_loss_mlp": 1.03638709, "epoch": 0.08705585282270185, "flos": 23149422982560.0, "grad_norm": 1.9117490530188335, "language_loss": 0.88766122, "learning_rate": 3.96601352430495e-06, "loss": 0.91063845, "num_input_tokens_seen": 15347405, "step": 724, "time_per_iteration": 3.4927773475646973 }, { "auxiliary_loss_clip": 0.01230427, "auxiliary_loss_mlp": 0.01042028, "balance_loss_clip": 1.06939697, "balance_loss_mlp": 1.03082812, "epoch": 0.08717609571334095, "flos": 29497603852800.0, "grad_norm": 1.5393287929728863, "language_loss": 0.8330785, "learning_rate": 3.965870379869735e-06, "loss": 0.85580313, "num_input_tokens_seen": 15369450, "step": 725, "time_per_iteration": 2.7017087936401367 }, { "auxiliary_loss_clip": 0.01246914, "auxiliary_loss_mlp": 0.01040309, "balance_loss_clip": 1.06624246, "balance_loss_mlp": 1.02897215, "epoch": 0.08729633860398003, "flos": 20667492614880.0, "grad_norm": 2.055364775932136, "language_loss": 0.87002528, "learning_rate": 3.965726937213547e-06, "loss": 0.89289755, "num_input_tokens_seen": 15388085, "step": 726, "time_per_iteration": 2.606701135635376 }, { "auxiliary_loss_clip": 0.01246099, "auxiliary_loss_mlp": 0.01040973, "balance_loss_clip": 1.06596982, "balance_loss_mlp": 1.02939785, "epoch": 0.08741658149461913, "flos": 18369496974720.0, "grad_norm": 2.1548989280092723, "language_loss": 0.81150401, "learning_rate": 3.965583196358144e-06, "loss": 0.83437479, "num_input_tokens_seen": 15407120, "step": 727, "time_per_iteration": 2.5921289920806885 }, { "auxiliary_loss_clip": 0.01263811, "auxiliary_loss_mlp": 0.01042253, "balance_loss_clip": 1.07186604, "balance_loss_mlp": 1.03117812, "epoch": 0.08753682438525823, "flos": 18729680200320.0, "grad_norm": 2.282842696265938, "language_loss": 0.74731988, "learning_rate": 3.965439157325335e-06, "loss": 0.7703805, "num_input_tokens_seen": 15424485, "step": 728, "time_per_iteration": 2.5430901050567627 }, { "auxiliary_loss_clip": 0.01223214, "auxiliary_loss_mlp": 0.01041835, "balance_loss_clip": 1.06311536, "balance_loss_mlp": 1.03068256, "epoch": 0.08765706727589731, "flos": 27776116875360.0, "grad_norm": 1.7351992677988433, "language_loss": 0.7583493, "learning_rate": 3.965294820136968e-06, "loss": 0.78099978, "num_input_tokens_seen": 15446285, "step": 729, "time_per_iteration": 2.731264591217041 }, { "auxiliary_loss_clip": 0.01230316, "auxiliary_loss_mlp": 0.01045862, "balance_loss_clip": 1.07037592, "balance_loss_mlp": 1.03411973, "epoch": 0.08777731016653641, "flos": 24389131072800.0, "grad_norm": 1.9694577355233964, "language_loss": 0.86959857, "learning_rate": 3.965150184814938e-06, "loss": 0.89236033, "num_input_tokens_seen": 15465770, "step": 730, "time_per_iteration": 2.6599161624908447 }, { "auxiliary_loss_clip": 0.01218882, "auxiliary_loss_mlp": 0.01036718, "balance_loss_clip": 1.06563854, "balance_loss_mlp": 1.02517259, "epoch": 0.08789755305717549, "flos": 21981859288320.0, "grad_norm": 2.0468922117510875, "language_loss": 0.7686013, "learning_rate": 3.965005251381189e-06, "loss": 0.79115731, "num_input_tokens_seen": 15483705, "step": 731, "time_per_iteration": 2.6491644382476807 }, { "auxiliary_loss_clip": 0.01146048, "auxiliary_loss_mlp": 0.0100397, "balance_loss_clip": 1.03356433, "balance_loss_mlp": 1.0010252, "epoch": 0.08801779594781459, "flos": 58360131257760.0, "grad_norm": 0.8914969198226523, "language_loss": 0.64685982, "learning_rate": 3.964860019857705e-06, "loss": 0.66835999, "num_input_tokens_seen": 15548620, "step": 732, "time_per_iteration": 3.188491106033325 }, { "auxiliary_loss_clip": 0.01263199, "auxiliary_loss_mlp": 0.01047992, "balance_loss_clip": 1.07521129, "balance_loss_mlp": 1.03658354, "epoch": 0.08813803883845367, "flos": 23294789283360.0, "grad_norm": 1.7456551250293355, "language_loss": 0.84029067, "learning_rate": 3.964714490266518e-06, "loss": 0.86340261, "num_input_tokens_seen": 15569265, "step": 733, "time_per_iteration": 2.594614028930664 }, { "auxiliary_loss_clip": 0.01137066, "auxiliary_loss_mlp": 0.0100214, "balance_loss_clip": 1.02988958, "balance_loss_mlp": 0.99924326, "epoch": 0.08825828172909277, "flos": 63424935402720.0, "grad_norm": 0.8800853511586075, "language_loss": 0.64646608, "learning_rate": 3.964568662629706e-06, "loss": 0.66785812, "num_input_tokens_seen": 15630570, "step": 734, "time_per_iteration": 3.0598623752593994 }, { "auxiliary_loss_clip": 0.01240497, "auxiliary_loss_mlp": 0.01035886, "balance_loss_clip": 1.06661963, "balance_loss_mlp": 1.0239476, "epoch": 0.08837852461973186, "flos": 26720990017920.0, "grad_norm": 2.0885491683497506, "language_loss": 0.84568274, "learning_rate": 3.9644225369693895e-06, "loss": 0.86844659, "num_input_tokens_seen": 15650870, "step": 735, "time_per_iteration": 2.6526997089385986 }, { "auxiliary_loss_clip": 0.01261834, "auxiliary_loss_mlp": 0.01048388, "balance_loss_clip": 1.07306266, "balance_loss_mlp": 1.03748024, "epoch": 0.08849876751037095, "flos": 27265431223680.0, "grad_norm": 2.0051671669312516, "language_loss": 0.866947, "learning_rate": 3.964276113307735e-06, "loss": 0.89004922, "num_input_tokens_seen": 15670835, "step": 736, "time_per_iteration": 2.601654052734375 }, { "auxiliary_loss_clip": 0.01209653, "auxiliary_loss_mlp": 0.01034424, "balance_loss_clip": 1.06691086, "balance_loss_mlp": 1.02274716, "epoch": 0.08861901040101004, "flos": 19828763028480.0, "grad_norm": 1.738630278292907, "language_loss": 0.80643851, "learning_rate": 3.9641293916669574e-06, "loss": 0.8288793, "num_input_tokens_seen": 15689795, "step": 737, "time_per_iteration": 2.6905412673950195 }, { "auxiliary_loss_clip": 0.01214075, "auxiliary_loss_mlp": 0.01046599, "balance_loss_clip": 1.06921387, "balance_loss_mlp": 1.03401685, "epoch": 0.08873925329164913, "flos": 23658707872800.0, "grad_norm": 1.608158924019315, "language_loss": 0.83014953, "learning_rate": 3.9639823720693115e-06, "loss": 0.8527562, "num_input_tokens_seen": 15711650, "step": 738, "time_per_iteration": 2.6838138103485107 }, { "auxiliary_loss_clip": 0.0110935, "auxiliary_loss_mlp": 0.01016474, "balance_loss_clip": 1.03053737, "balance_loss_mlp": 1.01381612, "epoch": 0.08885949618228822, "flos": 71831554591200.0, "grad_norm": 0.8376220304353359, "language_loss": 0.60045445, "learning_rate": 3.963835054537102e-06, "loss": 0.62171268, "num_input_tokens_seen": 15780615, "step": 739, "time_per_iteration": 3.331188917160034 }, { "auxiliary_loss_clip": 0.01230363, "auxiliary_loss_mlp": 0.01045305, "balance_loss_clip": 1.06579161, "balance_loss_mlp": 1.03445745, "epoch": 0.08897973907292732, "flos": 22346172964320.0, "grad_norm": 2.081032253204623, "language_loss": 0.60596859, "learning_rate": 3.963687439092676e-06, "loss": 0.62872529, "num_input_tokens_seen": 15801300, "step": 740, "time_per_iteration": 2.616345167160034 }, { "auxiliary_loss_clip": 0.01243873, "auxiliary_loss_mlp": 0.01041864, "balance_loss_clip": 1.07071424, "balance_loss_mlp": 1.03074217, "epoch": 0.0890999819635664, "flos": 21251831174880.0, "grad_norm": 1.9250871521580135, "language_loss": 0.80762994, "learning_rate": 3.963539525758427e-06, "loss": 0.83048725, "num_input_tokens_seen": 15820860, "step": 741, "time_per_iteration": 2.5935487747192383 }, { "auxiliary_loss_clip": 0.01233097, "auxiliary_loss_mlp": 0.01037964, "balance_loss_clip": 1.06873202, "balance_loss_mlp": 1.02694941, "epoch": 0.0892202248542055, "flos": 25370892357600.0, "grad_norm": 1.9629189231414295, "language_loss": 0.68243265, "learning_rate": 3.9633913145567925e-06, "loss": 0.70514327, "num_input_tokens_seen": 15841350, "step": 742, "time_per_iteration": 2.6234467029571533 }, { "auxiliary_loss_clip": 0.01231023, "auxiliary_loss_mlp": 0.01044191, "balance_loss_clip": 1.06837416, "balance_loss_mlp": 1.03287196, "epoch": 0.08934046774484458, "flos": 24457899274560.0, "grad_norm": 1.7540886357060097, "language_loss": 0.81687677, "learning_rate": 3.9632428055102575e-06, "loss": 0.83962893, "num_input_tokens_seen": 15861360, "step": 743, "time_per_iteration": 2.6511242389678955 }, { "auxiliary_loss_clip": 0.01250628, "auxiliary_loss_mlp": 0.01049806, "balance_loss_clip": 1.0747236, "balance_loss_mlp": 1.03756356, "epoch": 0.08946071063548368, "flos": 35772778320000.0, "grad_norm": 1.9666127108424067, "language_loss": 0.6718843, "learning_rate": 3.9630939986413495e-06, "loss": 0.69488871, "num_input_tokens_seen": 15883160, "step": 744, "time_per_iteration": 2.712554454803467 }, { "auxiliary_loss_clip": 0.01199774, "auxiliary_loss_mlp": 0.01050158, "balance_loss_clip": 1.06409907, "balance_loss_mlp": 1.03892839, "epoch": 0.08958095352612276, "flos": 14356587160800.0, "grad_norm": 1.9184574518972797, "language_loss": 0.78229558, "learning_rate": 3.962944893972643e-06, "loss": 0.80479485, "num_input_tokens_seen": 15901610, "step": 745, "time_per_iteration": 3.377922296524048 }, { "auxiliary_loss_clip": 0.01228768, "auxiliary_loss_mlp": 0.01039113, "balance_loss_clip": 1.06935775, "balance_loss_mlp": 1.02770448, "epoch": 0.08970119641676186, "flos": 17853280111200.0, "grad_norm": 2.8172746724498436, "language_loss": 0.90707886, "learning_rate": 3.962795491526756e-06, "loss": 0.92975765, "num_input_tokens_seen": 15918770, "step": 746, "time_per_iteration": 2.6041882038116455 }, { "auxiliary_loss_clip": 0.01268735, "auxiliary_loss_mlp": 0.01052978, "balance_loss_clip": 1.07514095, "balance_loss_mlp": 1.0411942, "epoch": 0.08982143930740095, "flos": 20811673656000.0, "grad_norm": 2.4219497505576197, "language_loss": 0.89545721, "learning_rate": 3.962645791326354e-06, "loss": 0.91867435, "num_input_tokens_seen": 15938025, "step": 747, "time_per_iteration": 2.5432653427124023 }, { "auxiliary_loss_clip": 0.01242871, "auxiliary_loss_mlp": 0.01037875, "balance_loss_clip": 1.07353735, "balance_loss_mlp": 1.02776575, "epoch": 0.08994168219804004, "flos": 24097716048960.0, "grad_norm": 1.7956410883545904, "language_loss": 0.83301574, "learning_rate": 3.962495793394146e-06, "loss": 0.85582316, "num_input_tokens_seen": 15957215, "step": 748, "time_per_iteration": 3.3703391551971436 }, { "auxiliary_loss_clip": 0.01144781, "auxiliary_loss_mlp": 0.01004886, "balance_loss_clip": 1.02665067, "balance_loss_mlp": 1.00217962, "epoch": 0.09006192508867913, "flos": 57188904033600.0, "grad_norm": 0.7503670737964726, "language_loss": 0.61255711, "learning_rate": 3.9623454977528864e-06, "loss": 0.63405377, "num_input_tokens_seen": 16015870, "step": 749, "time_per_iteration": 2.9820303916931152 }, { "auxiliary_loss_clip": 0.01216023, "auxiliary_loss_mlp": 0.01045427, "balance_loss_clip": 1.06497264, "balance_loss_mlp": 1.03358912, "epoch": 0.09018216797931822, "flos": 20487508752960.0, "grad_norm": 1.7407241655831363, "language_loss": 0.85190523, "learning_rate": 3.962194904425375e-06, "loss": 0.87451971, "num_input_tokens_seen": 16036500, "step": 750, "time_per_iteration": 3.5352847576141357 }, { "auxiliary_loss_clip": 0.01239616, "auxiliary_loss_mlp": 0.0104251, "balance_loss_clip": 1.06859827, "balance_loss_mlp": 1.0323894, "epoch": 0.09030241086995731, "flos": 22638126742560.0, "grad_norm": 1.8709491046381719, "language_loss": 0.68065846, "learning_rate": 3.9620440134344566e-06, "loss": 0.70347977, "num_input_tokens_seen": 16054655, "step": 751, "time_per_iteration": 3.465172290802002 }, { "auxiliary_loss_clip": 0.01212778, "auxiliary_loss_mlp": 0.01046382, "balance_loss_clip": 1.06787205, "balance_loss_mlp": 1.03449106, "epoch": 0.09042265376059641, "flos": 21871505635200.0, "grad_norm": 2.332790195608918, "language_loss": 0.82312357, "learning_rate": 3.9618928248030215e-06, "loss": 0.84571522, "num_input_tokens_seen": 16074165, "step": 752, "time_per_iteration": 2.6442437171936035 }, { "auxiliary_loss_clip": 0.01243565, "auxiliary_loss_mlp": 0.01038147, "balance_loss_clip": 1.07156682, "balance_loss_mlp": 1.02661943, "epoch": 0.0905428966512355, "flos": 24316196503680.0, "grad_norm": 2.1670848442727895, "language_loss": 0.82691824, "learning_rate": 3.961741338554005e-06, "loss": 0.84973538, "num_input_tokens_seen": 16092505, "step": 753, "time_per_iteration": 2.5901458263397217 }, { "auxiliary_loss_clip": 0.01236185, "auxiliary_loss_mlp": 0.01042413, "balance_loss_clip": 1.07084107, "balance_loss_mlp": 1.03072476, "epoch": 0.09066313954187459, "flos": 35845066383840.0, "grad_norm": 1.7396269616845146, "language_loss": 0.75887275, "learning_rate": 3.9615895547103865e-06, "loss": 0.78165877, "num_input_tokens_seen": 16116150, "step": 754, "time_per_iteration": 2.732250690460205 }, { "auxiliary_loss_clip": 0.01227676, "auxiliary_loss_mlp": 0.01045535, "balance_loss_clip": 1.0677141, "balance_loss_mlp": 1.03443027, "epoch": 0.09078338243251367, "flos": 29168733828000.0, "grad_norm": 2.5567383633950698, "language_loss": 0.77987111, "learning_rate": 3.961437473295193e-06, "loss": 0.80260324, "num_input_tokens_seen": 16136295, "step": 755, "time_per_iteration": 2.6740026473999023 }, { "auxiliary_loss_clip": 0.01184344, "auxiliary_loss_mlp": 0.0104452, "balance_loss_clip": 1.05930972, "balance_loss_mlp": 1.03396988, "epoch": 0.09090362532315277, "flos": 21907703542560.0, "grad_norm": 2.337529844002445, "language_loss": 0.72667301, "learning_rate": 3.961285094331495e-06, "loss": 0.74896163, "num_input_tokens_seen": 16154210, "step": 756, "time_per_iteration": 2.650094747543335 }, { "auxiliary_loss_clip": 0.01256748, "auxiliary_loss_mlp": 0.01046529, "balance_loss_clip": 1.07032979, "balance_loss_mlp": 1.03665829, "epoch": 0.09102386821379185, "flos": 27344507592960.0, "grad_norm": 1.8374957449490106, "language_loss": 0.86061811, "learning_rate": 3.961132417842406e-06, "loss": 0.88365078, "num_input_tokens_seen": 16173995, "step": 757, "time_per_iteration": 2.6105575561523438 }, { "auxiliary_loss_clip": 0.01233117, "auxiliary_loss_mlp": 0.01040914, "balance_loss_clip": 1.06741297, "balance_loss_mlp": 1.03032207, "epoch": 0.09114411110443095, "flos": 20813505420960.0, "grad_norm": 2.4309084618631394, "language_loss": 0.75723898, "learning_rate": 3.960979443851089e-06, "loss": 0.77997923, "num_input_tokens_seen": 16191020, "step": 758, "time_per_iteration": 2.5695114135742188 }, { "auxiliary_loss_clip": 0.01220675, "auxiliary_loss_mlp": 0.01037112, "balance_loss_clip": 1.06402206, "balance_loss_mlp": 1.02553678, "epoch": 0.09126435399507005, "flos": 26145953950560.0, "grad_norm": 1.862051814067506, "language_loss": 0.7895503, "learning_rate": 3.96082617238075e-06, "loss": 0.81212819, "num_input_tokens_seen": 16213645, "step": 759, "time_per_iteration": 2.6692254543304443 }, { "auxiliary_loss_clip": 0.01228194, "auxiliary_loss_mlp": 0.01034046, "balance_loss_clip": 1.06689715, "balance_loss_mlp": 1.02357984, "epoch": 0.09138459688570913, "flos": 24388915571040.0, "grad_norm": 3.911753939030352, "language_loss": 0.79901069, "learning_rate": 3.960672603454639e-06, "loss": 0.8216331, "num_input_tokens_seen": 16233625, "step": 760, "time_per_iteration": 2.640109062194824 }, { "auxiliary_loss_clip": 0.01234739, "auxiliary_loss_mlp": 0.01045252, "balance_loss_clip": 1.06727099, "balance_loss_mlp": 1.03496981, "epoch": 0.09150483977634823, "flos": 21032668297920.0, "grad_norm": 2.694484771575343, "language_loss": 0.77114177, "learning_rate": 3.960518737096054e-06, "loss": 0.79394174, "num_input_tokens_seen": 16253255, "step": 761, "time_per_iteration": 2.5761590003967285 }, { "auxiliary_loss_clip": 0.01245232, "auxiliary_loss_mlp": 0.01041583, "balance_loss_clip": 1.0701524, "balance_loss_mlp": 1.03093171, "epoch": 0.09162508266698731, "flos": 22857002283840.0, "grad_norm": 2.432752116550217, "language_loss": 0.72908258, "learning_rate": 3.960364573328334e-06, "loss": 0.75195074, "num_input_tokens_seen": 16272580, "step": 762, "time_per_iteration": 2.6022911071777344 }, { "auxiliary_loss_clip": 0.01211952, "auxiliary_loss_mlp": 0.01039883, "balance_loss_clip": 1.06364369, "balance_loss_mlp": 1.02915454, "epoch": 0.0917453255576264, "flos": 21724415320320.0, "grad_norm": 9.814010246477697, "language_loss": 0.88668752, "learning_rate": 3.9602101121748675e-06, "loss": 0.90920579, "num_input_tokens_seen": 16293075, "step": 763, "time_per_iteration": 2.674577474594116 }, { "auxiliary_loss_clip": 0.01225799, "auxiliary_loss_mlp": 0.01047348, "balance_loss_clip": 1.06806254, "balance_loss_mlp": 1.03713799, "epoch": 0.0918655684482655, "flos": 14609218256160.0, "grad_norm": 1.8828218711429403, "language_loss": 0.72616506, "learning_rate": 3.960055353659085e-06, "loss": 0.74889654, "num_input_tokens_seen": 16310185, "step": 764, "time_per_iteration": 2.6008472442626953 }, { "auxiliary_loss_clip": 0.0121469, "auxiliary_loss_mlp": 0.01040029, "balance_loss_clip": 1.06720173, "balance_loss_mlp": 1.03032541, "epoch": 0.09198581133890459, "flos": 23435019458880.0, "grad_norm": 1.7488165424359174, "language_loss": 0.83844054, "learning_rate": 3.959900297804465e-06, "loss": 0.86098772, "num_input_tokens_seen": 16330355, "step": 765, "time_per_iteration": 2.706517219543457 }, { "auxiliary_loss_clip": 0.01206994, "auxiliary_loss_mlp": 0.01046752, "balance_loss_clip": 1.06123805, "balance_loss_mlp": 1.03650618, "epoch": 0.09210605422954368, "flos": 16795890485280.0, "grad_norm": 1.796021211138546, "language_loss": 0.77212018, "learning_rate": 3.9597449446345276e-06, "loss": 0.79465765, "num_input_tokens_seen": 16347600, "step": 766, "time_per_iteration": 2.5868942737579346 }, { "auxiliary_loss_clip": 0.01214539, "auxiliary_loss_mlp": 0.01038908, "balance_loss_clip": 1.06450033, "balance_loss_mlp": 1.02932382, "epoch": 0.09222629712018277, "flos": 22674252816000.0, "grad_norm": 2.3506884565169828, "language_loss": 0.83886874, "learning_rate": 3.95958929417284e-06, "loss": 0.86140323, "num_input_tokens_seen": 16365755, "step": 767, "time_per_iteration": 2.64499831199646 }, { "auxiliary_loss_clip": 0.01139161, "auxiliary_loss_mlp": 0.01003844, "balance_loss_clip": 1.02948177, "balance_loss_mlp": 1.00118601, "epoch": 0.09234654001082186, "flos": 69976769411520.0, "grad_norm": 0.7352891201754634, "language_loss": 0.58766925, "learning_rate": 3.9594333464430145e-06, "loss": 0.60909933, "num_input_tokens_seen": 16435245, "step": 768, "time_per_iteration": 3.299705743789673 }, { "auxiliary_loss_clip": 0.01152654, "auxiliary_loss_mlp": 0.01036939, "balance_loss_clip": 1.05676699, "balance_loss_mlp": 1.02662194, "epoch": 0.09246678290146094, "flos": 20011440662400.0, "grad_norm": 1.8070912789919542, "language_loss": 0.88227165, "learning_rate": 3.959277101468709e-06, "loss": 0.90416753, "num_input_tokens_seen": 16454795, "step": 769, "time_per_iteration": 2.761648178100586 }, { "auxiliary_loss_clip": 0.01207894, "auxiliary_loss_mlp": 0.01037137, "balance_loss_clip": 1.0626514, "balance_loss_mlp": 1.02642012, "epoch": 0.09258702579210004, "flos": 17747451995040.0, "grad_norm": 2.3553878128479298, "language_loss": 0.78791636, "learning_rate": 3.959120559273624e-06, "loss": 0.81036669, "num_input_tokens_seen": 16472580, "step": 770, "time_per_iteration": 2.610466480255127 }, { "auxiliary_loss_clip": 0.0120533, "auxiliary_loss_mlp": 0.01036415, "balance_loss_clip": 1.06151128, "balance_loss_mlp": 1.02676511, "epoch": 0.09270726868273914, "flos": 20886547740960.0, "grad_norm": 1.7545847891073714, "language_loss": 0.83497083, "learning_rate": 3.958963719881509e-06, "loss": 0.85738832, "num_input_tokens_seen": 16490670, "step": 771, "time_per_iteration": 3.3899662494659424 }, { "auxiliary_loss_clip": 0.012397, "auxiliary_loss_mlp": 0.01045123, "balance_loss_clip": 1.06995177, "balance_loss_mlp": 1.03406668, "epoch": 0.09282751157337822, "flos": 17015699867520.0, "grad_norm": 1.872604129099228, "language_loss": 0.93481416, "learning_rate": 3.958806583316154e-06, "loss": 0.9576624, "num_input_tokens_seen": 16508640, "step": 772, "time_per_iteration": 2.5507311820983887 }, { "auxiliary_loss_clip": 0.01256076, "auxiliary_loss_mlp": 0.01040844, "balance_loss_clip": 1.07039976, "balance_loss_mlp": 1.03062749, "epoch": 0.09294775446401732, "flos": 32523652173600.0, "grad_norm": 2.675851984798906, "language_loss": 0.78556442, "learning_rate": 3.9586491496013985e-06, "loss": 0.80853361, "num_input_tokens_seen": 16531035, "step": 773, "time_per_iteration": 2.65266752243042 }, { "auxiliary_loss_clip": 0.01249838, "auxiliary_loss_mlp": 0.01041631, "balance_loss_clip": 1.0731051, "balance_loss_mlp": 1.03065825, "epoch": 0.0930679973546564, "flos": 18259753909920.0, "grad_norm": 2.176928936441126, "language_loss": 0.83295858, "learning_rate": 3.958491418761124e-06, "loss": 0.85587323, "num_input_tokens_seen": 16548605, "step": 774, "time_per_iteration": 3.294247627258301 }, { "auxiliary_loss_clip": 0.0122388, "auxiliary_loss_mlp": 0.01038586, "balance_loss_clip": 1.06183767, "balance_loss_mlp": 1.02829778, "epoch": 0.0931882402452955, "flos": 21099748402560.0, "grad_norm": 2.4302365033545166, "language_loss": 0.72518325, "learning_rate": 3.958333390819258e-06, "loss": 0.74780786, "num_input_tokens_seen": 16565535, "step": 775, "time_per_iteration": 2.6381092071533203 }, { "auxiliary_loss_clip": 0.01258664, "auxiliary_loss_mlp": 0.01042935, "balance_loss_clip": 1.07363153, "balance_loss_mlp": 1.03358328, "epoch": 0.0933084831359346, "flos": 24207279528960.0, "grad_norm": 1.9608174804318805, "language_loss": 0.80320752, "learning_rate": 3.9581750657997754e-06, "loss": 0.82622349, "num_input_tokens_seen": 16584900, "step": 776, "time_per_iteration": 4.290300607681274 }, { "auxiliary_loss_clip": 0.01220892, "auxiliary_loss_mlp": 0.01037528, "balance_loss_clip": 1.06371796, "balance_loss_mlp": 1.02715099, "epoch": 0.09342872602657368, "flos": 25480276252800.0, "grad_norm": 1.7467438468769263, "language_loss": 0.89562333, "learning_rate": 3.95801644372669e-06, "loss": 0.91820759, "num_input_tokens_seen": 16604805, "step": 777, "time_per_iteration": 2.740816593170166 }, { "auxiliary_loss_clip": 0.01230058, "auxiliary_loss_mlp": 0.01044711, "balance_loss_clip": 1.06362486, "balance_loss_mlp": 1.03439307, "epoch": 0.09354896891721277, "flos": 23149063812960.0, "grad_norm": 1.9764397138018452, "language_loss": 0.84648013, "learning_rate": 3.957857524624068e-06, "loss": 0.86922783, "num_input_tokens_seen": 16623685, "step": 778, "time_per_iteration": 2.6416049003601074 }, { "auxiliary_loss_clip": 0.01223093, "auxiliary_loss_mlp": 0.01040957, "balance_loss_clip": 1.06524611, "balance_loss_mlp": 1.03091335, "epoch": 0.09366921180785186, "flos": 24279567592800.0, "grad_norm": 1.587271326130455, "language_loss": 0.89614582, "learning_rate": 3.957698308516016e-06, "loss": 0.91878629, "num_input_tokens_seen": 16644985, "step": 779, "time_per_iteration": 2.6697945594787598 }, { "auxiliary_loss_clip": 0.0123846, "auxiliary_loss_mlp": 0.00765137, "balance_loss_clip": 1.07283437, "balance_loss_mlp": 1.00025511, "epoch": 0.09378945469849095, "flos": 18730039369920.0, "grad_norm": 1.7663487663128943, "language_loss": 0.82434928, "learning_rate": 3.957538795426688e-06, "loss": 0.84438521, "num_input_tokens_seen": 16662410, "step": 780, "time_per_iteration": 2.572355031967163 }, { "auxiliary_loss_clip": 0.01225965, "auxiliary_loss_mlp": 0.01032772, "balance_loss_clip": 1.0657829, "balance_loss_mlp": 1.02191758, "epoch": 0.09390969758913004, "flos": 23218837689600.0, "grad_norm": 2.150568049475754, "language_loss": 0.77071285, "learning_rate": 3.9573789853802804e-06, "loss": 0.79330027, "num_input_tokens_seen": 16680885, "step": 781, "time_per_iteration": 2.6412341594696045 }, { "auxiliary_loss_clip": 0.01225279, "auxiliary_loss_mlp": 0.00765115, "balance_loss_clip": 1.06888008, "balance_loss_mlp": 1.00020552, "epoch": 0.09402994047976913, "flos": 19646731899840.0, "grad_norm": 3.4674002118743803, "language_loss": 0.74804193, "learning_rate": 3.957218878401037e-06, "loss": 0.76794589, "num_input_tokens_seen": 16699375, "step": 782, "time_per_iteration": 2.609994411468506 }, { "auxiliary_loss_clip": 0.01258219, "auxiliary_loss_mlp": 0.01038915, "balance_loss_clip": 1.07278228, "balance_loss_mlp": 1.0273037, "epoch": 0.09415018337040823, "flos": 29420466999360.0, "grad_norm": 1.8745032342874968, "language_loss": 0.89338177, "learning_rate": 3.957058474513246e-06, "loss": 0.91635311, "num_input_tokens_seen": 16719230, "step": 783, "time_per_iteration": 2.6332828998565674 }, { "auxiliary_loss_clip": 0.01239293, "auxiliary_loss_mlp": 0.01046234, "balance_loss_clip": 1.07203102, "balance_loss_mlp": 1.03631544, "epoch": 0.09427042626104731, "flos": 24572096042400.0, "grad_norm": 1.7887246585432066, "language_loss": 0.78786814, "learning_rate": 3.956897773741241e-06, "loss": 0.81072342, "num_input_tokens_seen": 16738220, "step": 784, "time_per_iteration": 2.6199662685394287 }, { "auxiliary_loss_clip": 0.0121418, "auxiliary_loss_mlp": 0.0103861, "balance_loss_clip": 1.06522346, "balance_loss_mlp": 1.02814937, "epoch": 0.09439066915168641, "flos": 26359585615680.0, "grad_norm": 1.8177504219201464, "language_loss": 0.7161845, "learning_rate": 3.956736776109398e-06, "loss": 0.73871237, "num_input_tokens_seen": 16759395, "step": 785, "time_per_iteration": 2.6895411014556885 }, { "auxiliary_loss_clip": 0.01230293, "auxiliary_loss_mlp": 0.00765402, "balance_loss_clip": 1.06607795, "balance_loss_mlp": 1.0002377, "epoch": 0.09451091204232549, "flos": 19427281687200.0, "grad_norm": 1.791149454318554, "language_loss": 0.8373909, "learning_rate": 3.956575481642143e-06, "loss": 0.85734773, "num_input_tokens_seen": 16778285, "step": 786, "time_per_iteration": 2.56343150138855 }, { "auxiliary_loss_clip": 0.01183597, "auxiliary_loss_mlp": 0.01037921, "balance_loss_clip": 1.05593979, "balance_loss_mlp": 1.02804446, "epoch": 0.09463115493296459, "flos": 25368054917760.0, "grad_norm": 2.605047247264696, "language_loss": 0.74559796, "learning_rate": 3.956413890363943e-06, "loss": 0.76781321, "num_input_tokens_seen": 16795265, "step": 787, "time_per_iteration": 2.7479825019836426 }, { "auxiliary_loss_clip": 0.01237569, "auxiliary_loss_mlp": 0.01043035, "balance_loss_clip": 1.06931782, "balance_loss_mlp": 1.03124535, "epoch": 0.09475139782360369, "flos": 10123257376320.0, "grad_norm": 1.99716796785436, "language_loss": 0.81768787, "learning_rate": 3.956252002299312e-06, "loss": 0.84049386, "num_input_tokens_seen": 16811165, "step": 788, "time_per_iteration": 2.5395922660827637 }, { "auxiliary_loss_clip": 0.01254274, "auxiliary_loss_mlp": 0.01038326, "balance_loss_clip": 1.07013464, "balance_loss_mlp": 1.02772212, "epoch": 0.09487164071424277, "flos": 17231091463680.0, "grad_norm": 1.8937573638683478, "language_loss": 0.90560579, "learning_rate": 3.956089817472807e-06, "loss": 0.92853189, "num_input_tokens_seen": 16828470, "step": 789, "time_per_iteration": 2.573096513748169 }, { "auxiliary_loss_clip": 0.01215254, "auxiliary_loss_mlp": 0.01044247, "balance_loss_clip": 1.06496632, "balance_loss_mlp": 1.03254604, "epoch": 0.09499188360488187, "flos": 30849569195040.0, "grad_norm": 1.9189864694947423, "language_loss": 0.85159338, "learning_rate": 3.955927335909032e-06, "loss": 0.87418842, "num_input_tokens_seen": 16851680, "step": 790, "time_per_iteration": 2.700805425643921 }, { "auxiliary_loss_clip": 0.01186098, "auxiliary_loss_mlp": 0.01040154, "balance_loss_clip": 1.06195986, "balance_loss_mlp": 1.02965808, "epoch": 0.09511212649552095, "flos": 29351698797600.0, "grad_norm": 2.2116562683086682, "language_loss": 0.7609601, "learning_rate": 3.955764557632634e-06, "loss": 0.78322256, "num_input_tokens_seen": 16871490, "step": 791, "time_per_iteration": 2.731508731842041 }, { "auxiliary_loss_clip": 0.01219362, "auxiliary_loss_mlp": 0.01039885, "balance_loss_clip": 1.06537783, "balance_loss_mlp": 1.02976382, "epoch": 0.09523236938616005, "flos": 10378689994560.0, "grad_norm": 3.9112229320398706, "language_loss": 0.95023346, "learning_rate": 3.955601482668309e-06, "loss": 0.97282594, "num_input_tokens_seen": 16889350, "step": 792, "time_per_iteration": 2.597620725631714 }, { "auxiliary_loss_clip": 0.01184268, "auxiliary_loss_mlp": 0.01042344, "balance_loss_clip": 1.05693364, "balance_loss_mlp": 1.03207374, "epoch": 0.09535261227679913, "flos": 19061818668480.0, "grad_norm": 1.9507923733417103, "language_loss": 0.88413012, "learning_rate": 3.955438111040794e-06, "loss": 0.90639627, "num_input_tokens_seen": 16907625, "step": 793, "time_per_iteration": 2.699491262435913 }, { "auxiliary_loss_clip": 0.01184501, "auxiliary_loss_mlp": 0.01033239, "balance_loss_clip": 1.05968356, "balance_loss_mlp": 1.02242625, "epoch": 0.09547285516743823, "flos": 20922997067040.0, "grad_norm": 1.7605740227428384, "language_loss": 0.80257964, "learning_rate": 3.955274442774873e-06, "loss": 0.82475698, "num_input_tokens_seen": 16926205, "step": 794, "time_per_iteration": 2.6635377407073975 }, { "auxiliary_loss_clip": 0.01235798, "auxiliary_loss_mlp": 0.01040151, "balance_loss_clip": 1.06727171, "balance_loss_mlp": 1.02927303, "epoch": 0.09559309805807732, "flos": 30154410061440.0, "grad_norm": 2.14633659009695, "language_loss": 0.70823753, "learning_rate": 3.9551104778953725e-06, "loss": 0.73099709, "num_input_tokens_seen": 16946500, "step": 795, "time_per_iteration": 2.65962815284729 }, { "auxiliary_loss_clip": 0.0120673, "auxiliary_loss_mlp": 0.01042365, "balance_loss_clip": 1.06283379, "balance_loss_mlp": 1.03212535, "epoch": 0.0957133409487164, "flos": 21066747104640.0, "grad_norm": 1.6965827336305437, "language_loss": 0.85372871, "learning_rate": 3.954946216427167e-06, "loss": 0.87621969, "num_input_tokens_seen": 16966960, "step": 796, "time_per_iteration": 2.6854734420776367 }, { "auxiliary_loss_clip": 0.01100112, "auxiliary_loss_mlp": 0.01003535, "balance_loss_clip": 1.02385688, "balance_loss_mlp": 1.00096059, "epoch": 0.0958335838393555, "flos": 71297989836000.0, "grad_norm": 0.873776757385297, "language_loss": 0.61597902, "learning_rate": 3.954781658395176e-06, "loss": 0.63701546, "num_input_tokens_seen": 17023215, "step": 797, "time_per_iteration": 3.889244794845581 }, { "auxiliary_loss_clip": 0.01227155, "auxiliary_loss_mlp": 0.01049276, "balance_loss_clip": 1.06388414, "balance_loss_mlp": 1.03819537, "epoch": 0.09595382672999458, "flos": 21872978230560.0, "grad_norm": 1.9236919001516608, "language_loss": 0.92351079, "learning_rate": 3.95461680382436e-06, "loss": 0.94627506, "num_input_tokens_seen": 17042140, "step": 798, "time_per_iteration": 2.6383681297302246 }, { "auxiliary_loss_clip": 0.01245986, "auxiliary_loss_mlp": 0.01049261, "balance_loss_clip": 1.0740726, "balance_loss_mlp": 1.03739333, "epoch": 0.09607406962063368, "flos": 18695565476640.0, "grad_norm": 2.900276760137428, "language_loss": 0.86525106, "learning_rate": 3.9544516527397295e-06, "loss": 0.88820356, "num_input_tokens_seen": 17058490, "step": 799, "time_per_iteration": 3.3062150478363037 }, { "auxiliary_loss_clip": 0.01210001, "auxiliary_loss_mlp": 0.01034215, "balance_loss_clip": 1.06367481, "balance_loss_mlp": 1.02336085, "epoch": 0.09619431251127276, "flos": 22568460616800.0, "grad_norm": 1.7997266741696125, "language_loss": 0.80738759, "learning_rate": 3.954286205166338e-06, "loss": 0.82982975, "num_input_tokens_seen": 17079655, "step": 800, "time_per_iteration": 2.679720878601074 }, { "auxiliary_loss_clip": 0.01244439, "auxiliary_loss_mlp": 0.01043782, "balance_loss_clip": 1.07221329, "balance_loss_mlp": 1.03300524, "epoch": 0.09631455540191186, "flos": 14246233507680.0, "grad_norm": 2.018195199712037, "language_loss": 0.84122062, "learning_rate": 3.954120461129282e-06, "loss": 0.86410284, "num_input_tokens_seen": 17097065, "step": 801, "time_per_iteration": 2.563901424407959 }, { "auxiliary_loss_clip": 0.01257746, "auxiliary_loss_mlp": 0.01040217, "balance_loss_clip": 1.07419491, "balance_loss_mlp": 1.0303818, "epoch": 0.09643479829255096, "flos": 20740391267040.0, "grad_norm": 1.905083109699503, "language_loss": 0.83766204, "learning_rate": 3.953954420653706e-06, "loss": 0.86064172, "num_input_tokens_seen": 17114090, "step": 802, "time_per_iteration": 4.340804100036621 }, { "auxiliary_loss_clip": 0.01240918, "auxiliary_loss_mlp": 0.01037611, "balance_loss_clip": 1.0692718, "balance_loss_mlp": 1.02641082, "epoch": 0.09655504118319004, "flos": 24420480190560.0, "grad_norm": 1.8812504214955392, "language_loss": 0.88210475, "learning_rate": 3.953788083764798e-06, "loss": 0.90489, "num_input_tokens_seen": 17133325, "step": 803, "time_per_iteration": 2.6172091960906982 }, { "auxiliary_loss_clip": 0.01193382, "auxiliary_loss_mlp": 0.01042221, "balance_loss_clip": 1.06506133, "balance_loss_mlp": 1.03140855, "epoch": 0.09667528407382914, "flos": 18441964623360.0, "grad_norm": 6.856974070981249, "language_loss": 0.92428386, "learning_rate": 3.953621450487792e-06, "loss": 0.94663978, "num_input_tokens_seen": 17151945, "step": 804, "time_per_iteration": 2.628161907196045 }, { "auxiliary_loss_clip": 0.01150233, "auxiliary_loss_mlp": 0.0100708, "balance_loss_clip": 1.0341841, "balance_loss_mlp": 1.00462425, "epoch": 0.09679552696446822, "flos": 70816468755840.0, "grad_norm": 5.041548532235385, "language_loss": 0.61291295, "learning_rate": 3.953454520847964e-06, "loss": 0.63448608, "num_input_tokens_seen": 17216790, "step": 805, "time_per_iteration": 3.297152280807495 }, { "auxiliary_loss_clip": 0.01218424, "auxiliary_loss_mlp": 0.01052296, "balance_loss_clip": 1.06510544, "balance_loss_mlp": 1.04003537, "epoch": 0.09691576985510732, "flos": 21945517713120.0, "grad_norm": 2.2463470813373005, "language_loss": 0.73095262, "learning_rate": 3.9532872948706395e-06, "loss": 0.75365984, "num_input_tokens_seen": 17236285, "step": 806, "time_per_iteration": 2.6347782611846924 }, { "auxiliary_loss_clip": 0.01224961, "auxiliary_loss_mlp": 0.01049531, "balance_loss_clip": 1.06543493, "balance_loss_mlp": 1.03738391, "epoch": 0.09703601274574641, "flos": 17965214110560.0, "grad_norm": 2.3312838307252286, "language_loss": 0.82532746, "learning_rate": 3.9531197725811845e-06, "loss": 0.84807241, "num_input_tokens_seen": 17251670, "step": 807, "time_per_iteration": 2.5696370601654053 }, { "auxiliary_loss_clip": 0.01257239, "auxiliary_loss_mlp": 0.01040082, "balance_loss_clip": 1.07538247, "balance_loss_mlp": 1.02985978, "epoch": 0.0971562556363855, "flos": 22162166402880.0, "grad_norm": 1.80867873605041, "language_loss": 0.87973273, "learning_rate": 3.952951954005013e-06, "loss": 0.90270591, "num_input_tokens_seen": 17271355, "step": 808, "time_per_iteration": 2.558849334716797 }, { "auxiliary_loss_clip": 0.01216583, "auxiliary_loss_mlp": 0.01038773, "balance_loss_clip": 1.06175315, "balance_loss_mlp": 1.02781725, "epoch": 0.0972764985270246, "flos": 25848720379200.0, "grad_norm": 1.6084194918685506, "language_loss": 0.84786308, "learning_rate": 3.952783839167584e-06, "loss": 0.87041664, "num_input_tokens_seen": 17291400, "step": 809, "time_per_iteration": 2.642679452896118 }, { "auxiliary_loss_clip": 0.01236697, "auxiliary_loss_mlp": 0.01049344, "balance_loss_clip": 1.06728053, "balance_loss_mlp": 1.03725612, "epoch": 0.09739674141766368, "flos": 20339376846240.0, "grad_norm": 3.3146322016514143, "language_loss": 0.74906147, "learning_rate": 3.952615428094398e-06, "loss": 0.77192187, "num_input_tokens_seen": 17310920, "step": 810, "time_per_iteration": 2.642301082611084 }, { "auxiliary_loss_clip": 0.01181726, "auxiliary_loss_mlp": 0.01051808, "balance_loss_clip": 1.05555403, "balance_loss_mlp": 1.04016697, "epoch": 0.09751698430830277, "flos": 15743062313280.0, "grad_norm": 2.1194606920715526, "language_loss": 0.73372781, "learning_rate": 3.952446720811004e-06, "loss": 0.75606316, "num_input_tokens_seen": 17329245, "step": 811, "time_per_iteration": 2.655493974685669 }, { "auxiliary_loss_clip": 0.01103334, "auxiliary_loss_mlp": 0.01005456, "balance_loss_clip": 1.02862787, "balance_loss_mlp": 1.00298846, "epoch": 0.09763722719894186, "flos": 63716817347040.0, "grad_norm": 0.8425430300891095, "language_loss": 0.63623053, "learning_rate": 3.952277717342995e-06, "loss": 0.65731847, "num_input_tokens_seen": 17395680, "step": 812, "time_per_iteration": 3.3294599056243896 }, { "auxiliary_loss_clip": 0.0122691, "auxiliary_loss_mlp": 0.01043718, "balance_loss_clip": 1.06595731, "balance_loss_mlp": 1.03130817, "epoch": 0.09775747008958095, "flos": 22090919930880.0, "grad_norm": 2.3397871798426433, "language_loss": 0.85656703, "learning_rate": 3.952108417716009e-06, "loss": 0.8792733, "num_input_tokens_seen": 17415135, "step": 813, "time_per_iteration": 2.6098837852478027 }, { "auxiliary_loss_clip": 0.01242468, "auxiliary_loss_mlp": 0.01047723, "balance_loss_clip": 1.07094538, "balance_loss_mlp": 1.03538465, "epoch": 0.09787771298022005, "flos": 21286053649440.0, "grad_norm": 4.791996110084267, "language_loss": 0.84791011, "learning_rate": 3.951938821955727e-06, "loss": 0.870812, "num_input_tokens_seen": 17434535, "step": 814, "time_per_iteration": 2.5890963077545166 }, { "auxiliary_loss_clip": 0.01219313, "auxiliary_loss_mlp": 0.01052252, "balance_loss_clip": 1.06494021, "balance_loss_mlp": 1.04021192, "epoch": 0.09799795587085913, "flos": 22054578355680.0, "grad_norm": 1.5883905071637856, "language_loss": 0.76439965, "learning_rate": 3.9517689300878786e-06, "loss": 0.78711534, "num_input_tokens_seen": 17454270, "step": 815, "time_per_iteration": 2.5987350940704346 }, { "auxiliary_loss_clip": 0.01256707, "auxiliary_loss_mlp": 0.01038743, "balance_loss_clip": 1.07016861, "balance_loss_mlp": 1.02809143, "epoch": 0.09811819876149823, "flos": 22163746749120.0, "grad_norm": 1.7619191095671676, "language_loss": 0.78627867, "learning_rate": 3.951598742138236e-06, "loss": 0.80923319, "num_input_tokens_seen": 17472995, "step": 816, "time_per_iteration": 2.56121563911438 }, { "auxiliary_loss_clip": 0.012271, "auxiliary_loss_mlp": 0.01039104, "balance_loss_clip": 1.06383133, "balance_loss_mlp": 1.02852416, "epoch": 0.09823844165213731, "flos": 22231114189440.0, "grad_norm": 2.4774787625276202, "language_loss": 0.80093133, "learning_rate": 3.951428258132615e-06, "loss": 0.82359344, "num_input_tokens_seen": 17491115, "step": 817, "time_per_iteration": 2.633676052093506 }, { "auxiliary_loss_clip": 0.01228666, "auxiliary_loss_mlp": 0.01037684, "balance_loss_clip": 1.068856, "balance_loss_mlp": 1.0263952, "epoch": 0.09835868454277641, "flos": 22487732067360.0, "grad_norm": 2.2168738830112673, "language_loss": 0.84764385, "learning_rate": 3.951257478096879e-06, "loss": 0.87030739, "num_input_tokens_seen": 17509480, "step": 818, "time_per_iteration": 2.6420743465423584 }, { "auxiliary_loss_clip": 0.01226682, "auxiliary_loss_mlp": 0.0076642, "balance_loss_clip": 1.06795156, "balance_loss_mlp": 1.0002054, "epoch": 0.0984789274334155, "flos": 16362557188800.0, "grad_norm": 2.7224527151693527, "language_loss": 0.68384683, "learning_rate": 3.951086402056936e-06, "loss": 0.70377785, "num_input_tokens_seen": 17524080, "step": 819, "time_per_iteration": 2.5646188259124756 }, { "auxiliary_loss_clip": 0.01155562, "auxiliary_loss_mlp": 0.00766595, "balance_loss_clip": 1.06220043, "balance_loss_mlp": 1.00022769, "epoch": 0.09859917032405459, "flos": 24243549270240.0, "grad_norm": 1.6209893521657313, "language_loss": 0.83738971, "learning_rate": 3.950915030038735e-06, "loss": 0.85661125, "num_input_tokens_seen": 17543875, "step": 820, "time_per_iteration": 2.8057687282562256 }, { "auxiliary_loss_clip": 0.01234233, "auxiliary_loss_mlp": 0.01038263, "balance_loss_clip": 1.06779933, "balance_loss_mlp": 1.02751052, "epoch": 0.09871941321469369, "flos": 17420198233440.0, "grad_norm": 2.200996272382442, "language_loss": 0.83373523, "learning_rate": 3.9507433620682765e-06, "loss": 0.85646021, "num_input_tokens_seen": 17560810, "step": 821, "time_per_iteration": 2.665475845336914 }, { "auxiliary_loss_clip": 0.01202319, "auxiliary_loss_mlp": 0.01046723, "balance_loss_clip": 1.06111443, "balance_loss_mlp": 1.03518367, "epoch": 0.09883965610533277, "flos": 28477345975200.0, "grad_norm": 1.5421940620333527, "language_loss": 0.88434213, "learning_rate": 3.9505713981716e-06, "loss": 0.90683258, "num_input_tokens_seen": 17583640, "step": 822, "time_per_iteration": 2.7087714672088623 }, { "auxiliary_loss_clip": 0.01219277, "auxiliary_loss_mlp": 0.01040066, "balance_loss_clip": 1.06695175, "balance_loss_mlp": 1.03003466, "epoch": 0.09895989899597187, "flos": 23693936022240.0, "grad_norm": 2.260393356751564, "language_loss": 0.81139243, "learning_rate": 3.950399138374795e-06, "loss": 0.83398592, "num_input_tokens_seen": 17602720, "step": 823, "time_per_iteration": 3.4117772579193115 }, { "auxiliary_loss_clip": 0.01236932, "auxiliary_loss_mlp": 0.01043959, "balance_loss_clip": 1.06737208, "balance_loss_mlp": 1.03327799, "epoch": 0.09908014188661095, "flos": 24679612255680.0, "grad_norm": 1.6199856639024015, "language_loss": 0.74128324, "learning_rate": 3.95022658270399e-06, "loss": 0.76409221, "num_input_tokens_seen": 17623085, "step": 824, "time_per_iteration": 2.6015853881835938 }, { "auxiliary_loss_clip": 0.01215423, "auxiliary_loss_mlp": 0.01041646, "balance_loss_clip": 1.06429052, "balance_loss_mlp": 1.0308634, "epoch": 0.09920038477725004, "flos": 14064310129920.0, "grad_norm": 1.9351144892511412, "language_loss": 0.78171992, "learning_rate": 3.9500537311853635e-06, "loss": 0.80429065, "num_input_tokens_seen": 17641040, "step": 825, "time_per_iteration": 3.3471908569335938 }, { "auxiliary_loss_clip": 0.01232568, "auxiliary_loss_mlp": 0.01041045, "balance_loss_clip": 1.06352329, "balance_loss_mlp": 1.02971387, "epoch": 0.09932062766788914, "flos": 13407072917760.0, "grad_norm": 2.250864798185332, "language_loss": 0.83497214, "learning_rate": 3.949880583845136e-06, "loss": 0.85770828, "num_input_tokens_seen": 17659115, "step": 826, "time_per_iteration": 2.5761120319366455 }, { "auxiliary_loss_clip": 0.01221424, "auxiliary_loss_mlp": 0.01052744, "balance_loss_clip": 1.06473923, "balance_loss_mlp": 1.04124665, "epoch": 0.09944087055852822, "flos": 19500755010720.0, "grad_norm": 1.7303089448817142, "language_loss": 0.81056762, "learning_rate": 3.949707140709575e-06, "loss": 0.83330929, "num_input_tokens_seen": 17678845, "step": 827, "time_per_iteration": 2.6395576000213623 }, { "auxiliary_loss_clip": 0.01239322, "auxiliary_loss_mlp": 0.0103679, "balance_loss_clip": 1.06528878, "balance_loss_mlp": 1.02608466, "epoch": 0.09956111344916732, "flos": 17749104175200.0, "grad_norm": 3.253087936376964, "language_loss": 0.83056134, "learning_rate": 3.949533401804991e-06, "loss": 0.85332251, "num_input_tokens_seen": 17695750, "step": 828, "time_per_iteration": 4.344284534454346 }, { "auxiliary_loss_clip": 0.01232649, "auxiliary_loss_mlp": 0.00766808, "balance_loss_clip": 1.0659616, "balance_loss_mlp": 1.00030088, "epoch": 0.0996813563398064, "flos": 17967584629920.0, "grad_norm": 2.038752141745556, "language_loss": 0.90675151, "learning_rate": 3.949359367157739e-06, "loss": 0.92674607, "num_input_tokens_seen": 17714445, "step": 829, "time_per_iteration": 2.5577025413513184 }, { "auxiliary_loss_clip": 0.01240226, "auxiliary_loss_mlp": 0.01043466, "balance_loss_clip": 1.06750822, "balance_loss_mlp": 1.03299975, "epoch": 0.0998015992304455, "flos": 17457042646080.0, "grad_norm": 2.2227250919971455, "language_loss": 0.75680405, "learning_rate": 3.949185036794222e-06, "loss": 0.77964091, "num_input_tokens_seen": 17732455, "step": 830, "time_per_iteration": 2.56009578704834 }, { "auxiliary_loss_clip": 0.01250637, "auxiliary_loss_mlp": 0.01042496, "balance_loss_clip": 1.06915784, "balance_loss_mlp": 1.03167772, "epoch": 0.0999218421210846, "flos": 25888761401280.0, "grad_norm": 1.6392841262546414, "language_loss": 0.78594804, "learning_rate": 3.949010410740884e-06, "loss": 0.80887938, "num_input_tokens_seen": 17755280, "step": 831, "time_per_iteration": 2.605551242828369 }, { "auxiliary_loss_clip": 0.01213106, "auxiliary_loss_mlp": 0.00766668, "balance_loss_clip": 1.06603742, "balance_loss_mlp": 1.00025678, "epoch": 0.10004208501172368, "flos": 21215920603200.0, "grad_norm": 1.7296216700638345, "language_loss": 0.86503214, "learning_rate": 3.948835489024216e-06, "loss": 0.88482988, "num_input_tokens_seen": 17775015, "step": 832, "time_per_iteration": 2.627446413040161 }, { "auxiliary_loss_clip": 0.01236857, "auxiliary_loss_mlp": 0.01044523, "balance_loss_clip": 1.06377769, "balance_loss_mlp": 1.03456855, "epoch": 0.10016232790236278, "flos": 17348413007040.0, "grad_norm": 1.9912188917127802, "language_loss": 0.90617895, "learning_rate": 3.948660271670755e-06, "loss": 0.92899275, "num_input_tokens_seen": 17792165, "step": 833, "time_per_iteration": 2.5568525791168213 }, { "auxiliary_loss_clip": 0.01216025, "auxiliary_loss_mlp": 0.01042781, "balance_loss_clip": 1.06365967, "balance_loss_mlp": 1.03257048, "epoch": 0.10028257079300186, "flos": 25666545582720.0, "grad_norm": 2.365054385004174, "language_loss": 0.83941293, "learning_rate": 3.948484758707079e-06, "loss": 0.862001, "num_input_tokens_seen": 17811765, "step": 834, "time_per_iteration": 2.669034719467163 }, { "auxiliary_loss_clip": 0.011941, "auxiliary_loss_mlp": 0.01037895, "balance_loss_clip": 1.05846655, "balance_loss_mlp": 1.02717257, "epoch": 0.10040281368364096, "flos": 25156039515840.0, "grad_norm": 2.0161862414361518, "language_loss": 0.83378649, "learning_rate": 3.948308950159815e-06, "loss": 0.85610652, "num_input_tokens_seen": 17830445, "step": 835, "time_per_iteration": 2.6703972816467285 }, { "auxiliary_loss_clip": 0.01198301, "auxiliary_loss_mlp": 0.01038459, "balance_loss_clip": 1.05939436, "balance_loss_mlp": 1.0266875, "epoch": 0.10052305657428004, "flos": 17603306870880.0, "grad_norm": 2.1831716687550835, "language_loss": 0.7596128, "learning_rate": 3.9481328460556326e-06, "loss": 0.7819804, "num_input_tokens_seen": 17847665, "step": 836, "time_per_iteration": 2.6271910667419434 }, { "auxiliary_loss_clip": 0.01209366, "auxiliary_loss_mlp": 0.01040749, "balance_loss_clip": 1.06243384, "balance_loss_mlp": 1.02947128, "epoch": 0.10064329946491914, "flos": 18660157742400.0, "grad_norm": 3.265935715409078, "language_loss": 0.88967204, "learning_rate": 3.9479564464212455e-06, "loss": 0.91217315, "num_input_tokens_seen": 17866825, "step": 837, "time_per_iteration": 2.6181938648223877 }, { "auxiliary_loss_clip": 0.01254643, "auxiliary_loss_mlp": 0.01039653, "balance_loss_clip": 1.06815124, "balance_loss_mlp": 1.02802384, "epoch": 0.10076354235555823, "flos": 17199059923680.0, "grad_norm": 2.2930719101120083, "language_loss": 0.76201141, "learning_rate": 3.947779751283414e-06, "loss": 0.78495437, "num_input_tokens_seen": 17883995, "step": 838, "time_per_iteration": 2.5157809257507324 }, { "auxiliary_loss_clip": 0.01239807, "auxiliary_loss_mlp": 0.00766243, "balance_loss_clip": 1.07294488, "balance_loss_mlp": 1.00020099, "epoch": 0.10088378524619732, "flos": 22962255728640.0, "grad_norm": 1.7692896142682384, "language_loss": 0.76184452, "learning_rate": 3.947602760668944e-06, "loss": 0.78190494, "num_input_tokens_seen": 17903785, "step": 839, "time_per_iteration": 2.5953593254089355 }, { "auxiliary_loss_clip": 0.01238917, "auxiliary_loss_mlp": 0.01057181, "balance_loss_clip": 1.07130384, "balance_loss_mlp": 1.04635072, "epoch": 0.10100402813683641, "flos": 37885833557760.0, "grad_norm": 1.6675727069022852, "language_loss": 0.71610761, "learning_rate": 3.947425474604684e-06, "loss": 0.73906857, "num_input_tokens_seen": 17927720, "step": 840, "time_per_iteration": 2.731945753097534 }, { "auxiliary_loss_clip": 0.01222535, "auxiliary_loss_mlp": 0.01042184, "balance_loss_clip": 1.06490386, "balance_loss_mlp": 1.03211129, "epoch": 0.1011242710274755, "flos": 21543461700480.0, "grad_norm": 1.9753177804284603, "language_loss": 0.92190397, "learning_rate": 3.947247893117528e-06, "loss": 0.94455117, "num_input_tokens_seen": 17946225, "step": 841, "time_per_iteration": 2.613039016723633 }, { "auxiliary_loss_clip": 0.01231899, "auxiliary_loss_mlp": 0.01037179, "balance_loss_clip": 1.06628752, "balance_loss_mlp": 1.0251205, "epoch": 0.10124451391811459, "flos": 13621458839040.0, "grad_norm": 17.379877468198597, "language_loss": 0.69521058, "learning_rate": 3.947070016234413e-06, "loss": 0.71790141, "num_input_tokens_seen": 17962015, "step": 842, "time_per_iteration": 2.5403358936309814 }, { "auxiliary_loss_clip": 0.01233363, "auxiliary_loss_mlp": 0.01048444, "balance_loss_clip": 1.06889939, "balance_loss_mlp": 1.03701806, "epoch": 0.10136475680875369, "flos": 16649231173920.0, "grad_norm": 2.825118580770558, "language_loss": 0.74925756, "learning_rate": 3.946891843982326e-06, "loss": 0.77207565, "num_input_tokens_seen": 17979680, "step": 843, "time_per_iteration": 2.645108938217163 }, { "auxiliary_loss_clip": 0.01238989, "auxiliary_loss_mlp": 0.01039951, "balance_loss_clip": 1.06967402, "balance_loss_mlp": 1.02820861, "epoch": 0.10148499969939277, "flos": 19461037241280.0, "grad_norm": 2.188727036451565, "language_loss": 0.7438845, "learning_rate": 3.9467133763882935e-06, "loss": 0.76667392, "num_input_tokens_seen": 17998145, "step": 844, "time_per_iteration": 2.571070671081543 }, { "auxiliary_loss_clip": 0.01226764, "auxiliary_loss_mlp": 0.01046558, "balance_loss_clip": 1.06652427, "balance_loss_mlp": 1.03543532, "epoch": 0.10160524259003187, "flos": 21104992278720.0, "grad_norm": 2.58725506387157, "language_loss": 0.86145401, "learning_rate": 3.9465346134793905e-06, "loss": 0.88418722, "num_input_tokens_seen": 18017955, "step": 845, "time_per_iteration": 2.622009038925171 }, { "auxiliary_loss_clip": 0.01203844, "auxiliary_loss_mlp": 0.01049537, "balance_loss_clip": 1.06533384, "balance_loss_mlp": 1.03899336, "epoch": 0.10172548548067095, "flos": 17712690766080.0, "grad_norm": 2.4805612302675306, "language_loss": 0.79819268, "learning_rate": 3.9463555552827335e-06, "loss": 0.82072651, "num_input_tokens_seen": 18035125, "step": 846, "time_per_iteration": 2.604830026626587 }, { "auxiliary_loss_clip": 0.01223561, "auxiliary_loss_mlp": 0.01036756, "balance_loss_clip": 1.06383669, "balance_loss_mlp": 1.0270282, "epoch": 0.10184572837131005, "flos": 21104848610880.0, "grad_norm": 2.4854992046928124, "language_loss": 0.86416018, "learning_rate": 3.946176201825487e-06, "loss": 0.88676333, "num_input_tokens_seen": 18053160, "step": 847, "time_per_iteration": 2.596400022506714 }, { "auxiliary_loss_clip": 0.0122198, "auxiliary_loss_mlp": 0.01040348, "balance_loss_clip": 1.06861377, "balance_loss_mlp": 1.0294764, "epoch": 0.10196597126194913, "flos": 26067595920480.0, "grad_norm": 1.9881777439431438, "language_loss": 0.83516961, "learning_rate": 3.9459965531348575e-06, "loss": 0.85779291, "num_input_tokens_seen": 18072815, "step": 848, "time_per_iteration": 2.6492419242858887 }, { "auxiliary_loss_clip": 0.01223313, "auxiliary_loss_mlp": 0.00765676, "balance_loss_clip": 1.0667448, "balance_loss_mlp": 1.00012815, "epoch": 0.10208621415258823, "flos": 29314638883200.0, "grad_norm": 3.03172928621201, "language_loss": 0.85141414, "learning_rate": 3.945816609238098e-06, "loss": 0.87130404, "num_input_tokens_seen": 18092225, "step": 849, "time_per_iteration": 3.440535068511963 }, { "auxiliary_loss_clip": 0.01183111, "auxiliary_loss_mlp": 0.01048247, "balance_loss_clip": 1.0605756, "balance_loss_mlp": 1.03676128, "epoch": 0.10220645704322733, "flos": 23805798187680.0, "grad_norm": 1.7832696235925676, "language_loss": 0.85567385, "learning_rate": 3.945636370162507e-06, "loss": 0.8779875, "num_input_tokens_seen": 18112335, "step": 850, "time_per_iteration": 2.73577880859375 }, { "auxiliary_loss_clip": 0.01235032, "auxiliary_loss_mlp": 0.01041356, "balance_loss_clip": 1.06769824, "balance_loss_mlp": 1.03109181, "epoch": 0.10232669993386641, "flos": 23218550353920.0, "grad_norm": 1.7063895208300819, "language_loss": 0.79108989, "learning_rate": 3.945455835935425e-06, "loss": 0.81385374, "num_input_tokens_seen": 18131520, "step": 851, "time_per_iteration": 3.375110626220703 }, { "auxiliary_loss_clip": 0.01223277, "auxiliary_loss_mlp": 0.01043305, "balance_loss_clip": 1.06662834, "balance_loss_mlp": 1.03268313, "epoch": 0.1024469428245055, "flos": 22922933045760.0, "grad_norm": 2.9809763814152137, "language_loss": 0.7526747, "learning_rate": 3.94527500658424e-06, "loss": 0.77534056, "num_input_tokens_seen": 18149185, "step": 852, "time_per_iteration": 2.6498820781707764 }, { "auxiliary_loss_clip": 0.01191703, "auxiliary_loss_mlp": 0.01038694, "balance_loss_clip": 1.06520748, "balance_loss_mlp": 1.02823377, "epoch": 0.10256718571514459, "flos": 31359500590560.0, "grad_norm": 1.658637373787579, "language_loss": 0.81034338, "learning_rate": 3.945093882136382e-06, "loss": 0.83264732, "num_input_tokens_seen": 18172960, "step": 853, "time_per_iteration": 3.6480398178100586 }, { "auxiliary_loss_clip": 0.01223426, "auxiliary_loss_mlp": 0.00765278, "balance_loss_clip": 1.06944036, "balance_loss_mlp": 1.00014365, "epoch": 0.10268742860578368, "flos": 23474880896160.0, "grad_norm": 2.072924329281072, "language_loss": 0.84441352, "learning_rate": 3.944912462619329e-06, "loss": 0.86430055, "num_input_tokens_seen": 18191925, "step": 854, "time_per_iteration": 3.4866795539855957 }, { "auxiliary_loss_clip": 0.01226669, "auxiliary_loss_mlp": 0.01046676, "balance_loss_clip": 1.06735694, "balance_loss_mlp": 1.03458786, "epoch": 0.10280767149642277, "flos": 25520317274880.0, "grad_norm": 2.2400149070584594, "language_loss": 0.80764484, "learning_rate": 3.9447307480606025e-06, "loss": 0.83037823, "num_input_tokens_seen": 18212010, "step": 855, "time_per_iteration": 2.707169771194458 }, { "auxiliary_loss_clip": 0.01214442, "auxiliary_loss_mlp": 0.01039159, "balance_loss_clip": 1.06741953, "balance_loss_mlp": 1.02847791, "epoch": 0.10292791438706186, "flos": 17347694667840.0, "grad_norm": 2.1821570272903292, "language_loss": 0.90080523, "learning_rate": 3.944548738487767e-06, "loss": 0.92334127, "num_input_tokens_seen": 18229525, "step": 856, "time_per_iteration": 2.6214683055877686 }, { "auxiliary_loss_clip": 0.01256585, "auxiliary_loss_mlp": 0.01053547, "balance_loss_clip": 1.07382572, "balance_loss_mlp": 1.04268169, "epoch": 0.10304815727770096, "flos": 27052697482560.0, "grad_norm": 1.9379360282081355, "language_loss": 0.90661979, "learning_rate": 3.944366433928434e-06, "loss": 0.92972112, "num_input_tokens_seen": 18249505, "step": 857, "time_per_iteration": 2.6017746925354004 }, { "auxiliary_loss_clip": 0.0121236, "auxiliary_loss_mlp": 0.01041461, "balance_loss_clip": 1.06339729, "balance_loss_mlp": 1.03159022, "epoch": 0.10316840016834004, "flos": 22782595119360.0, "grad_norm": 1.4588410587279683, "language_loss": 0.83536375, "learning_rate": 3.9441838344102594e-06, "loss": 0.85790193, "num_input_tokens_seen": 18269230, "step": 858, "time_per_iteration": 2.6096394062042236 }, { "auxiliary_loss_clip": 0.01227202, "auxiliary_loss_mlp": 0.01036416, "balance_loss_clip": 1.06826603, "balance_loss_mlp": 1.02553225, "epoch": 0.10328864305897914, "flos": 20704588446240.0, "grad_norm": 2.4494085532462444, "language_loss": 0.67059451, "learning_rate": 3.944000939960943e-06, "loss": 0.69323069, "num_input_tokens_seen": 18287955, "step": 859, "time_per_iteration": 2.6095762252807617 }, { "auxiliary_loss_clip": 0.01237128, "auxiliary_loss_mlp": 0.01044465, "balance_loss_clip": 1.06675601, "balance_loss_mlp": 1.03420126, "epoch": 0.10340888594961822, "flos": 28478818570560.0, "grad_norm": 1.8650282486837761, "language_loss": 0.80240095, "learning_rate": 3.943817750608229e-06, "loss": 0.82521683, "num_input_tokens_seen": 18310505, "step": 860, "time_per_iteration": 2.6509101390838623 }, { "auxiliary_loss_clip": 0.01240209, "auxiliary_loss_mlp": 0.01030513, "balance_loss_clip": 1.07141435, "balance_loss_mlp": 1.02018392, "epoch": 0.10352912884025732, "flos": 13370336256000.0, "grad_norm": 2.133704982881556, "language_loss": 0.82085145, "learning_rate": 3.943634266379908e-06, "loss": 0.84355873, "num_input_tokens_seen": 18327400, "step": 861, "time_per_iteration": 2.590475082397461 }, { "auxiliary_loss_clip": 0.01238942, "auxiliary_loss_mlp": 0.01038533, "balance_loss_clip": 1.06845474, "balance_loss_mlp": 1.02842975, "epoch": 0.10364937173089642, "flos": 25558562448960.0, "grad_norm": 1.976501356399388, "language_loss": 0.853701, "learning_rate": 3.943450487303815e-06, "loss": 0.87647581, "num_input_tokens_seen": 18347895, "step": 862, "time_per_iteration": 2.64089298248291 }, { "auxiliary_loss_clip": 0.01232161, "auxiliary_loss_mlp": 0.0104227, "balance_loss_clip": 1.06831098, "balance_loss_mlp": 1.03209591, "epoch": 0.1037696146215355, "flos": 21215489599680.0, "grad_norm": 2.2747011186010018, "language_loss": 0.85358351, "learning_rate": 3.943266413407827e-06, "loss": 0.87632775, "num_input_tokens_seen": 18367170, "step": 863, "time_per_iteration": 2.600259304046631 }, { "auxiliary_loss_clip": 0.01241505, "auxiliary_loss_mlp": 0.01045421, "balance_loss_clip": 1.06934237, "balance_loss_mlp": 1.03562188, "epoch": 0.1038898575121746, "flos": 25807386346560.0, "grad_norm": 2.2813759407131196, "language_loss": 0.84987485, "learning_rate": 3.94308204471987e-06, "loss": 0.87274408, "num_input_tokens_seen": 18386185, "step": 864, "time_per_iteration": 2.6040830612182617 }, { "auxiliary_loss_clip": 0.01203563, "auxiliary_loss_mlp": 0.01039031, "balance_loss_clip": 1.0637598, "balance_loss_mlp": 1.02902389, "epoch": 0.10401010040281368, "flos": 19062429256800.0, "grad_norm": 2.139320873001561, "language_loss": 0.74676788, "learning_rate": 3.942897381267912e-06, "loss": 0.76919377, "num_input_tokens_seen": 18402550, "step": 865, "time_per_iteration": 2.618257999420166 }, { "auxiliary_loss_clip": 0.01236686, "auxiliary_loss_mlp": 0.01042177, "balance_loss_clip": 1.06961298, "balance_loss_mlp": 1.03046489, "epoch": 0.10413034329345278, "flos": 16355122378080.0, "grad_norm": 2.4582168757993506, "language_loss": 0.65839499, "learning_rate": 3.942712423079965e-06, "loss": 0.68118358, "num_input_tokens_seen": 18418940, "step": 866, "time_per_iteration": 2.5540568828582764 }, { "auxiliary_loss_clip": 0.01185064, "auxiliary_loss_mlp": 0.01033187, "balance_loss_clip": 1.05503869, "balance_loss_mlp": 1.02343535, "epoch": 0.10425058618409186, "flos": 17236514924640.0, "grad_norm": 2.344430950550804, "language_loss": 0.89932406, "learning_rate": 3.942527170184088e-06, "loss": 0.92150652, "num_input_tokens_seen": 18435560, "step": 867, "time_per_iteration": 2.636551856994629 }, { "auxiliary_loss_clip": 0.01252525, "auxiliary_loss_mlp": 0.01041391, "balance_loss_clip": 1.07041073, "balance_loss_mlp": 1.03082895, "epoch": 0.10437082907473096, "flos": 17967369128160.0, "grad_norm": 2.1723638098596307, "language_loss": 0.77193773, "learning_rate": 3.942341622608385e-06, "loss": 0.79487687, "num_input_tokens_seen": 18452590, "step": 868, "time_per_iteration": 2.5422873497009277 }, { "auxiliary_loss_clip": 0.01223098, "auxiliary_loss_mlp": 0.01042333, "balance_loss_clip": 1.0708375, "balance_loss_mlp": 1.03167605, "epoch": 0.10449107196537005, "flos": 36283320303840.0, "grad_norm": 1.5597080979070723, "language_loss": 0.78033531, "learning_rate": 3.942155780381001e-06, "loss": 0.80298966, "num_input_tokens_seen": 18476325, "step": 869, "time_per_iteration": 2.7345569133758545 }, { "auxiliary_loss_clip": 0.01224399, "auxiliary_loss_mlp": 0.01039967, "balance_loss_clip": 1.06691813, "balance_loss_mlp": 1.02925003, "epoch": 0.10461131485600914, "flos": 23802098740800.0, "grad_norm": 1.9131710034245415, "language_loss": 0.75834703, "learning_rate": 3.94196964353013e-06, "loss": 0.78099072, "num_input_tokens_seen": 18495775, "step": 870, "time_per_iteration": 2.6333866119384766 }, { "auxiliary_loss_clip": 0.01214576, "auxiliary_loss_mlp": 0.00765401, "balance_loss_clip": 1.06347203, "balance_loss_mlp": 1.00005841, "epoch": 0.10473155774664823, "flos": 18405479380320.0, "grad_norm": 1.9638452279829646, "language_loss": 0.80848503, "learning_rate": 3.941783212084008e-06, "loss": 0.82828474, "num_input_tokens_seen": 18513530, "step": 871, "time_per_iteration": 2.5909266471862793 }, { "auxiliary_loss_clip": 0.01203707, "auxiliary_loss_mlp": 0.01032325, "balance_loss_clip": 1.06249452, "balance_loss_mlp": 1.02237701, "epoch": 0.10485180063728732, "flos": 25592641255680.0, "grad_norm": 2.5662355803628616, "language_loss": 0.78925908, "learning_rate": 3.941596486070916e-06, "loss": 0.8116194, "num_input_tokens_seen": 18531575, "step": 872, "time_per_iteration": 2.640979766845703 }, { "auxiliary_loss_clip": 0.0118064, "auxiliary_loss_mlp": 0.01045577, "balance_loss_clip": 1.06253147, "balance_loss_mlp": 1.03477097, "epoch": 0.10497204352792641, "flos": 27088751722080.0, "grad_norm": 2.2103988144511386, "language_loss": 0.58606392, "learning_rate": 3.941409465519182e-06, "loss": 0.60832608, "num_input_tokens_seen": 18552100, "step": 873, "time_per_iteration": 2.692180871963501 }, { "auxiliary_loss_clip": 0.01220627, "auxiliary_loss_mlp": 0.01032936, "balance_loss_clip": 1.06247532, "balance_loss_mlp": 1.02284503, "epoch": 0.10509228641856551, "flos": 32858484413760.0, "grad_norm": 1.7030017155054742, "language_loss": 0.85417002, "learning_rate": 3.941222150457176e-06, "loss": 0.87670565, "num_input_tokens_seen": 18575355, "step": 874, "time_per_iteration": 2.712693691253662 }, { "auxiliary_loss_clip": 0.01241953, "auxiliary_loss_mlp": 0.01029583, "balance_loss_clip": 1.06937206, "balance_loss_mlp": 1.01928878, "epoch": 0.10521252930920459, "flos": 14319168076800.0, "grad_norm": 2.344607423065796, "language_loss": 0.71190971, "learning_rate": 3.941034540913311e-06, "loss": 0.7346251, "num_input_tokens_seen": 18592885, "step": 875, "time_per_iteration": 3.336188793182373 }, { "auxiliary_loss_clip": 0.01234504, "auxiliary_loss_mlp": 0.00766483, "balance_loss_clip": 1.06715035, "balance_loss_mlp": 1.00009608, "epoch": 0.10533277219984369, "flos": 21687032153280.0, "grad_norm": 1.6615709438790682, "language_loss": 0.82479763, "learning_rate": 3.940846636916051e-06, "loss": 0.84480745, "num_input_tokens_seen": 18612920, "step": 876, "time_per_iteration": 3.387345552444458 }, { "auxiliary_loss_clip": 0.01222126, "auxiliary_loss_mlp": 0.010446, "balance_loss_clip": 1.07234263, "balance_loss_mlp": 1.03535557, "epoch": 0.10545301509048277, "flos": 22269790367040.0, "grad_norm": 1.938725415857724, "language_loss": 0.86405033, "learning_rate": 3.940658438493899e-06, "loss": 0.88671756, "num_input_tokens_seen": 18630765, "step": 877, "time_per_iteration": 2.6121037006378174 }, { "auxiliary_loss_clip": 0.0125254, "auxiliary_loss_mlp": 0.01036962, "balance_loss_clip": 1.06884861, "balance_loss_mlp": 1.02646017, "epoch": 0.10557325798112187, "flos": 22199729154720.0, "grad_norm": 2.0328067801026544, "language_loss": 0.75697041, "learning_rate": 3.940469945675405e-06, "loss": 0.77986538, "num_input_tokens_seen": 18649150, "step": 878, "time_per_iteration": 2.545165538787842 }, { "auxiliary_loss_clip": 0.01163528, "auxiliary_loss_mlp": 0.01040564, "balance_loss_clip": 1.05651975, "balance_loss_mlp": 1.03038907, "epoch": 0.10569350087176095, "flos": 25775893560960.0, "grad_norm": 2.023232886428141, "language_loss": 0.91370797, "learning_rate": 3.940281158489163e-06, "loss": 0.93574893, "num_input_tokens_seen": 18668380, "step": 879, "time_per_iteration": 2.700580358505249 }, { "auxiliary_loss_clip": 0.01166434, "auxiliary_loss_mlp": 0.01042713, "balance_loss_clip": 1.05526578, "balance_loss_mlp": 1.03265142, "epoch": 0.10581374376240005, "flos": 17311389009600.0, "grad_norm": 1.817167049379018, "language_loss": 0.82838893, "learning_rate": 3.940092076963812e-06, "loss": 0.85048032, "num_input_tokens_seen": 18685875, "step": 880, "time_per_iteration": 4.454898834228516 }, { "auxiliary_loss_clip": 0.01218735, "auxiliary_loss_mlp": 0.01043844, "balance_loss_clip": 1.06388998, "balance_loss_mlp": 1.03262079, "epoch": 0.10593398665303914, "flos": 34349458754880.0, "grad_norm": 2.093421013084938, "language_loss": 0.7874409, "learning_rate": 3.9399027011280355e-06, "loss": 0.8100667, "num_input_tokens_seen": 18707970, "step": 881, "time_per_iteration": 2.694491386413574 }, { "auxiliary_loss_clip": 0.01220591, "auxiliary_loss_mlp": 0.01040584, "balance_loss_clip": 1.06950474, "balance_loss_mlp": 1.029796, "epoch": 0.10605422954367823, "flos": 23257980787680.0, "grad_norm": 2.2688743031521454, "language_loss": 0.77078092, "learning_rate": 3.939713031010561e-06, "loss": 0.79339266, "num_input_tokens_seen": 18726335, "step": 882, "time_per_iteration": 2.613335132598877 }, { "auxiliary_loss_clip": 0.01202302, "auxiliary_loss_mlp": 0.01038426, "balance_loss_clip": 1.06492043, "balance_loss_mlp": 1.02785778, "epoch": 0.10617447243431732, "flos": 22820121954240.0, "grad_norm": 2.155681362317227, "language_loss": 0.78379339, "learning_rate": 3.939523066640163e-06, "loss": 0.80620068, "num_input_tokens_seen": 18745230, "step": 883, "time_per_iteration": 2.6236586570739746 }, { "auxiliary_loss_clip": 0.01235818, "auxiliary_loss_mlp": 0.0103923, "balance_loss_clip": 1.06944871, "balance_loss_mlp": 1.02891219, "epoch": 0.10629471532495641, "flos": 24386580968640.0, "grad_norm": 1.7766111675315786, "language_loss": 0.80980664, "learning_rate": 3.939332808045657e-06, "loss": 0.83255714, "num_input_tokens_seen": 18764880, "step": 884, "time_per_iteration": 2.610018253326416 }, { "auxiliary_loss_clip": 0.01204817, "auxiliary_loss_mlp": 0.0104132, "balance_loss_clip": 1.06695509, "balance_loss_mlp": 1.03073406, "epoch": 0.1064149582155955, "flos": 21105495116160.0, "grad_norm": 1.7481237220758572, "language_loss": 0.84615922, "learning_rate": 3.939142255255906e-06, "loss": 0.86862051, "num_input_tokens_seen": 18785765, "step": 885, "time_per_iteration": 2.6512482166290283 }, { "auxiliary_loss_clip": 0.01239153, "auxiliary_loss_mlp": 0.01036043, "balance_loss_clip": 1.07242799, "balance_loss_mlp": 1.02559412, "epoch": 0.1065352011062346, "flos": 20702038342080.0, "grad_norm": 2.014012714065302, "language_loss": 0.86682022, "learning_rate": 3.938951408299817e-06, "loss": 0.88957214, "num_input_tokens_seen": 18804605, "step": 886, "time_per_iteration": 2.597073793411255 }, { "auxiliary_loss_clip": 0.01096843, "auxiliary_loss_mlp": 0.01003133, "balance_loss_clip": 1.03991807, "balance_loss_mlp": 1.00085604, "epoch": 0.10665544399687368, "flos": 62659643222880.0, "grad_norm": 0.7970268950274886, "language_loss": 0.544981, "learning_rate": 3.938760267206342e-06, "loss": 0.56598079, "num_input_tokens_seen": 18866425, "step": 887, "time_per_iteration": 3.233159303665161 }, { "auxiliary_loss_clip": 0.01252044, "auxiliary_loss_mlp": 0.01037621, "balance_loss_clip": 1.07289124, "balance_loss_mlp": 1.0275358, "epoch": 0.10677568688751278, "flos": 26140386821760.0, "grad_norm": 2.086335383265417, "language_loss": 0.79156369, "learning_rate": 3.938568832004475e-06, "loss": 0.81446034, "num_input_tokens_seen": 18885130, "step": 888, "time_per_iteration": 2.784860610961914 }, { "auxiliary_loss_clip": 0.01208213, "auxiliary_loss_mlp": 0.01038547, "balance_loss_clip": 1.06323624, "balance_loss_mlp": 1.02844954, "epoch": 0.10689592977815186, "flos": 12786536450400.0, "grad_norm": 2.0254136112837813, "language_loss": 0.75517589, "learning_rate": 3.938377102723257e-06, "loss": 0.77764344, "num_input_tokens_seen": 18902265, "step": 889, "time_per_iteration": 2.582963466644287 }, { "auxiliary_loss_clip": 0.01173059, "auxiliary_loss_mlp": 0.01037107, "balance_loss_clip": 1.05782747, "balance_loss_mlp": 1.02445865, "epoch": 0.10701617266879096, "flos": 22126686834720.0, "grad_norm": 1.9631319656726067, "language_loss": 0.83458799, "learning_rate": 3.938185079391774e-06, "loss": 0.85668963, "num_input_tokens_seen": 18919310, "step": 890, "time_per_iteration": 2.7366526126861572 }, { "auxiliary_loss_clip": 0.012515, "auxiliary_loss_mlp": 0.01049529, "balance_loss_clip": 1.07054818, "balance_loss_mlp": 1.03918147, "epoch": 0.10713641555943004, "flos": 19745628042720.0, "grad_norm": 2.330587560752462, "language_loss": 1.06176841, "learning_rate": 3.937992762039157e-06, "loss": 1.08477879, "num_input_tokens_seen": 18932635, "step": 891, "time_per_iteration": 2.5067639350891113 }, { "auxiliary_loss_clip": 0.01237457, "auxiliary_loss_mlp": 0.01038186, "balance_loss_clip": 1.07190549, "balance_loss_mlp": 1.02783871, "epoch": 0.10725665845006914, "flos": 23952996253440.0, "grad_norm": 1.6989353536100822, "language_loss": 0.8057462, "learning_rate": 3.937800150694577e-06, "loss": 0.8285026, "num_input_tokens_seen": 18953810, "step": 892, "time_per_iteration": 2.617762804031372 }, { "auxiliary_loss_clip": 0.0119166, "auxiliary_loss_mlp": 0.01036457, "balance_loss_clip": 1.06654406, "balance_loss_mlp": 1.0255909, "epoch": 0.10737690134070824, "flos": 18551707688160.0, "grad_norm": 2.0530732172914754, "language_loss": 0.76402271, "learning_rate": 3.937607245387255e-06, "loss": 0.78630388, "num_input_tokens_seen": 18973175, "step": 893, "time_per_iteration": 2.6895763874053955 }, { "auxiliary_loss_clip": 0.01224846, "auxiliary_loss_mlp": 0.01046077, "balance_loss_clip": 1.06411576, "balance_loss_mlp": 1.03619504, "epoch": 0.10749714423134732, "flos": 22707613283520.0, "grad_norm": 1.892180004780693, "language_loss": 0.71826488, "learning_rate": 3.937414046146455e-06, "loss": 0.74097419, "num_input_tokens_seen": 18991130, "step": 894, "time_per_iteration": 2.6081061363220215 }, { "auxiliary_loss_clip": 0.01255132, "auxiliary_loss_mlp": 0.01047957, "balance_loss_clip": 1.07425785, "balance_loss_mlp": 1.03639984, "epoch": 0.10761738712198642, "flos": 21106069787520.0, "grad_norm": 2.119482028896176, "language_loss": 0.75708294, "learning_rate": 3.9372205530014845e-06, "loss": 0.78011382, "num_input_tokens_seen": 19009610, "step": 895, "time_per_iteration": 2.567944049835205 }, { "auxiliary_loss_clip": 0.01252737, "auxiliary_loss_mlp": 0.01037533, "balance_loss_clip": 1.07123256, "balance_loss_mlp": 1.02723956, "epoch": 0.1077376300126255, "flos": 23766726923520.0, "grad_norm": 2.1786945076442876, "language_loss": 0.71392936, "learning_rate": 3.937026765981696e-06, "loss": 0.73683202, "num_input_tokens_seen": 19029680, "step": 896, "time_per_iteration": 2.5604074001312256 }, { "auxiliary_loss_clip": 0.01209893, "auxiliary_loss_mlp": 0.01055339, "balance_loss_clip": 1.06933105, "balance_loss_mlp": 1.04407918, "epoch": 0.1078578729032646, "flos": 20919584955840.0, "grad_norm": 1.825190393671226, "language_loss": 0.79717886, "learning_rate": 3.936832685116488e-06, "loss": 0.81983125, "num_input_tokens_seen": 19047775, "step": 897, "time_per_iteration": 2.6395561695098877 }, { "auxiliary_loss_clip": 0.01254973, "auxiliary_loss_mlp": 0.01040724, "balance_loss_clip": 1.07357454, "balance_loss_mlp": 1.02996588, "epoch": 0.10797811579390369, "flos": 14829889645440.0, "grad_norm": 2.3504575801271548, "language_loss": 0.90048021, "learning_rate": 3.936638310435301e-06, "loss": 0.92343712, "num_input_tokens_seen": 19065640, "step": 898, "time_per_iteration": 2.528818368911743 }, { "auxiliary_loss_clip": 0.01243054, "auxiliary_loss_mlp": 0.01041955, "balance_loss_clip": 1.07178783, "balance_loss_mlp": 1.03026688, "epoch": 0.10809835868454278, "flos": 19536988835040.0, "grad_norm": 2.3563478241127265, "language_loss": 0.81137282, "learning_rate": 3.936443641967623e-06, "loss": 0.83422291, "num_input_tokens_seen": 19084470, "step": 899, "time_per_iteration": 2.5874667167663574 }, { "auxiliary_loss_clip": 0.01224666, "auxiliary_loss_mlp": 0.01047673, "balance_loss_clip": 1.07002735, "balance_loss_mlp": 1.03703344, "epoch": 0.10821860157518187, "flos": 18442323792960.0, "grad_norm": 2.135021348188451, "language_loss": 0.83436161, "learning_rate": 3.936248679742983e-06, "loss": 0.85708499, "num_input_tokens_seen": 19102965, "step": 900, "time_per_iteration": 2.6015784740448 }, { "auxiliary_loss_clip": 0.01098624, "auxiliary_loss_mlp": 0.01010601, "balance_loss_clip": 1.02931011, "balance_loss_mlp": 1.00816941, "epoch": 0.10833884446582095, "flos": 49359475930080.0, "grad_norm": 1.0572483683804346, "language_loss": 0.7019676, "learning_rate": 3.936053423790959e-06, "loss": 0.72305977, "num_input_tokens_seen": 19151285, "step": 901, "time_per_iteration": 3.7035372257232666 }, { "auxiliary_loss_clip": 0.01255096, "auxiliary_loss_mlp": 0.01037877, "balance_loss_clip": 1.07522476, "balance_loss_mlp": 1.02767277, "epoch": 0.10845908735646005, "flos": 20411916328800.0, "grad_norm": 1.663549613225642, "language_loss": 0.77371043, "learning_rate": 3.935857874141168e-06, "loss": 0.79664016, "num_input_tokens_seen": 19170120, "step": 902, "time_per_iteration": 3.2826485633850098 }, { "auxiliary_loss_clip": 0.01217396, "auxiliary_loss_mlp": 0.0104007, "balance_loss_clip": 1.06985116, "balance_loss_mlp": 1.02878046, "epoch": 0.10857933024709913, "flos": 14027753052960.0, "grad_norm": 2.3331066941828413, "language_loss": 0.83815461, "learning_rate": 3.935662030823279e-06, "loss": 0.86072928, "num_input_tokens_seen": 19186305, "step": 903, "time_per_iteration": 2.5736825466156006 }, { "auxiliary_loss_clip": 0.01241131, "auxiliary_loss_mlp": 0.01040761, "balance_loss_clip": 1.0713172, "balance_loss_mlp": 1.03033018, "epoch": 0.10869957313773823, "flos": 13369007328480.0, "grad_norm": 2.5320055349811943, "language_loss": 0.72729582, "learning_rate": 3.935465893866998e-06, "loss": 0.75011474, "num_input_tokens_seen": 19204530, "step": 904, "time_per_iteration": 2.576080322265625 }, { "auxiliary_loss_clip": 0.01221902, "auxiliary_loss_mlp": 0.01043063, "balance_loss_clip": 1.07001317, "balance_loss_mlp": 1.03296638, "epoch": 0.10881981602837733, "flos": 25807099010880.0, "grad_norm": 2.207541795974762, "language_loss": 0.80079925, "learning_rate": 3.935269463302079e-06, "loss": 0.8234489, "num_input_tokens_seen": 19222735, "step": 905, "time_per_iteration": 3.5053229331970215 }, { "auxiliary_loss_clip": 0.01245474, "auxiliary_loss_mlp": 0.01042736, "balance_loss_clip": 1.0730288, "balance_loss_mlp": 1.03142273, "epoch": 0.10894005891901641, "flos": 20777558932320.0, "grad_norm": 1.637791152761445, "language_loss": 0.76773405, "learning_rate": 3.935072739158322e-06, "loss": 0.79061615, "num_input_tokens_seen": 19242445, "step": 906, "time_per_iteration": 3.4795081615448 }, { "auxiliary_loss_clip": 0.01227434, "auxiliary_loss_mlp": 0.01036519, "balance_loss_clip": 1.07040143, "balance_loss_mlp": 1.02671969, "epoch": 0.10906030180965551, "flos": 26649887213760.0, "grad_norm": 1.740684205168762, "language_loss": 0.79983139, "learning_rate": 3.934875721465569e-06, "loss": 0.8224709, "num_input_tokens_seen": 19262865, "step": 907, "time_per_iteration": 2.6693809032440186 }, { "auxiliary_loss_clip": 0.01217421, "auxiliary_loss_mlp": 0.01039243, "balance_loss_clip": 1.06601596, "balance_loss_mlp": 1.02796614, "epoch": 0.10918054470029459, "flos": 36534406969920.0, "grad_norm": 3.1547919759412286, "language_loss": 0.71986908, "learning_rate": 3.9346784102537076e-06, "loss": 0.74243569, "num_input_tokens_seen": 19285000, "step": 908, "time_per_iteration": 2.7307016849517822 }, { "auxiliary_loss_clip": 0.01249919, "auxiliary_loss_mlp": 0.01040511, "balance_loss_clip": 1.06966925, "balance_loss_mlp": 1.03013945, "epoch": 0.10930078759093369, "flos": 21762552743520.0, "grad_norm": 1.7409971757709946, "language_loss": 0.7840966, "learning_rate": 3.934480805552669e-06, "loss": 0.80700094, "num_input_tokens_seen": 19306010, "step": 909, "time_per_iteration": 2.5724833011627197 }, { "auxiliary_loss_clip": 0.01248804, "auxiliary_loss_mlp": 0.00765778, "balance_loss_clip": 1.06940043, "balance_loss_mlp": 1.00012767, "epoch": 0.10942103048157277, "flos": 22601785167360.0, "grad_norm": 1.981446083760712, "language_loss": 0.88205564, "learning_rate": 3.93428290739243e-06, "loss": 0.90220141, "num_input_tokens_seen": 19325380, "step": 910, "time_per_iteration": 2.5584378242492676 }, { "auxiliary_loss_clip": 0.01224221, "auxiliary_loss_mlp": 0.01037465, "balance_loss_clip": 1.06911242, "balance_loss_mlp": 1.02684355, "epoch": 0.10954127337221187, "flos": 15045784079040.0, "grad_norm": 2.179656106099128, "language_loss": 0.79735005, "learning_rate": 3.9340847158030125e-06, "loss": 0.81996697, "num_input_tokens_seen": 19338960, "step": 911, "time_per_iteration": 2.641862392425537 }, { "auxiliary_loss_clip": 0.01237343, "auxiliary_loss_mlp": 0.01036506, "balance_loss_clip": 1.06840694, "balance_loss_mlp": 1.02646828, "epoch": 0.10966151626285096, "flos": 21650977913760.0, "grad_norm": 1.9349605279667441, "language_loss": 0.75317872, "learning_rate": 3.9338862308144814e-06, "loss": 0.77591717, "num_input_tokens_seen": 19357780, "step": 912, "time_per_iteration": 2.6174371242523193 }, { "auxiliary_loss_clip": 0.01250509, "auxiliary_loss_mlp": 0.01041285, "balance_loss_clip": 1.07233429, "balance_loss_mlp": 1.0313549, "epoch": 0.10978175915349005, "flos": 20121363312000.0, "grad_norm": 1.6403361133402552, "language_loss": 0.84731752, "learning_rate": 3.933687452456946e-06, "loss": 0.87023544, "num_input_tokens_seen": 19377680, "step": 913, "time_per_iteration": 2.5718891620635986 }, { "auxiliary_loss_clip": 0.01202877, "auxiliary_loss_mlp": 0.01042297, "balance_loss_clip": 1.06330538, "balance_loss_mlp": 1.03066206, "epoch": 0.10990200204412914, "flos": 20412670584960.0, "grad_norm": 2.869027808925682, "language_loss": 0.86366785, "learning_rate": 3.933488380760562e-06, "loss": 0.8861196, "num_input_tokens_seen": 19397040, "step": 914, "time_per_iteration": 2.636831283569336 }, { "auxiliary_loss_clip": 0.01250566, "auxiliary_loss_mlp": 0.00766043, "balance_loss_clip": 1.07050014, "balance_loss_mlp": 1.00013566, "epoch": 0.11002224493476823, "flos": 17530120883040.0, "grad_norm": 2.0619088909520866, "language_loss": 0.87483805, "learning_rate": 3.9332890157555286e-06, "loss": 0.89500415, "num_input_tokens_seen": 19413975, "step": 915, "time_per_iteration": 2.5368924140930176 }, { "auxiliary_loss_clip": 0.01228315, "auxiliary_loss_mlp": 0.01049387, "balance_loss_clip": 1.06924176, "balance_loss_mlp": 1.03853869, "epoch": 0.11014248782540732, "flos": 12203095814400.0, "grad_norm": 2.482390985637965, "language_loss": 0.76116818, "learning_rate": 3.933089357472088e-06, "loss": 0.7839452, "num_input_tokens_seen": 19432005, "step": 916, "time_per_iteration": 2.6114792823791504 }, { "auxiliary_loss_clip": 0.01252727, "auxiliary_loss_mlp": 0.01037137, "balance_loss_clip": 1.07417178, "balance_loss_mlp": 1.02659917, "epoch": 0.11026273071604642, "flos": 22382981460000.0, "grad_norm": 1.8194093214523412, "language_loss": 0.85633665, "learning_rate": 3.932889405940529e-06, "loss": 0.87923533, "num_input_tokens_seen": 19450100, "step": 917, "time_per_iteration": 2.5730907917022705 }, { "auxiliary_loss_clip": 0.01221664, "auxiliary_loss_mlp": 0.01040864, "balance_loss_clip": 1.0696975, "balance_loss_mlp": 1.02967072, "epoch": 0.1103829736066855, "flos": 19829050364160.0, "grad_norm": 2.1365093515066222, "language_loss": 0.80546343, "learning_rate": 3.932689161191184e-06, "loss": 0.82808876, "num_input_tokens_seen": 19467805, "step": 918, "time_per_iteration": 2.593223810195923 }, { "auxiliary_loss_clip": 0.01232569, "auxiliary_loss_mlp": 0.0103971, "balance_loss_clip": 1.06579888, "balance_loss_mlp": 1.02888608, "epoch": 0.1105032164973246, "flos": 22669619528160.0, "grad_norm": 2.0326483801419517, "language_loss": 0.88247025, "learning_rate": 3.93248862325443e-06, "loss": 0.90519309, "num_input_tokens_seen": 19486710, "step": 919, "time_per_iteration": 2.5968987941741943 }, { "auxiliary_loss_clip": 0.01122325, "auxiliary_loss_mlp": 0.01006318, "balance_loss_clip": 1.02477181, "balance_loss_mlp": 1.00387371, "epoch": 0.11062345938796368, "flos": 66483518100960.0, "grad_norm": 0.9411012702810337, "language_loss": 0.64540964, "learning_rate": 3.932287792160688e-06, "loss": 0.66669607, "num_input_tokens_seen": 19545170, "step": 920, "time_per_iteration": 3.0609219074249268 }, { "auxiliary_loss_clip": 0.01240322, "auxiliary_loss_mlp": 0.01044048, "balance_loss_clip": 1.06773734, "balance_loss_mlp": 1.03238916, "epoch": 0.11074370227860278, "flos": 21907775376480.0, "grad_norm": 2.320591907785837, "language_loss": 0.8068254, "learning_rate": 3.932086667940424e-06, "loss": 0.82966912, "num_input_tokens_seen": 19561875, "step": 921, "time_per_iteration": 2.600083589553833 }, { "auxiliary_loss_clip": 0.01233752, "auxiliary_loss_mlp": 0.00765343, "balance_loss_clip": 1.07057667, "balance_loss_mlp": 1.0001514, "epoch": 0.11086394516924186, "flos": 28658120010240.0, "grad_norm": 1.7887776900440606, "language_loss": 0.81656957, "learning_rate": 3.93188525062415e-06, "loss": 0.83656055, "num_input_tokens_seen": 19582340, "step": 922, "time_per_iteration": 2.6622154712677 }, { "auxiliary_loss_clip": 0.01237334, "auxiliary_loss_mlp": 0.01043038, "balance_loss_clip": 1.0690434, "balance_loss_mlp": 1.03177261, "epoch": 0.11098418805988096, "flos": 24535251629760.0, "grad_norm": 2.055603585206955, "language_loss": 0.86017549, "learning_rate": 3.931683540242418e-06, "loss": 0.88297927, "num_input_tokens_seen": 19603405, "step": 923, "time_per_iteration": 2.636016368865967 }, { "auxiliary_loss_clip": 0.01228293, "auxiliary_loss_mlp": 0.01042062, "balance_loss_clip": 1.06535745, "balance_loss_mlp": 1.03100586, "epoch": 0.11110443095052006, "flos": 22960388046720.0, "grad_norm": 2.7719468091471913, "language_loss": 0.90873849, "learning_rate": 3.9314815368258295e-06, "loss": 0.93144202, "num_input_tokens_seen": 19619885, "step": 924, "time_per_iteration": 2.636169910430908 }, { "auxiliary_loss_clip": 0.01240051, "auxiliary_loss_mlp": 0.01038543, "balance_loss_clip": 1.07111931, "balance_loss_mlp": 1.02812362, "epoch": 0.11122467384115914, "flos": 18950028336960.0, "grad_norm": 1.6863999680790498, "language_loss": 0.78893286, "learning_rate": 3.9312792404050275e-06, "loss": 0.81171876, "num_input_tokens_seen": 19637940, "step": 925, "time_per_iteration": 2.625399351119995 }, { "auxiliary_loss_clip": 0.01245682, "auxiliary_loss_mlp": 0.01035784, "balance_loss_clip": 1.06956959, "balance_loss_mlp": 1.02653933, "epoch": 0.11134491673179824, "flos": 25082961278880.0, "grad_norm": 1.7232145246491963, "language_loss": 0.77404535, "learning_rate": 3.9310766510107e-06, "loss": 0.79685998, "num_input_tokens_seen": 19657115, "step": 926, "time_per_iteration": 2.5951454639434814 }, { "auxiliary_loss_clip": 0.01209188, "auxiliary_loss_mlp": 0.01035088, "balance_loss_clip": 1.0642997, "balance_loss_mlp": 1.02353072, "epoch": 0.11146515962243732, "flos": 24499125556320.0, "grad_norm": 1.79973908875984, "language_loss": 0.92084306, "learning_rate": 3.9308737686735806e-06, "loss": 0.94328582, "num_input_tokens_seen": 19677075, "step": 927, "time_per_iteration": 3.3919551372528076 }, { "auxiliary_loss_clip": 0.01252173, "auxiliary_loss_mlp": 0.01039168, "balance_loss_clip": 1.07055736, "balance_loss_mlp": 1.0289638, "epoch": 0.11158540251307641, "flos": 22343766528000.0, "grad_norm": 2.019395680297531, "language_loss": 0.82812297, "learning_rate": 3.9306705934244455e-06, "loss": 0.85103643, "num_input_tokens_seen": 19697155, "step": 928, "time_per_iteration": 3.292135238647461 }, { "auxiliary_loss_clip": 0.0120587, "auxiliary_loss_mlp": 0.0103944, "balance_loss_clip": 1.06410408, "balance_loss_mlp": 1.02967131, "epoch": 0.11170564540371551, "flos": 19902308185920.0, "grad_norm": 1.6258517460728352, "language_loss": 0.882061, "learning_rate": 3.930467125294116e-06, "loss": 0.90451407, "num_input_tokens_seen": 19716705, "step": 929, "time_per_iteration": 2.6237895488739014 }, { "auxiliary_loss_clip": 0.01066651, "auxiliary_loss_mlp": 0.0100818, "balance_loss_clip": 1.02064276, "balance_loss_mlp": 1.00572479, "epoch": 0.1118258882943546, "flos": 64586249545920.0, "grad_norm": 0.9300433403861943, "language_loss": 0.60437286, "learning_rate": 3.930263364313458e-06, "loss": 0.62512112, "num_input_tokens_seen": 19767275, "step": 930, "time_per_iteration": 3.172553300857544 }, { "auxiliary_loss_clip": 0.0120449, "auxiliary_loss_mlp": 0.01046851, "balance_loss_clip": 1.06372452, "balance_loss_mlp": 1.03445959, "epoch": 0.11194613118499369, "flos": 17201969197440.0, "grad_norm": 2.2822841635108104, "language_loss": 0.82753849, "learning_rate": 3.930059310513384e-06, "loss": 0.85005188, "num_input_tokens_seen": 19786315, "step": 931, "time_per_iteration": 3.6042213439941406 }, { "auxiliary_loss_clip": 0.01187102, "auxiliary_loss_mlp": 0.00765587, "balance_loss_clip": 1.05767465, "balance_loss_mlp": 1.0002104, "epoch": 0.11206637407563277, "flos": 31863469770720.0, "grad_norm": 1.8187221784596128, "language_loss": 0.83855122, "learning_rate": 3.929854963924846e-06, "loss": 0.85807812, "num_input_tokens_seen": 19806580, "step": 932, "time_per_iteration": 3.6244394779205322 }, { "auxiliary_loss_clip": 0.01199907, "auxiliary_loss_mlp": 0.01040697, "balance_loss_clip": 1.05893683, "balance_loss_mlp": 1.03065324, "epoch": 0.11218661696627187, "flos": 21945625464000.0, "grad_norm": 1.813103066083334, "language_loss": 0.77426708, "learning_rate": 3.929650324578845e-06, "loss": 0.79667312, "num_input_tokens_seen": 19826045, "step": 933, "time_per_iteration": 2.64279842376709 }, { "auxiliary_loss_clip": 0.01224716, "auxiliary_loss_mlp": 0.01045037, "balance_loss_clip": 1.06750584, "balance_loss_mlp": 1.03359926, "epoch": 0.11230685985691095, "flos": 25878201815040.0, "grad_norm": 2.50351689872848, "language_loss": 0.81885326, "learning_rate": 3.929445392506423e-06, "loss": 0.84155071, "num_input_tokens_seen": 19843985, "step": 934, "time_per_iteration": 2.64932918548584 }, { "auxiliary_loss_clip": 0.01233696, "auxiliary_loss_mlp": 0.01043114, "balance_loss_clip": 1.0710578, "balance_loss_mlp": 1.03248, "epoch": 0.11242710274755005, "flos": 22231509276000.0, "grad_norm": 1.8987384153665845, "language_loss": 0.75843006, "learning_rate": 3.92924016773867e-06, "loss": 0.78119814, "num_input_tokens_seen": 19860480, "step": 935, "time_per_iteration": 2.5621440410614014 }, { "auxiliary_loss_clip": 0.01217223, "auxiliary_loss_mlp": 0.00766195, "balance_loss_clip": 1.06329703, "balance_loss_mlp": 1.00015688, "epoch": 0.11254734563818915, "flos": 17712187928640.0, "grad_norm": 2.1634363572283704, "language_loss": 0.7356981, "learning_rate": 3.9290346503067175e-06, "loss": 0.75553226, "num_input_tokens_seen": 19877145, "step": 936, "time_per_iteration": 2.6251070499420166 }, { "auxiliary_loss_clip": 0.01238199, "auxiliary_loss_mlp": 0.01037712, "balance_loss_clip": 1.06655097, "balance_loss_mlp": 1.02671552, "epoch": 0.11266758852882823, "flos": 54930404272800.0, "grad_norm": 1.9252312054764007, "language_loss": 0.78940868, "learning_rate": 3.9288288402417415e-06, "loss": 0.81216776, "num_input_tokens_seen": 19903405, "step": 937, "time_per_iteration": 2.893050193786621 }, { "auxiliary_loss_clip": 0.0123611, "auxiliary_loss_mlp": 0.010412, "balance_loss_clip": 1.06799912, "balance_loss_mlp": 1.02974439, "epoch": 0.11278783141946733, "flos": 18878135359680.0, "grad_norm": 2.3498809419896958, "language_loss": 0.70430249, "learning_rate": 3.928622737574964e-06, "loss": 0.72707564, "num_input_tokens_seen": 19918740, "step": 938, "time_per_iteration": 2.5680646896362305 }, { "auxiliary_loss_clip": 0.01215422, "auxiliary_loss_mlp": 0.01037463, "balance_loss_clip": 1.06264663, "balance_loss_mlp": 1.02757478, "epoch": 0.11290807431010641, "flos": 26469256846560.0, "grad_norm": 1.9466043216341948, "language_loss": 0.91106123, "learning_rate": 3.928416342337652e-06, "loss": 0.93359005, "num_input_tokens_seen": 19938475, "step": 939, "time_per_iteration": 2.628816843032837 }, { "auxiliary_loss_clip": 0.01221757, "auxiliary_loss_mlp": 0.01045611, "balance_loss_clip": 1.06640315, "balance_loss_mlp": 1.0352875, "epoch": 0.1130283172007455, "flos": 22710594391200.0, "grad_norm": 1.7425286414222505, "language_loss": 0.82621866, "learning_rate": 3.928209654561113e-06, "loss": 0.84889233, "num_input_tokens_seen": 19959310, "step": 940, "time_per_iteration": 2.6317458152770996 }, { "auxiliary_loss_clip": 0.01211168, "auxiliary_loss_mlp": 0.01045164, "balance_loss_clip": 1.06389439, "balance_loss_mlp": 1.03499579, "epoch": 0.1131485600913846, "flos": 23219915198400.0, "grad_norm": 1.943662286398371, "language_loss": 0.81610584, "learning_rate": 3.928002674276703e-06, "loss": 0.83866918, "num_input_tokens_seen": 19978700, "step": 941, "time_per_iteration": 2.604008436203003 }, { "auxiliary_loss_clip": 0.01163565, "auxiliary_loss_mlp": 0.01033082, "balance_loss_clip": 1.05190396, "balance_loss_mlp": 1.02210832, "epoch": 0.11326880298202369, "flos": 14064274212960.0, "grad_norm": 2.0053553106444713, "language_loss": 0.75369084, "learning_rate": 3.92779540151582e-06, "loss": 0.7756573, "num_input_tokens_seen": 19995785, "step": 942, "time_per_iteration": 2.6077256202697754 }, { "auxiliary_loss_clip": 0.01218962, "auxiliary_loss_mlp": 0.01033212, "balance_loss_clip": 1.06651926, "balance_loss_mlp": 1.02262592, "epoch": 0.11338904587266278, "flos": 16325389523520.0, "grad_norm": 1.857698307221872, "language_loss": 0.86048025, "learning_rate": 3.927587836309907e-06, "loss": 0.88300198, "num_input_tokens_seen": 20013615, "step": 943, "time_per_iteration": 2.5497775077819824 }, { "auxiliary_loss_clip": 0.01209991, "auxiliary_loss_mlp": 0.01037769, "balance_loss_clip": 1.06117034, "balance_loss_mlp": 1.02684367, "epoch": 0.11350928876330187, "flos": 24426262821120.0, "grad_norm": 1.823730009635493, "language_loss": 0.78032029, "learning_rate": 3.927379978690452e-06, "loss": 0.80279779, "num_input_tokens_seen": 20032880, "step": 944, "time_per_iteration": 2.625469923019409 }, { "auxiliary_loss_clip": 0.01190007, "auxiliary_loss_mlp": 0.01042685, "balance_loss_clip": 1.05499148, "balance_loss_mlp": 1.03151536, "epoch": 0.11362953165394096, "flos": 24497078289600.0, "grad_norm": 1.9140979823887125, "language_loss": 0.87614381, "learning_rate": 3.927171828688987e-06, "loss": 0.89847076, "num_input_tokens_seen": 20052405, "step": 945, "time_per_iteration": 2.6448943614959717 }, { "auxiliary_loss_clip": 0.01249416, "auxiliary_loss_mlp": 0.01042021, "balance_loss_clip": 1.06861496, "balance_loss_mlp": 1.03139353, "epoch": 0.11374977454458005, "flos": 24060835719360.0, "grad_norm": 1.896350719265469, "language_loss": 0.82016528, "learning_rate": 3.926963386337088e-06, "loss": 0.84307969, "num_input_tokens_seen": 20070635, "step": 946, "time_per_iteration": 2.5786471366882324 }, { "auxiliary_loss_clip": 0.01253516, "auxiliary_loss_mlp": 0.01037919, "balance_loss_clip": 1.07060742, "balance_loss_mlp": 1.02609921, "epoch": 0.11387001743521914, "flos": 39457644199200.0, "grad_norm": 2.1688236677502304, "language_loss": 0.69920087, "learning_rate": 3.926754651666375e-06, "loss": 0.72211522, "num_input_tokens_seen": 20091195, "step": 947, "time_per_iteration": 2.687347173690796 }, { "auxiliary_loss_clip": 0.01203373, "auxiliary_loss_mlp": 0.01045362, "balance_loss_clip": 1.06317484, "balance_loss_mlp": 1.03506851, "epoch": 0.11399026032585824, "flos": 25082458441440.0, "grad_norm": 2.4099136646061163, "language_loss": 0.78314352, "learning_rate": 3.926545624708513e-06, "loss": 0.80563092, "num_input_tokens_seen": 20110435, "step": 948, "time_per_iteration": 2.666694164276123 }, { "auxiliary_loss_clip": 0.01197885, "auxiliary_loss_mlp": 0.01045022, "balance_loss_clip": 1.06045949, "balance_loss_mlp": 1.03447747, "epoch": 0.11411050321649732, "flos": 17961837916320.0, "grad_norm": 1.7646694782902344, "language_loss": 0.8570441, "learning_rate": 3.926336305495213e-06, "loss": 0.87947309, "num_input_tokens_seen": 20128995, "step": 949, "time_per_iteration": 2.6286401748657227 }, { "auxiliary_loss_clip": 0.01187793, "auxiliary_loss_mlp": 0.01042201, "balance_loss_clip": 1.05956817, "balance_loss_mlp": 1.03122795, "epoch": 0.11423074610713642, "flos": 22455413191680.0, "grad_norm": 2.3786422061226027, "language_loss": 0.88945067, "learning_rate": 3.926126694058226e-06, "loss": 0.91175067, "num_input_tokens_seen": 20148145, "step": 950, "time_per_iteration": 2.654972791671753 }, { "auxiliary_loss_clip": 0.01182113, "auxiliary_loss_mlp": 0.01028307, "balance_loss_clip": 1.06183386, "balance_loss_mlp": 1.01863933, "epoch": 0.1143509889977755, "flos": 19717691036160.0, "grad_norm": 1.522377617018036, "language_loss": 0.82177377, "learning_rate": 3.92591679042935e-06, "loss": 0.84387791, "num_input_tokens_seen": 20168035, "step": 951, "time_per_iteration": 2.686969041824341 }, { "auxiliary_loss_clip": 0.01231892, "auxiliary_loss_mlp": 0.0104003, "balance_loss_clip": 1.06832612, "balance_loss_mlp": 1.02908683, "epoch": 0.1144712318884146, "flos": 19822872647040.0, "grad_norm": 1.6187721490401008, "language_loss": 0.82473987, "learning_rate": 3.92570659464043e-06, "loss": 0.84745908, "num_input_tokens_seen": 20186095, "step": 952, "time_per_iteration": 2.5580146312713623 }, { "auxiliary_loss_clip": 0.01228627, "auxiliary_loss_mlp": 0.0076615, "balance_loss_clip": 1.06711602, "balance_loss_mlp": 1.0001204, "epoch": 0.1145914747790537, "flos": 14939201706720.0, "grad_norm": 1.8520353253882988, "language_loss": 0.79539919, "learning_rate": 3.925496106723349e-06, "loss": 0.81534702, "num_input_tokens_seen": 20203535, "step": 953, "time_per_iteration": 3.282984495162964 }, { "auxiliary_loss_clip": 0.01230318, "auxiliary_loss_mlp": 0.01035915, "balance_loss_clip": 1.06397367, "balance_loss_mlp": 1.02552009, "epoch": 0.11471171766969278, "flos": 19865032769760.0, "grad_norm": 2.3817096417529937, "language_loss": 0.83863354, "learning_rate": 3.9252853267100405e-06, "loss": 0.86129582, "num_input_tokens_seen": 20222780, "step": 954, "time_per_iteration": 3.2817542552948 }, { "auxiliary_loss_clip": 0.01193893, "auxiliary_loss_mlp": 0.01040707, "balance_loss_clip": 1.06238019, "balance_loss_mlp": 1.03022826, "epoch": 0.11483196056033187, "flos": 22526480078880.0, "grad_norm": 1.866138163194873, "language_loss": 0.83916473, "learning_rate": 3.9250742546324786e-06, "loss": 0.86151075, "num_input_tokens_seen": 20243015, "step": 955, "time_per_iteration": 2.673243522644043 }, { "auxiliary_loss_clip": 0.01210896, "auxiliary_loss_mlp": 0.01038401, "balance_loss_clip": 1.06227696, "balance_loss_mlp": 1.02908444, "epoch": 0.11495220345097096, "flos": 28220297093760.0, "grad_norm": 1.7926717745874454, "language_loss": 0.86974978, "learning_rate": 3.924862890522683e-06, "loss": 0.89224273, "num_input_tokens_seen": 20263025, "step": 956, "time_per_iteration": 2.6690001487731934 }, { "auxiliary_loss_clip": 0.01232231, "auxiliary_loss_mlp": 0.01038061, "balance_loss_clip": 1.06685042, "balance_loss_mlp": 1.02663493, "epoch": 0.11507244634161005, "flos": 17492270795520.0, "grad_norm": 2.059871985364134, "language_loss": 0.86174977, "learning_rate": 3.9246512344127174e-06, "loss": 0.88445276, "num_input_tokens_seen": 20280685, "step": 957, "time_per_iteration": 3.4905755519866943 }, { "auxiliary_loss_clip": 0.0114886, "auxiliary_loss_mlp": 0.01036855, "balance_loss_clip": 1.05568409, "balance_loss_mlp": 1.02647185, "epoch": 0.11519268923224914, "flos": 22564940754720.0, "grad_norm": 2.381283157978416, "language_loss": 0.81953889, "learning_rate": 3.9244392863346895e-06, "loss": 0.84139603, "num_input_tokens_seen": 20300090, "step": 958, "time_per_iteration": 3.6417155265808105 }, { "auxiliary_loss_clip": 0.01219965, "auxiliary_loss_mlp": 0.01039017, "balance_loss_clip": 1.06859481, "balance_loss_mlp": 1.02779317, "epoch": 0.11531293212288823, "flos": 16982842237440.0, "grad_norm": 1.81695956618723, "language_loss": 0.92481112, "learning_rate": 3.9242270463207524e-06, "loss": 0.94740093, "num_input_tokens_seen": 20318480, "step": 959, "time_per_iteration": 2.8482091426849365 }, { "auxiliary_loss_clip": 0.01167889, "auxiliary_loss_mlp": 0.01043151, "balance_loss_clip": 1.05518436, "balance_loss_mlp": 1.03242779, "epoch": 0.11543317501352733, "flos": 12422007272640.0, "grad_norm": 2.4625208968375083, "language_loss": 0.85207891, "learning_rate": 3.924014514403102e-06, "loss": 0.87418932, "num_input_tokens_seen": 20334635, "step": 960, "time_per_iteration": 2.6824491024017334 }, { "auxiliary_loss_clip": 0.01174993, "auxiliary_loss_mlp": 0.01039016, "balance_loss_clip": 1.05663109, "balance_loss_mlp": 1.02702904, "epoch": 0.11555341790416641, "flos": 19821651470400.0, "grad_norm": 2.1789480915093864, "language_loss": 0.91199607, "learning_rate": 3.92380169061398e-06, "loss": 0.93413615, "num_input_tokens_seen": 20352415, "step": 961, "time_per_iteration": 2.718966245651245 }, { "auxiliary_loss_clip": 0.01191519, "auxiliary_loss_mlp": 0.00766033, "balance_loss_clip": 1.05724478, "balance_loss_mlp": 1.00013208, "epoch": 0.11567366079480551, "flos": 25738869563520.0, "grad_norm": 1.927938613205274, "language_loss": 0.83553219, "learning_rate": 3.9235885749856705e-06, "loss": 0.85510778, "num_input_tokens_seen": 20371095, "step": 962, "time_per_iteration": 2.6670122146606445 }, { "auxiliary_loss_clip": 0.01215752, "auxiliary_loss_mlp": 0.0104651, "balance_loss_clip": 1.06676126, "balance_loss_mlp": 1.03522706, "epoch": 0.1157939036854446, "flos": 18223304583840.0, "grad_norm": 2.116969012727706, "language_loss": 0.8275671, "learning_rate": 3.9233751675505035e-06, "loss": 0.85018981, "num_input_tokens_seen": 20389805, "step": 963, "time_per_iteration": 2.6052236557006836 }, { "auxiliary_loss_clip": 0.01208895, "auxiliary_loss_mlp": 0.0104227, "balance_loss_clip": 1.06548429, "balance_loss_mlp": 1.03048635, "epoch": 0.11591414657608369, "flos": 23073758724480.0, "grad_norm": 1.7733774210667823, "language_loss": 0.84931403, "learning_rate": 3.923161468340853e-06, "loss": 0.8718257, "num_input_tokens_seen": 20409640, "step": 964, "time_per_iteration": 2.6227962970733643 }, { "auxiliary_loss_clip": 0.01169506, "auxiliary_loss_mlp": 0.01035315, "balance_loss_clip": 1.05352664, "balance_loss_mlp": 1.02447319, "epoch": 0.11603438946672277, "flos": 19461719663520.0, "grad_norm": 1.8621264434431974, "language_loss": 0.81864429, "learning_rate": 3.9229474773891374e-06, "loss": 0.84069252, "num_input_tokens_seen": 20428180, "step": 965, "time_per_iteration": 2.6708545684814453 }, { "auxiliary_loss_clip": 0.01207921, "auxiliary_loss_mlp": 0.01032832, "balance_loss_clip": 1.05808723, "balance_loss_mlp": 1.02170444, "epoch": 0.11615463235736187, "flos": 26831990176320.0, "grad_norm": 2.586054234900975, "language_loss": 0.83954871, "learning_rate": 3.922733194727818e-06, "loss": 0.8619563, "num_input_tokens_seen": 20447975, "step": 966, "time_per_iteration": 2.6877503395080566 }, { "auxiliary_loss_clip": 0.01234565, "auxiliary_loss_mlp": 0.01039588, "balance_loss_clip": 1.06671143, "balance_loss_mlp": 1.02949071, "epoch": 0.11627487524800097, "flos": 18580327116960.0, "grad_norm": 1.9491487227674371, "language_loss": 0.87263441, "learning_rate": 3.922518620389402e-06, "loss": 0.89537591, "num_input_tokens_seen": 20464840, "step": 967, "time_per_iteration": 2.566600799560547 }, { "auxiliary_loss_clip": 0.01121405, "auxiliary_loss_mlp": 0.01042996, "balance_loss_clip": 1.05097628, "balance_loss_mlp": 1.03367376, "epoch": 0.11639511813864005, "flos": 18150477765600.0, "grad_norm": 1.8057899033484908, "language_loss": 0.89592826, "learning_rate": 3.922303754406439e-06, "loss": 0.91757232, "num_input_tokens_seen": 20482680, "step": 968, "time_per_iteration": 2.7244739532470703 }, { "auxiliary_loss_clip": 0.01175718, "auxiliary_loss_mlp": 0.01036416, "balance_loss_clip": 1.05412006, "balance_loss_mlp": 1.02579498, "epoch": 0.11651536102927915, "flos": 20922027309120.0, "grad_norm": 1.691121310981471, "language_loss": 0.78983831, "learning_rate": 3.922088596811526e-06, "loss": 0.81195974, "num_input_tokens_seen": 20501810, "step": 969, "time_per_iteration": 2.651287794113159 }, { "auxiliary_loss_clip": 0.01219709, "auxiliary_loss_mlp": 0.0104127, "balance_loss_clip": 1.06345916, "balance_loss_mlp": 1.03022552, "epoch": 0.11663560391991823, "flos": 16508605911840.0, "grad_norm": 2.2080493987500356, "language_loss": 0.86883545, "learning_rate": 3.9218731476373e-06, "loss": 0.89144522, "num_input_tokens_seen": 20517995, "step": 970, "time_per_iteration": 2.5517470836639404 }, { "auxiliary_loss_clip": 0.0123439, "auxiliary_loss_mlp": 0.01037594, "balance_loss_clip": 1.06884575, "balance_loss_mlp": 1.02739561, "epoch": 0.11675584681055733, "flos": 19865032769760.0, "grad_norm": 1.9521327537098012, "language_loss": 0.84794611, "learning_rate": 3.9216574069164455e-06, "loss": 0.87066591, "num_input_tokens_seen": 20536970, "step": 971, "time_per_iteration": 2.5738625526428223 }, { "auxiliary_loss_clip": 0.01240127, "auxiliary_loss_mlp": 0.01036566, "balance_loss_clip": 1.06528461, "balance_loss_mlp": 1.02681518, "epoch": 0.11687608970119642, "flos": 21944368370400.0, "grad_norm": 1.5260722010607342, "language_loss": 0.80411649, "learning_rate": 3.921441374681691e-06, "loss": 0.82688349, "num_input_tokens_seen": 20557030, "step": 972, "time_per_iteration": 2.551624298095703 }, { "auxiliary_loss_clip": 0.01208493, "auxiliary_loss_mlp": 0.01040848, "balance_loss_clip": 1.06328809, "balance_loss_mlp": 1.03067923, "epoch": 0.1169963325918355, "flos": 24061158972000.0, "grad_norm": 1.8689085591137033, "language_loss": 0.64922094, "learning_rate": 3.921225050965808e-06, "loss": 0.67171431, "num_input_tokens_seen": 20576915, "step": 973, "time_per_iteration": 2.636671543121338 }, { "auxiliary_loss_clip": 0.01194088, "auxiliary_loss_mlp": 0.01034439, "balance_loss_clip": 1.06190753, "balance_loss_mlp": 1.0239073, "epoch": 0.1171165754824746, "flos": 23368154856000.0, "grad_norm": 2.0484292080506075, "language_loss": 0.75037521, "learning_rate": 3.921008435801612e-06, "loss": 0.77266049, "num_input_tokens_seen": 20596000, "step": 974, "time_per_iteration": 2.631430149078369 }, { "auxiliary_loss_clip": 0.01217906, "auxiliary_loss_mlp": 0.01042691, "balance_loss_clip": 1.06289768, "balance_loss_mlp": 1.03174794, "epoch": 0.11723681837311369, "flos": 18552246442560.0, "grad_norm": 3.783354013580105, "language_loss": 0.75940025, "learning_rate": 3.920791529221963e-06, "loss": 0.78200626, "num_input_tokens_seen": 20614675, "step": 975, "time_per_iteration": 2.575665235519409 }, { "auxiliary_loss_clip": 0.01214828, "auxiliary_loss_mlp": 0.00765639, "balance_loss_clip": 1.06336236, "balance_loss_mlp": 1.00008237, "epoch": 0.11735706126375278, "flos": 23550545154240.0, "grad_norm": 1.8843221587499712, "language_loss": 0.7676264, "learning_rate": 3.920574331259768e-06, "loss": 0.78743112, "num_input_tokens_seen": 20635875, "step": 976, "time_per_iteration": 2.6461870670318604 }, { "auxiliary_loss_clip": 0.01203419, "auxiliary_loss_mlp": 0.01041389, "balance_loss_clip": 1.06248617, "balance_loss_mlp": 1.03135705, "epoch": 0.11747730415439187, "flos": 22381544781600.0, "grad_norm": 2.0344810895445775, "language_loss": 0.7982651, "learning_rate": 3.9203568419479716e-06, "loss": 0.8207131, "num_input_tokens_seen": 20656430, "step": 977, "time_per_iteration": 2.6307032108306885 }, { "auxiliary_loss_clip": 0.01213448, "auxiliary_loss_mlp": 0.01039536, "balance_loss_clip": 1.06436586, "balance_loss_mlp": 1.02996325, "epoch": 0.11759754704503096, "flos": 22200734829600.0, "grad_norm": 1.70737286260215, "language_loss": 0.7570464, "learning_rate": 3.92013906131957e-06, "loss": 0.7795763, "num_input_tokens_seen": 20675360, "step": 978, "time_per_iteration": 2.6113264560699463 }, { "auxiliary_loss_clip": 0.01193591, "auxiliary_loss_mlp": 0.01045295, "balance_loss_clip": 1.05957508, "balance_loss_mlp": 1.03571057, "epoch": 0.11771778993567006, "flos": 22309759555200.0, "grad_norm": 1.492533618847831, "language_loss": 0.82424009, "learning_rate": 3.9199209894076e-06, "loss": 0.8466289, "num_input_tokens_seen": 20695675, "step": 979, "time_per_iteration": 2.6465048789978027 }, { "auxiliary_loss_clip": 0.01246794, "auxiliary_loss_mlp": 0.01040198, "balance_loss_clip": 1.06625497, "balance_loss_mlp": 1.02895057, "epoch": 0.11783803282630914, "flos": 21288172750080.0, "grad_norm": 2.117319580920693, "language_loss": 0.89923286, "learning_rate": 3.919702626245142e-06, "loss": 0.92210281, "num_input_tokens_seen": 20715330, "step": 980, "time_per_iteration": 4.049605369567871 }, { "auxiliary_loss_clip": 0.01197279, "auxiliary_loss_mlp": 0.01042282, "balance_loss_clip": 1.05810857, "balance_loss_mlp": 1.03188741, "epoch": 0.11795827571694824, "flos": 25371538862880.0, "grad_norm": 2.031979392826253, "language_loss": 0.66063535, "learning_rate": 3.919483971865322e-06, "loss": 0.68303096, "num_input_tokens_seen": 20735325, "step": 981, "time_per_iteration": 2.639383316040039 }, { "auxiliary_loss_clip": 0.01209905, "auxiliary_loss_mlp": 0.0104639, "balance_loss_clip": 1.06486535, "balance_loss_mlp": 1.03610826, "epoch": 0.11807851860758732, "flos": 23622222629760.0, "grad_norm": 1.9482345857582088, "language_loss": 0.87704432, "learning_rate": 3.91926502630131e-06, "loss": 0.89960724, "num_input_tokens_seen": 20755940, "step": 982, "time_per_iteration": 2.6461434364318848 }, { "auxiliary_loss_clip": 0.01232499, "auxiliary_loss_mlp": 0.01041854, "balance_loss_clip": 1.07021332, "balance_loss_mlp": 1.03248394, "epoch": 0.11819876149822642, "flos": 24972499874880.0, "grad_norm": 1.7655391201738813, "language_loss": 0.71974075, "learning_rate": 3.91904578958632e-06, "loss": 0.74248433, "num_input_tokens_seen": 20775355, "step": 983, "time_per_iteration": 3.5337941646575928 }, { "auxiliary_loss_clip": 0.01246275, "auxiliary_loss_mlp": 0.01035611, "balance_loss_clip": 1.06964469, "balance_loss_mlp": 1.02577615, "epoch": 0.11831900438886551, "flos": 23003230591680.0, "grad_norm": 1.8905301711174414, "language_loss": 0.84073955, "learning_rate": 3.918826261753608e-06, "loss": 0.86355847, "num_input_tokens_seen": 20794935, "step": 984, "time_per_iteration": 2.6135125160217285 }, { "auxiliary_loss_clip": 0.01215855, "auxiliary_loss_mlp": 0.01035638, "balance_loss_clip": 1.06456494, "balance_loss_mlp": 1.02639914, "epoch": 0.1184392472795046, "flos": 27965151811200.0, "grad_norm": 2.7646232069379533, "language_loss": 0.70962262, "learning_rate": 3.918606442836478e-06, "loss": 0.73213756, "num_input_tokens_seen": 20817155, "step": 985, "time_per_iteration": 3.5109994411468506 }, { "auxiliary_loss_clip": 0.01229812, "auxiliary_loss_mlp": 0.01037869, "balance_loss_clip": 1.06996536, "balance_loss_mlp": 1.02832043, "epoch": 0.1185594901701437, "flos": 19898500988160.0, "grad_norm": 1.7587649626682864, "language_loss": 0.77413046, "learning_rate": 3.918386332868277e-06, "loss": 0.79680729, "num_input_tokens_seen": 20835125, "step": 986, "time_per_iteration": 2.5735373497009277 }, { "auxiliary_loss_clip": 0.01219923, "auxiliary_loss_mlp": 0.01040081, "balance_loss_clip": 1.0649879, "balance_loss_mlp": 1.0294776, "epoch": 0.11867973306078278, "flos": 18912357834240.0, "grad_norm": 1.8593265851732703, "language_loss": 0.94558042, "learning_rate": 3.918165931882394e-06, "loss": 0.96818054, "num_input_tokens_seen": 20853525, "step": 987, "time_per_iteration": 2.586812734603882 }, { "auxiliary_loss_clip": 0.01154362, "auxiliary_loss_mlp": 0.01040961, "balance_loss_clip": 1.05517149, "balance_loss_mlp": 1.03112042, "epoch": 0.11879997595142187, "flos": 16982806320480.0, "grad_norm": 3.4947410178902505, "language_loss": 0.75397182, "learning_rate": 3.917945239912264e-06, "loss": 0.7759251, "num_input_tokens_seen": 20871000, "step": 988, "time_per_iteration": 2.696259021759033 }, { "auxiliary_loss_clip": 0.01182774, "auxiliary_loss_mlp": 0.01040771, "balance_loss_clip": 1.05969071, "balance_loss_mlp": 1.03158617, "epoch": 0.11892021884206096, "flos": 17530372301760.0, "grad_norm": 2.1848555933202283, "language_loss": 0.75420994, "learning_rate": 3.917724256991367e-06, "loss": 0.77644539, "num_input_tokens_seen": 20889745, "step": 989, "time_per_iteration": 2.700300455093384 }, { "auxiliary_loss_clip": 0.01201907, "auxiliary_loss_mlp": 0.01035575, "balance_loss_clip": 1.06139779, "balance_loss_mlp": 1.02473891, "epoch": 0.11904046173270005, "flos": 30955900148640.0, "grad_norm": 2.515787503972399, "language_loss": 0.81671554, "learning_rate": 3.9175029831532245e-06, "loss": 0.83909035, "num_input_tokens_seen": 20909260, "step": 990, "time_per_iteration": 2.686183214187622 }, { "auxiliary_loss_clip": 0.01200373, "auxiliary_loss_mlp": 0.01040957, "balance_loss_clip": 1.06395006, "balance_loss_mlp": 1.0313127, "epoch": 0.11916070462333915, "flos": 20157237966720.0, "grad_norm": 2.039328076258691, "language_loss": 0.88783985, "learning_rate": 3.917281418431404e-06, "loss": 0.91025317, "num_input_tokens_seen": 20928305, "step": 991, "time_per_iteration": 2.6383352279663086 }, { "auxiliary_loss_clip": 0.01212988, "auxiliary_loss_mlp": 0.01036062, "balance_loss_clip": 1.06784654, "balance_loss_mlp": 1.0254997, "epoch": 0.11928094751397823, "flos": 23551119825600.0, "grad_norm": 2.0529612736582106, "language_loss": 0.76832592, "learning_rate": 3.917059562859516e-06, "loss": 0.79081643, "num_input_tokens_seen": 20947630, "step": 992, "time_per_iteration": 2.6359071731567383 }, { "auxiliary_loss_clip": 0.01200321, "auxiliary_loss_mlp": 0.01039755, "balance_loss_clip": 1.06278837, "balance_loss_mlp": 1.02970588, "epoch": 0.11940119040461733, "flos": 23908429694400.0, "grad_norm": 2.0087042864067954, "language_loss": 0.89111996, "learning_rate": 3.916837416471218e-06, "loss": 0.91352075, "num_input_tokens_seen": 20964250, "step": 993, "time_per_iteration": 2.617818832397461 }, { "auxiliary_loss_clip": 0.01219514, "auxiliary_loss_mlp": 0.01037612, "balance_loss_clip": 1.06084967, "balance_loss_mlp": 1.02767634, "epoch": 0.11952143329525641, "flos": 13844536664640.0, "grad_norm": 2.510809333212766, "language_loss": 0.72034162, "learning_rate": 3.916614979300207e-06, "loss": 0.74291289, "num_input_tokens_seen": 20979095, "step": 994, "time_per_iteration": 2.5599846839904785 }, { "auxiliary_loss_clip": 0.01172616, "auxiliary_loss_mlp": 0.0104148, "balance_loss_clip": 1.06011045, "balance_loss_mlp": 1.03198481, "epoch": 0.11964167618589551, "flos": 27015529817280.0, "grad_norm": 1.6084034941070455, "language_loss": 0.78856581, "learning_rate": 3.9163922513802274e-06, "loss": 0.81070673, "num_input_tokens_seen": 21001430, "step": 995, "time_per_iteration": 2.7323594093322754 }, { "auxiliary_loss_clip": 0.01247494, "auxiliary_loss_mlp": 0.01040192, "balance_loss_clip": 1.0680387, "balance_loss_mlp": 1.02941561, "epoch": 0.1197619190765346, "flos": 12567624992160.0, "grad_norm": 2.4906554360987925, "language_loss": 0.82509351, "learning_rate": 3.916169232745067e-06, "loss": 0.84797037, "num_input_tokens_seen": 21019105, "step": 996, "time_per_iteration": 2.528564691543579 }, { "auxiliary_loss_clip": 0.01202402, "auxiliary_loss_mlp": 0.01039975, "balance_loss_clip": 1.06157231, "balance_loss_mlp": 1.02841187, "epoch": 0.11988216196717369, "flos": 16909440747840.0, "grad_norm": 2.417106384687524, "language_loss": 0.92226744, "learning_rate": 3.915945923428559e-06, "loss": 0.94469118, "num_input_tokens_seen": 21035630, "step": 997, "time_per_iteration": 2.5788238048553467 }, { "auxiliary_loss_clip": 0.01227399, "auxiliary_loss_mlp": 0.01041111, "balance_loss_clip": 1.06555152, "balance_loss_mlp": 1.03112173, "epoch": 0.12000240485781279, "flos": 16216580299680.0, "grad_norm": 2.234354323378954, "language_loss": 0.83038831, "learning_rate": 3.915722323464577e-06, "loss": 0.85307336, "num_input_tokens_seen": 21054235, "step": 998, "time_per_iteration": 2.5855844020843506 }, { "auxiliary_loss_clip": 0.01229255, "auxiliary_loss_mlp": 0.01040404, "balance_loss_clip": 1.06592798, "balance_loss_mlp": 1.02940083, "epoch": 0.12012264774845187, "flos": 49344893644320.0, "grad_norm": 2.386018269077979, "language_loss": 0.69761944, "learning_rate": 3.91549843288704e-06, "loss": 0.72031605, "num_input_tokens_seen": 21077915, "step": 999, "time_per_iteration": 2.8020005226135254 }, { "auxiliary_loss_clip": 0.01197605, "auxiliary_loss_mlp": 0.00765016, "balance_loss_clip": 1.05795884, "balance_loss_mlp": 1.0000875, "epoch": 0.12024289063909097, "flos": 26979439660800.0, "grad_norm": 4.3511694906623966, "language_loss": 0.79329818, "learning_rate": 3.915274251729916e-06, "loss": 0.81292439, "num_input_tokens_seen": 21099205, "step": 1000, "time_per_iteration": 2.7077507972717285 }, { "auxiliary_loss_clip": 0.01202846, "auxiliary_loss_mlp": 0.01033468, "balance_loss_clip": 1.06534624, "balance_loss_mlp": 1.02355564, "epoch": 0.12036313352973005, "flos": 19537312087680.0, "grad_norm": 2.1416837013240135, "language_loss": 0.90116692, "learning_rate": 3.91504978002721e-06, "loss": 0.92353004, "num_input_tokens_seen": 21118260, "step": 1001, "time_per_iteration": 2.6366045475006104 }, { "auxiliary_loss_clip": 0.01214223, "auxiliary_loss_mlp": 0.00765809, "balance_loss_clip": 1.06095815, "balance_loss_mlp": 1.00013351, "epoch": 0.12048337642036915, "flos": 17268259128960.0, "grad_norm": 2.0868516787729785, "language_loss": 0.76139212, "learning_rate": 3.914825017812974e-06, "loss": 0.78119242, "num_input_tokens_seen": 21134910, "step": 1002, "time_per_iteration": 2.619255304336548 }, { "auxiliary_loss_clip": 0.01213725, "auxiliary_loss_mlp": 0.01043915, "balance_loss_clip": 1.06646705, "balance_loss_mlp": 1.03364551, "epoch": 0.12060361931100824, "flos": 22856966366880.0, "grad_norm": 2.0826480086220442, "language_loss": 0.72963965, "learning_rate": 3.9145999651213065e-06, "loss": 0.75221604, "num_input_tokens_seen": 21154150, "step": 1003, "time_per_iteration": 2.610697031021118 }, { "auxiliary_loss_clip": 0.01227927, "auxiliary_loss_mlp": 0.01032566, "balance_loss_clip": 1.06693673, "balance_loss_mlp": 1.02144384, "epoch": 0.12072386220164733, "flos": 16726799030880.0, "grad_norm": 11.081054697138285, "language_loss": 0.88649142, "learning_rate": 3.9143746219863465e-06, "loss": 0.90909636, "num_input_tokens_seen": 21171255, "step": 1004, "time_per_iteration": 2.5854544639587402 }, { "auxiliary_loss_clip": 0.01120983, "auxiliary_loss_mlp": 0.0100398, "balance_loss_clip": 1.02993214, "balance_loss_mlp": 1.00142884, "epoch": 0.12084410509228642, "flos": 55144185994080.0, "grad_norm": 0.9453003248824734, "language_loss": 0.64829838, "learning_rate": 3.914148988442278e-06, "loss": 0.66954803, "num_input_tokens_seen": 21227045, "step": 1005, "time_per_iteration": 3.8996171951293945 }, { "auxiliary_loss_clip": 0.0120219, "auxiliary_loss_mlp": 0.01037304, "balance_loss_clip": 1.06380582, "balance_loss_mlp": 1.02619946, "epoch": 0.1209643479829255, "flos": 26760240866880.0, "grad_norm": 2.9284719720095427, "language_loss": 0.9563148, "learning_rate": 3.91392306452333e-06, "loss": 0.9787097, "num_input_tokens_seen": 21244120, "step": 1006, "time_per_iteration": 3.3982532024383545 }, { "auxiliary_loss_clip": 0.01245864, "auxiliary_loss_mlp": 0.01047018, "balance_loss_clip": 1.0684166, "balance_loss_mlp": 1.03521013, "epoch": 0.1210845908735646, "flos": 11035029282720.0, "grad_norm": 2.876164890201792, "language_loss": 0.66880971, "learning_rate": 3.913696850263774e-06, "loss": 0.69173849, "num_input_tokens_seen": 21258485, "step": 1007, "time_per_iteration": 2.5332441329956055 }, { "auxiliary_loss_clip": 0.01230066, "auxiliary_loss_mlp": 0.01043721, "balance_loss_clip": 1.06674719, "balance_loss_mlp": 1.03327191, "epoch": 0.1212048337642037, "flos": 20484635396160.0, "grad_norm": 1.8589154422830634, "language_loss": 0.79378378, "learning_rate": 3.913470345697929e-06, "loss": 0.8165217, "num_input_tokens_seen": 21277115, "step": 1008, "time_per_iteration": 2.650498867034912 }, { "auxiliary_loss_clip": 0.01181228, "auxiliary_loss_mlp": 0.01033532, "balance_loss_clip": 1.06027532, "balance_loss_mlp": 1.0242393, "epoch": 0.12132507665484278, "flos": 22346065213440.0, "grad_norm": 2.0940699085775236, "language_loss": 0.85613322, "learning_rate": 3.913243550860153e-06, "loss": 0.87828088, "num_input_tokens_seen": 21294880, "step": 1009, "time_per_iteration": 3.5613319873809814 }, { "auxiliary_loss_clip": 0.01231943, "auxiliary_loss_mlp": 0.01038344, "balance_loss_clip": 1.06983185, "balance_loss_mlp": 1.02703679, "epoch": 0.12144531954548188, "flos": 29314962135840.0, "grad_norm": 1.7651414204512883, "language_loss": 0.76565063, "learning_rate": 3.913016465784852e-06, "loss": 0.7883535, "num_input_tokens_seen": 21315555, "step": 1010, "time_per_iteration": 3.4910309314727783 }, { "auxiliary_loss_clip": 0.011759, "auxiliary_loss_mlp": 0.01037356, "balance_loss_clip": 1.05805469, "balance_loss_mlp": 1.02674007, "epoch": 0.12156556243612096, "flos": 20485245984480.0, "grad_norm": 2.975901981852568, "language_loss": 0.72095704, "learning_rate": 3.912789090506474e-06, "loss": 0.74308956, "num_input_tokens_seen": 21334815, "step": 1011, "time_per_iteration": 2.6527559757232666 }, { "auxiliary_loss_clip": 0.01206013, "auxiliary_loss_mlp": 0.01043359, "balance_loss_clip": 1.06145859, "balance_loss_mlp": 1.03330374, "epoch": 0.12168580532676006, "flos": 16472012917920.0, "grad_norm": 3.1896381252956663, "language_loss": 0.71903688, "learning_rate": 3.9125614250595114e-06, "loss": 0.7415306, "num_input_tokens_seen": 21351025, "step": 1012, "time_per_iteration": 2.629240036010742 }, { "auxiliary_loss_clip": 0.01228884, "auxiliary_loss_mlp": 0.01043102, "balance_loss_clip": 1.06497335, "balance_loss_mlp": 1.03242683, "epoch": 0.12180604821739914, "flos": 15341293636320.0, "grad_norm": 2.386538476675492, "language_loss": 0.88639355, "learning_rate": 3.912333469478502e-06, "loss": 0.90911347, "num_input_tokens_seen": 21368990, "step": 1013, "time_per_iteration": 2.5499460697174072 }, { "auxiliary_loss_clip": 0.01214706, "auxiliary_loss_mlp": 0.01043659, "balance_loss_clip": 1.06387067, "balance_loss_mlp": 1.03316879, "epoch": 0.12192629110803824, "flos": 19318041459840.0, "grad_norm": 1.9918014255714318, "language_loss": 0.77840543, "learning_rate": 3.912105223798025e-06, "loss": 0.80098903, "num_input_tokens_seen": 21388410, "step": 1014, "time_per_iteration": 2.6155850887298584 }, { "auxiliary_loss_clip": 0.01104903, "auxiliary_loss_mlp": 0.01004095, "balance_loss_clip": 1.02601087, "balance_loss_mlp": 1.00177062, "epoch": 0.12204653399867733, "flos": 47725362139680.0, "grad_norm": 1.0101362449896938, "language_loss": 0.67770219, "learning_rate": 3.9118766880527065e-06, "loss": 0.69879222, "num_input_tokens_seen": 21442845, "step": 1015, "time_per_iteration": 3.1035335063934326 }, { "auxiliary_loss_clip": 0.01171872, "auxiliary_loss_mlp": 0.01036338, "balance_loss_clip": 1.0581342, "balance_loss_mlp": 1.02645004, "epoch": 0.12216677688931642, "flos": 18221939739360.0, "grad_norm": 1.644739669524897, "language_loss": 0.73841232, "learning_rate": 3.9116478622772145e-06, "loss": 0.76049441, "num_input_tokens_seen": 21461420, "step": 1016, "time_per_iteration": 2.6724729537963867 }, { "auxiliary_loss_clip": 0.01226317, "auxiliary_loss_mlp": 0.01042115, "balance_loss_clip": 1.0687654, "balance_loss_mlp": 1.03130865, "epoch": 0.12228701977995551, "flos": 27525640797600.0, "grad_norm": 2.9036455600892537, "language_loss": 0.87813824, "learning_rate": 3.911418746506261e-06, "loss": 0.90082252, "num_input_tokens_seen": 21481550, "step": 1017, "time_per_iteration": 2.6409385204315186 }, { "auxiliary_loss_clip": 0.01233015, "auxiliary_loss_mlp": 0.01040172, "balance_loss_clip": 1.07304835, "balance_loss_mlp": 1.02941298, "epoch": 0.1224072626705946, "flos": 21798140062560.0, "grad_norm": 1.697436563952464, "language_loss": 0.78508556, "learning_rate": 3.911189340774604e-06, "loss": 0.8078174, "num_input_tokens_seen": 21501680, "step": 1018, "time_per_iteration": 2.5757715702056885 }, { "auxiliary_loss_clip": 0.01222094, "auxiliary_loss_mlp": 0.01039431, "balance_loss_clip": 1.06567788, "balance_loss_mlp": 1.02947664, "epoch": 0.1225275055612337, "flos": 20703762356160.0, "grad_norm": 3.8112941015726864, "language_loss": 0.7935766, "learning_rate": 3.910959645117043e-06, "loss": 0.81619185, "num_input_tokens_seen": 21521015, "step": 1019, "time_per_iteration": 2.621568202972412 }, { "auxiliary_loss_clip": 0.01112475, "auxiliary_loss_mlp": 0.00756474, "balance_loss_clip": 1.03093493, "balance_loss_mlp": 0.99958801, "epoch": 0.12264774845187278, "flos": 57745305587040.0, "grad_norm": 0.8175512836752886, "language_loss": 0.56795663, "learning_rate": 3.910729659568423e-06, "loss": 0.58664608, "num_input_tokens_seen": 21578200, "step": 1020, "time_per_iteration": 3.1621005535125732 }, { "auxiliary_loss_clip": 0.01210601, "auxiliary_loss_mlp": 0.01037332, "balance_loss_clip": 1.06478131, "balance_loss_mlp": 1.02796221, "epoch": 0.12276799134251187, "flos": 26396286360480.0, "grad_norm": 1.6477851360433187, "language_loss": 0.82260418, "learning_rate": 3.9104993841636344e-06, "loss": 0.84508348, "num_input_tokens_seen": 21598770, "step": 1021, "time_per_iteration": 2.668044328689575 }, { "auxiliary_loss_clip": 0.01211818, "auxiliary_loss_mlp": 0.00764397, "balance_loss_clip": 1.07003176, "balance_loss_mlp": 1.00000095, "epoch": 0.12288823423315097, "flos": 21064197000480.0, "grad_norm": 1.7989191012694121, "language_loss": 0.80889261, "learning_rate": 3.910268818937608e-06, "loss": 0.82865477, "num_input_tokens_seen": 21616925, "step": 1022, "time_per_iteration": 2.653303861618042 }, { "auxiliary_loss_clip": 0.01184209, "auxiliary_loss_mlp": 0.01039519, "balance_loss_clip": 1.06427145, "balance_loss_mlp": 1.0286231, "epoch": 0.12300847712379005, "flos": 12312443792640.0, "grad_norm": 2.310576655858731, "language_loss": 0.87501717, "learning_rate": 3.9100379639253196e-06, "loss": 0.89725447, "num_input_tokens_seen": 21633645, "step": 1023, "time_per_iteration": 2.6460819244384766 }, { "auxiliary_loss_clip": 0.01211166, "auxiliary_loss_mlp": 0.01041173, "balance_loss_clip": 1.06265378, "balance_loss_mlp": 1.02940154, "epoch": 0.12312872001442915, "flos": 16762242682080.0, "grad_norm": 2.5446772034532112, "language_loss": 0.86309808, "learning_rate": 3.909806819161791e-06, "loss": 0.88562149, "num_input_tokens_seen": 21649120, "step": 1024, "time_per_iteration": 2.571484088897705 }, { "auxiliary_loss_clip": 0.01203962, "auxiliary_loss_mlp": 0.0103717, "balance_loss_clip": 1.06230319, "balance_loss_mlp": 1.02647126, "epoch": 0.12324896290506823, "flos": 18404940625920.0, "grad_norm": 1.8727789055219595, "language_loss": 0.86264324, "learning_rate": 3.909575384682086e-06, "loss": 0.88505459, "num_input_tokens_seen": 21668000, "step": 1025, "time_per_iteration": 2.628757953643799 }, { "auxiliary_loss_clip": 0.01230908, "auxiliary_loss_mlp": 0.01038748, "balance_loss_clip": 1.06569242, "balance_loss_mlp": 1.02888346, "epoch": 0.12336920579570733, "flos": 18915410775840.0, "grad_norm": 1.7058286971186227, "language_loss": 0.69366229, "learning_rate": 3.9093436605213144e-06, "loss": 0.71635884, "num_input_tokens_seen": 21688500, "step": 1026, "time_per_iteration": 2.541470527648926 }, { "auxiliary_loss_clip": 0.01212234, "auxiliary_loss_mlp": 0.01038081, "balance_loss_clip": 1.06320202, "balance_loss_mlp": 1.02810931, "epoch": 0.12348944868634643, "flos": 23878373587200.0, "grad_norm": 1.7079268386337643, "language_loss": 0.79619211, "learning_rate": 3.909111646714627e-06, "loss": 0.81869525, "num_input_tokens_seen": 21709345, "step": 1027, "time_per_iteration": 2.655040979385376 }, { "auxiliary_loss_clip": 0.01242045, "auxiliary_loss_mlp": 0.01036877, "balance_loss_clip": 1.0678606, "balance_loss_mlp": 1.02704787, "epoch": 0.12360969157698551, "flos": 19026087681600.0, "grad_norm": 1.8754239431037787, "language_loss": 0.72466278, "learning_rate": 3.9088793432972206e-06, "loss": 0.74745196, "num_input_tokens_seen": 21728165, "step": 1028, "time_per_iteration": 2.498469829559326 }, { "auxiliary_loss_clip": 0.01184456, "auxiliary_loss_mlp": 0.01039699, "balance_loss_clip": 1.06171823, "balance_loss_mlp": 1.02980423, "epoch": 0.1237299344676246, "flos": 13224610785600.0, "grad_norm": 2.0098196193852447, "language_loss": 0.82303876, "learning_rate": 3.908646750304336e-06, "loss": 0.84528029, "num_input_tokens_seen": 21745850, "step": 1029, "time_per_iteration": 2.6881062984466553 }, { "auxiliary_loss_clip": 0.01218712, "auxiliary_loss_mlp": 0.01043444, "balance_loss_clip": 1.06833363, "balance_loss_mlp": 1.03288221, "epoch": 0.12385017735826369, "flos": 20485676988000.0, "grad_norm": 2.0304840897508356, "language_loss": 0.87230206, "learning_rate": 3.908413867771257e-06, "loss": 0.89492369, "num_input_tokens_seen": 21764760, "step": 1030, "time_per_iteration": 2.591869354248047 }, { "auxiliary_loss_clip": 0.01229636, "auxiliary_loss_mlp": 0.01044136, "balance_loss_clip": 1.06942904, "balance_loss_mlp": 1.03410411, "epoch": 0.12397042024890279, "flos": 17347838335680.0, "grad_norm": 1.5911670276513132, "language_loss": 0.8066535, "learning_rate": 3.908180695733311e-06, "loss": 0.82939124, "num_input_tokens_seen": 21784250, "step": 1031, "time_per_iteration": 3.3333442211151123 }, { "auxiliary_loss_clip": 0.01155281, "auxiliary_loss_mlp": 0.01045097, "balance_loss_clip": 1.05329609, "balance_loss_mlp": 1.03386736, "epoch": 0.12409066313954187, "flos": 20412347332320.0, "grad_norm": 1.7364078001580194, "language_loss": 0.82164538, "learning_rate": 3.907947234225871e-06, "loss": 0.84364915, "num_input_tokens_seen": 21803260, "step": 1032, "time_per_iteration": 3.466615676879883 }, { "auxiliary_loss_clip": 0.0116141, "auxiliary_loss_mlp": 0.01040098, "balance_loss_clip": 1.05679345, "balance_loss_mlp": 1.02981615, "epoch": 0.12421090603018096, "flos": 20736691820160.0, "grad_norm": 1.8162773992561894, "language_loss": 0.87181807, "learning_rate": 3.907713483284352e-06, "loss": 0.89383316, "num_input_tokens_seen": 21822735, "step": 1033, "time_per_iteration": 2.6846275329589844 }, { "auxiliary_loss_clip": 0.01135468, "auxiliary_loss_mlp": 0.01039626, "balance_loss_clip": 1.04968357, "balance_loss_mlp": 1.02891517, "epoch": 0.12433114892082006, "flos": 24498838220640.0, "grad_norm": 2.7924438139969245, "language_loss": 0.97292674, "learning_rate": 3.907479442944216e-06, "loss": 0.9946776, "num_input_tokens_seen": 21841140, "step": 1034, "time_per_iteration": 2.737757921218872 }, { "auxiliary_loss_clip": 0.01225233, "auxiliary_loss_mlp": 0.0103476, "balance_loss_clip": 1.06630027, "balance_loss_mlp": 1.02391744, "epoch": 0.12445139181145914, "flos": 19682319218880.0, "grad_norm": 2.0617348708621277, "language_loss": 0.92523587, "learning_rate": 3.907245113240963e-06, "loss": 0.9478358, "num_input_tokens_seen": 21859260, "step": 1035, "time_per_iteration": 3.482732057571411 }, { "auxiliary_loss_clip": 0.01193766, "auxiliary_loss_mlp": 0.01038506, "balance_loss_clip": 1.05922794, "balance_loss_mlp": 1.0276041, "epoch": 0.12457163470209824, "flos": 46423093093440.0, "grad_norm": 1.6716975452147838, "language_loss": 0.73604405, "learning_rate": 3.907010494210144e-06, "loss": 0.7583667, "num_input_tokens_seen": 21881920, "step": 1036, "time_per_iteration": 3.75349497795105 }, { "auxiliary_loss_clip": 0.01231598, "auxiliary_loss_mlp": 0.01037212, "balance_loss_clip": 1.06748211, "balance_loss_mlp": 1.02711511, "epoch": 0.12469187759273732, "flos": 20376293092800.0, "grad_norm": 2.0218227450719204, "language_loss": 0.91891563, "learning_rate": 3.9067755858873495e-06, "loss": 0.94160366, "num_input_tokens_seen": 21898720, "step": 1037, "time_per_iteration": 2.580411195755005 }, { "auxiliary_loss_clip": 0.01093264, "auxiliary_loss_mlp": 0.01007839, "balance_loss_clip": 1.02097833, "balance_loss_mlp": 1.005669, "epoch": 0.12481212048337642, "flos": 69224652367680.0, "grad_norm": 0.8527012571586651, "language_loss": 0.62826133, "learning_rate": 3.906540388308214e-06, "loss": 0.64927232, "num_input_tokens_seen": 21958305, "step": 1038, "time_per_iteration": 3.1960105895996094 }, { "auxiliary_loss_clip": 0.01162267, "auxiliary_loss_mlp": 0.01035621, "balance_loss_clip": 1.0579989, "balance_loss_mlp": 1.02580976, "epoch": 0.12493236337401552, "flos": 18223699670400.0, "grad_norm": 1.8667892546123406, "language_loss": 0.81057024, "learning_rate": 3.906304901508417e-06, "loss": 0.8325491, "num_input_tokens_seen": 21977205, "step": 1039, "time_per_iteration": 2.647974729537964 }, { "auxiliary_loss_clip": 0.01229803, "auxiliary_loss_mlp": 0.01035594, "balance_loss_clip": 1.06961918, "balance_loss_mlp": 1.02547884, "epoch": 0.12505260626465461, "flos": 30044379660960.0, "grad_norm": 2.362361922675102, "language_loss": 0.75808901, "learning_rate": 3.9060691255236835e-06, "loss": 0.780743, "num_input_tokens_seen": 21997770, "step": 1040, "time_per_iteration": 2.652308225631714 }, { "auxiliary_loss_clip": 0.0122166, "auxiliary_loss_mlp": 0.01036335, "balance_loss_clip": 1.06218469, "balance_loss_mlp": 1.02612507, "epoch": 0.1251728491552937, "flos": 24433984967520.0, "grad_norm": 1.614678392039214, "language_loss": 0.80661738, "learning_rate": 3.905833060389778e-06, "loss": 0.82919741, "num_input_tokens_seen": 22021890, "step": 1041, "time_per_iteration": 2.6699585914611816 }, { "auxiliary_loss_clip": 0.01241772, "auxiliary_loss_mlp": 0.00764978, "balance_loss_clip": 1.06773734, "balance_loss_mlp": 1.00002265, "epoch": 0.12529309204593278, "flos": 27119813504160.0, "grad_norm": 1.9899403878189943, "language_loss": 0.78444779, "learning_rate": 3.905596706142513e-06, "loss": 0.80451524, "num_input_tokens_seen": 22043300, "step": 1042, "time_per_iteration": 2.6123251914978027 }, { "auxiliary_loss_clip": 0.01190873, "auxiliary_loss_mlp": 0.01040499, "balance_loss_clip": 1.05882537, "balance_loss_mlp": 1.02952623, "epoch": 0.12541333493657186, "flos": 30774156355680.0, "grad_norm": 1.941712744002252, "language_loss": 0.86039692, "learning_rate": 3.9053600628177435e-06, "loss": 0.8827107, "num_input_tokens_seen": 22062910, "step": 1043, "time_per_iteration": 2.7127573490142822 }, { "auxiliary_loss_clip": 0.01237711, "auxiliary_loss_mlp": 0.01040044, "balance_loss_clip": 1.06501508, "balance_loss_mlp": 1.03004289, "epoch": 0.12553357782721097, "flos": 23659569879840.0, "grad_norm": 2.0571114075400443, "language_loss": 0.84352958, "learning_rate": 3.905123130451367e-06, "loss": 0.86630714, "num_input_tokens_seen": 22084010, "step": 1044, "time_per_iteration": 2.5719051361083984 }, { "auxiliary_loss_clip": 0.01241454, "auxiliary_loss_mlp": 0.01036607, "balance_loss_clip": 1.06737351, "balance_loss_mlp": 1.02618766, "epoch": 0.12565382071785006, "flos": 24863762484960.0, "grad_norm": 1.6705303961428097, "language_loss": 0.79318154, "learning_rate": 3.904885909079326e-06, "loss": 0.8159622, "num_input_tokens_seen": 22102795, "step": 1045, "time_per_iteration": 2.5693092346191406 }, { "auxiliary_loss_clip": 0.0122629, "auxiliary_loss_mlp": 0.0103782, "balance_loss_clip": 1.06400824, "balance_loss_mlp": 1.0275147, "epoch": 0.12577406360848914, "flos": 21360784066560.0, "grad_norm": 2.4950156924760027, "language_loss": 0.78445607, "learning_rate": 3.904648398737607e-06, "loss": 0.80709714, "num_input_tokens_seen": 22121360, "step": 1046, "time_per_iteration": 2.5783309936523438 }, { "auxiliary_loss_clip": 0.01240847, "auxiliary_loss_mlp": 0.01043349, "balance_loss_clip": 1.06765223, "balance_loss_mlp": 1.0337646, "epoch": 0.12589430649912825, "flos": 36138061753920.0, "grad_norm": 2.308508331542329, "language_loss": 0.7837677, "learning_rate": 3.9044105994622406e-06, "loss": 0.80660963, "num_input_tokens_seen": 22142505, "step": 1047, "time_per_iteration": 2.6729471683502197 }, { "auxiliary_loss_clip": 0.01213752, "auxiliary_loss_mlp": 0.00765949, "balance_loss_clip": 1.06441975, "balance_loss_mlp": 1.0001173, "epoch": 0.12601454938976733, "flos": 25337675557920.0, "grad_norm": 1.972762451232197, "language_loss": 0.81724858, "learning_rate": 3.9041725112893005e-06, "loss": 0.83704555, "num_input_tokens_seen": 22163730, "step": 1048, "time_per_iteration": 2.6607394218444824 }, { "auxiliary_loss_clip": 0.0118785, "auxiliary_loss_mlp": 0.0103233, "balance_loss_clip": 1.06200147, "balance_loss_mlp": 1.02243614, "epoch": 0.12613479228040642, "flos": 15560097343680.0, "grad_norm": 3.2527082797004847, "language_loss": 0.75171655, "learning_rate": 3.903934134254904e-06, "loss": 0.77391839, "num_input_tokens_seen": 22181520, "step": 1049, "time_per_iteration": 2.610858678817749 }, { "auxiliary_loss_clip": 0.01230072, "auxiliary_loss_mlp": 0.01034762, "balance_loss_clip": 1.06391561, "balance_loss_mlp": 1.02444434, "epoch": 0.1262550351710455, "flos": 21470598965280.0, "grad_norm": 2.0359887548397886, "language_loss": 0.85151786, "learning_rate": 3.903695468395213e-06, "loss": 0.87416619, "num_input_tokens_seen": 22199390, "step": 1050, "time_per_iteration": 2.589484691619873 }, { "auxiliary_loss_clip": 0.01214136, "auxiliary_loss_mlp": 0.01042586, "balance_loss_clip": 1.05967569, "balance_loss_mlp": 1.03265619, "epoch": 0.1263752780616846, "flos": 31576723951680.0, "grad_norm": 1.9557919803179251, "language_loss": 0.55706489, "learning_rate": 3.903456513746434e-06, "loss": 0.57963216, "num_input_tokens_seen": 22220365, "step": 1051, "time_per_iteration": 2.697896718978882 }, { "auxiliary_loss_clip": 0.01235562, "auxiliary_loss_mlp": 0.01039255, "balance_loss_clip": 1.06522548, "balance_loss_mlp": 1.02899754, "epoch": 0.1264955209523237, "flos": 28768222244640.0, "grad_norm": 3.276738235787745, "language_loss": 0.87378109, "learning_rate": 3.903217270344815e-06, "loss": 0.89652932, "num_input_tokens_seen": 22240615, "step": 1052, "time_per_iteration": 2.5991532802581787 }, { "auxiliary_loss_clip": 0.01186404, "auxiliary_loss_mlp": 0.01038265, "balance_loss_clip": 1.0588212, "balance_loss_mlp": 1.02850163, "epoch": 0.12661576384296278, "flos": 29241129642720.0, "grad_norm": 1.918178068012639, "language_loss": 0.82524729, "learning_rate": 3.902977738226648e-06, "loss": 0.84749401, "num_input_tokens_seen": 22261350, "step": 1053, "time_per_iteration": 2.7039742469787598 }, { "auxiliary_loss_clip": 0.01227225, "auxiliary_loss_mlp": 0.01046128, "balance_loss_clip": 1.06537974, "balance_loss_mlp": 1.03604853, "epoch": 0.12673600673360189, "flos": 20850349833600.0, "grad_norm": 1.830398047669258, "language_loss": 0.91559291, "learning_rate": 3.902737917428273e-06, "loss": 0.93832648, "num_input_tokens_seen": 22279515, "step": 1054, "time_per_iteration": 2.570446014404297 }, { "auxiliary_loss_clip": 0.01239518, "auxiliary_loss_mlp": 0.01035715, "balance_loss_clip": 1.06523693, "balance_loss_mlp": 1.02627349, "epoch": 0.12685624962424097, "flos": 25263699396960.0, "grad_norm": 1.6193903132881742, "language_loss": 0.83912665, "learning_rate": 3.902497807986068e-06, "loss": 0.86187893, "num_input_tokens_seen": 22299535, "step": 1055, "time_per_iteration": 2.580256938934326 }, { "auxiliary_loss_clip": 0.01194432, "auxiliary_loss_mlp": 0.0103385, "balance_loss_clip": 1.05652177, "balance_loss_mlp": 1.02394998, "epoch": 0.12697649251488005, "flos": 27527113392960.0, "grad_norm": 5.393035403035226, "language_loss": 0.83744597, "learning_rate": 3.902257409936458e-06, "loss": 0.85972881, "num_input_tokens_seen": 22320300, "step": 1056, "time_per_iteration": 2.6985204219818115 }, { "auxiliary_loss_clip": 0.01212686, "auxiliary_loss_mlp": 0.0103698, "balance_loss_clip": 1.06570649, "balance_loss_mlp": 1.02744913, "epoch": 0.12709673540551916, "flos": 21251867091840.0, "grad_norm": 1.7188254370753917, "language_loss": 0.8386721, "learning_rate": 3.902016723315912e-06, "loss": 0.86116874, "num_input_tokens_seen": 22338240, "step": 1057, "time_per_iteration": 2.6108360290527344 }, { "auxiliary_loss_clip": 0.012174, "auxiliary_loss_mlp": 0.01040874, "balance_loss_clip": 1.05914021, "balance_loss_mlp": 1.03149271, "epoch": 0.12721697829615825, "flos": 25337711474880.0, "grad_norm": 1.8834622134932428, "language_loss": 0.69293731, "learning_rate": 3.901775748160941e-06, "loss": 0.71552002, "num_input_tokens_seen": 22357420, "step": 1058, "time_per_iteration": 3.4411516189575195 }, { "auxiliary_loss_clip": 0.01098065, "auxiliary_loss_mlp": 0.01004632, "balance_loss_clip": 1.0231111, "balance_loss_mlp": 1.00242639, "epoch": 0.12733722118679733, "flos": 61943299471200.0, "grad_norm": 0.8020808107323115, "language_loss": 0.60884726, "learning_rate": 3.901534484508101e-06, "loss": 0.62987423, "num_input_tokens_seen": 22420095, "step": 1059, "time_per_iteration": 3.228116035461426 }, { "auxiliary_loss_clip": 0.01197878, "auxiliary_loss_mlp": 0.01037446, "balance_loss_clip": 1.05979872, "balance_loss_mlp": 1.02758741, "epoch": 0.1274574640774364, "flos": 26976745888800.0, "grad_norm": 1.8293306376681742, "language_loss": 0.74568379, "learning_rate": 3.901292932393991e-06, "loss": 0.76803702, "num_input_tokens_seen": 22438975, "step": 1060, "time_per_iteration": 2.643662929534912 }, { "auxiliary_loss_clip": 0.01240739, "auxiliary_loss_mlp": 0.01041182, "balance_loss_clip": 1.06830907, "balance_loss_mlp": 1.03050053, "epoch": 0.12757770696807552, "flos": 22236322148640.0, "grad_norm": 2.794665196377321, "language_loss": 0.85393542, "learning_rate": 3.9010510918552555e-06, "loss": 0.87675464, "num_input_tokens_seen": 22458050, "step": 1061, "time_per_iteration": 3.4046311378479004 }, { "auxiliary_loss_clip": 0.01205164, "auxiliary_loss_mlp": 0.0104451, "balance_loss_clip": 1.06048226, "balance_loss_mlp": 1.03362608, "epoch": 0.1276979498587146, "flos": 28547910024960.0, "grad_norm": 2.226582776859783, "language_loss": 0.74175787, "learning_rate": 3.900808962928581e-06, "loss": 0.76425457, "num_input_tokens_seen": 22475665, "step": 1062, "time_per_iteration": 3.5354795455932617 }, { "auxiliary_loss_clip": 0.01234634, "auxiliary_loss_mlp": 0.01040955, "balance_loss_clip": 1.06411028, "balance_loss_mlp": 1.03055406, "epoch": 0.1278181927493537, "flos": 17420341901280.0, "grad_norm": 2.0960175799093266, "language_loss": 0.88873851, "learning_rate": 3.900566545650698e-06, "loss": 0.91149443, "num_input_tokens_seen": 22493335, "step": 1063, "time_per_iteration": 2.5226306915283203 }, { "auxiliary_loss_clip": 0.01223218, "auxiliary_loss_mlp": 0.01035964, "balance_loss_clip": 1.06543779, "balance_loss_mlp": 1.02527142, "epoch": 0.1279384356399928, "flos": 21138640081920.0, "grad_norm": 2.04726395401181, "language_loss": 0.82062471, "learning_rate": 3.900323840058381e-06, "loss": 0.84321654, "num_input_tokens_seen": 22511045, "step": 1064, "time_per_iteration": 2.5743496417999268 }, { "auxiliary_loss_clip": 0.01222201, "auxiliary_loss_mlp": 0.01033429, "balance_loss_clip": 1.06064785, "balance_loss_mlp": 1.02413654, "epoch": 0.12805867853063188, "flos": 26576737142880.0, "grad_norm": 1.701711028546204, "language_loss": 0.82021415, "learning_rate": 3.900080846188449e-06, "loss": 0.8427704, "num_input_tokens_seen": 22529635, "step": 1065, "time_per_iteration": 2.601241111755371 }, { "auxiliary_loss_clip": 0.01236945, "auxiliary_loss_mlp": 0.01038326, "balance_loss_clip": 1.06328547, "balance_loss_mlp": 1.02884293, "epoch": 0.12817892142127096, "flos": 16436210097120.0, "grad_norm": 1.9641687232649985, "language_loss": 0.81318337, "learning_rate": 3.8998375640777625e-06, "loss": 0.83593607, "num_input_tokens_seen": 22547505, "step": 1066, "time_per_iteration": 2.5644633769989014 }, { "auxiliary_loss_clip": 0.0110272, "auxiliary_loss_mlp": 0.01000925, "balance_loss_clip": 1.02237797, "balance_loss_mlp": 0.99882692, "epoch": 0.12829916431191005, "flos": 60757058957760.0, "grad_norm": 0.7040938776021135, "language_loss": 0.5263797, "learning_rate": 3.899593993763229e-06, "loss": 0.54741621, "num_input_tokens_seen": 22608465, "step": 1067, "time_per_iteration": 3.114388942718506 }, { "auxiliary_loss_clip": 0.01182151, "auxiliary_loss_mlp": 0.01036918, "balance_loss_clip": 1.05721509, "balance_loss_mlp": 1.02587295, "epoch": 0.12841940720254916, "flos": 29786899776000.0, "grad_norm": 2.3310725771180385, "language_loss": 0.81820327, "learning_rate": 3.899350135281796e-06, "loss": 0.84039402, "num_input_tokens_seen": 22629465, "step": 1068, "time_per_iteration": 2.7113826274871826 }, { "auxiliary_loss_clip": 0.01193699, "auxiliary_loss_mlp": 0.01041932, "balance_loss_clip": 1.05928326, "balance_loss_mlp": 1.03233004, "epoch": 0.12853965009318824, "flos": 25951854723360.0, "grad_norm": 1.9430855088256658, "language_loss": 0.79775393, "learning_rate": 3.8991059886704585e-06, "loss": 0.82011026, "num_input_tokens_seen": 22648970, "step": 1069, "time_per_iteration": 2.6799185276031494 }, { "auxiliary_loss_clip": 0.01184163, "auxiliary_loss_mlp": 0.01046176, "balance_loss_clip": 1.05688965, "balance_loss_mlp": 1.03539932, "epoch": 0.12865989298382732, "flos": 30846875423040.0, "grad_norm": 2.1208562643134283, "language_loss": 0.82745922, "learning_rate": 3.898861553966252e-06, "loss": 0.84976262, "num_input_tokens_seen": 22668620, "step": 1070, "time_per_iteration": 2.690791130065918 }, { "auxiliary_loss_clip": 0.01146743, "auxiliary_loss_mlp": 0.0104225, "balance_loss_clip": 1.05306756, "balance_loss_mlp": 1.03311825, "epoch": 0.12878013587446643, "flos": 25885780293600.0, "grad_norm": 1.62324700832848, "language_loss": 0.88168252, "learning_rate": 3.898616831206257e-06, "loss": 0.90357244, "num_input_tokens_seen": 22689045, "step": 1071, "time_per_iteration": 2.7965850830078125 }, { "auxiliary_loss_clip": 0.01184462, "auxiliary_loss_mlp": 0.01043992, "balance_loss_clip": 1.05458689, "balance_loss_mlp": 1.03359675, "epoch": 0.12890037876510552, "flos": 23333142208320.0, "grad_norm": 1.948020317952066, "language_loss": 0.76770663, "learning_rate": 3.8983718204276e-06, "loss": 0.7899912, "num_input_tokens_seen": 22711265, "step": 1072, "time_per_iteration": 2.7039318084716797 }, { "auxiliary_loss_clip": 0.01204971, "auxiliary_loss_mlp": 0.01035319, "balance_loss_clip": 1.06119215, "balance_loss_mlp": 1.02476311, "epoch": 0.1290206216557446, "flos": 23587245899040.0, "grad_norm": 1.6198741518952386, "language_loss": 0.82360399, "learning_rate": 3.898126521667446e-06, "loss": 0.84600687, "num_input_tokens_seen": 22731420, "step": 1073, "time_per_iteration": 2.6523313522338867 }, { "auxiliary_loss_clip": 0.01221578, "auxiliary_loss_mlp": 0.01038715, "balance_loss_clip": 1.0624975, "balance_loss_mlp": 1.02911294, "epoch": 0.12914086454638368, "flos": 24170614701120.0, "grad_norm": 1.6023414875658106, "language_loss": 0.83098966, "learning_rate": 3.897880934963007e-06, "loss": 0.85359257, "num_input_tokens_seen": 22750970, "step": 1074, "time_per_iteration": 2.614434242248535 }, { "auxiliary_loss_clip": 0.01200986, "auxiliary_loss_mlp": 0.01038089, "balance_loss_clip": 1.05680013, "balance_loss_mlp": 1.02787316, "epoch": 0.1292611074370228, "flos": 20267160616320.0, "grad_norm": 2.047547119262316, "language_loss": 0.78760725, "learning_rate": 3.89763506035154e-06, "loss": 0.80999804, "num_input_tokens_seen": 22768820, "step": 1075, "time_per_iteration": 2.6199114322662354 }, { "auxiliary_loss_clip": 0.01210744, "auxiliary_loss_mlp": 0.01037447, "balance_loss_clip": 1.06112862, "balance_loss_mlp": 1.02703428, "epoch": 0.12938135032766188, "flos": 27377688475680.0, "grad_norm": 1.8149818270706557, "language_loss": 0.81602484, "learning_rate": 3.897388897870343e-06, "loss": 0.83850682, "num_input_tokens_seen": 22789460, "step": 1076, "time_per_iteration": 2.622260332107544 }, { "auxiliary_loss_clip": 0.01218907, "auxiliary_loss_mlp": 0.01038917, "balance_loss_clip": 1.06087303, "balance_loss_mlp": 1.02795529, "epoch": 0.12950159321830096, "flos": 29277112048320.0, "grad_norm": 1.7445560632857235, "language_loss": 0.74559426, "learning_rate": 3.89714244755676e-06, "loss": 0.7681725, "num_input_tokens_seen": 22810820, "step": 1077, "time_per_iteration": 2.7071852684020996 }, { "auxiliary_loss_clip": 0.0115963, "auxiliary_loss_mlp": 0.01034342, "balance_loss_clip": 1.04946208, "balance_loss_mlp": 1.0236485, "epoch": 0.12962183610894007, "flos": 24534892460160.0, "grad_norm": 2.3250350968652578, "language_loss": 0.86196899, "learning_rate": 3.896895709448175e-06, "loss": 0.88390875, "num_input_tokens_seen": 22830570, "step": 1078, "time_per_iteration": 2.6855568885803223 }, { "auxiliary_loss_clip": 0.01154515, "auxiliary_loss_mlp": 0.01042712, "balance_loss_clip": 1.05089521, "balance_loss_mlp": 1.03163719, "epoch": 0.12974207899957915, "flos": 11215946985600.0, "grad_norm": 2.3635093617109684, "language_loss": 0.76575553, "learning_rate": 3.896648683582019e-06, "loss": 0.78772777, "num_input_tokens_seen": 22845905, "step": 1079, "time_per_iteration": 2.706712484359741 }, { "auxiliary_loss_clip": 0.01171857, "auxiliary_loss_mlp": 0.01035867, "balance_loss_clip": 1.05826187, "balance_loss_mlp": 1.02578783, "epoch": 0.12986232189021824, "flos": 24717893346720.0, "grad_norm": 1.9267619455776894, "language_loss": 0.8055172, "learning_rate": 3.896401369995766e-06, "loss": 0.8275944, "num_input_tokens_seen": 22865710, "step": 1080, "time_per_iteration": 2.6808230876922607 }, { "auxiliary_loss_clip": 0.01236089, "auxiliary_loss_mlp": 0.01038942, "balance_loss_clip": 1.06501389, "balance_loss_mlp": 1.02879119, "epoch": 0.12998256478085732, "flos": 23915361667680.0, "grad_norm": 1.9632540712117217, "language_loss": 0.79364443, "learning_rate": 3.896153768726932e-06, "loss": 0.81639475, "num_input_tokens_seen": 22886020, "step": 1081, "time_per_iteration": 2.58870005607605 }, { "auxiliary_loss_clip": 0.01222975, "auxiliary_loss_mlp": 0.01037774, "balance_loss_clip": 1.06554174, "balance_loss_mlp": 1.02747393, "epoch": 0.13010280767149643, "flos": 18624211253760.0, "grad_norm": 2.4371893902067225, "language_loss": 0.88204765, "learning_rate": 3.8959058798130806e-06, "loss": 0.90465504, "num_input_tokens_seen": 22903995, "step": 1082, "time_per_iteration": 2.566941738128662 }, { "auxiliary_loss_clip": 0.01206629, "auxiliary_loss_mlp": 0.00765998, "balance_loss_clip": 1.06207108, "balance_loss_mlp": 1.00019598, "epoch": 0.1302230505621355, "flos": 22783995880800.0, "grad_norm": 1.6253004285514001, "language_loss": 0.75131786, "learning_rate": 3.895657703291814e-06, "loss": 0.77104414, "num_input_tokens_seen": 22924100, "step": 1083, "time_per_iteration": 3.3921191692352295 }, { "auxiliary_loss_clip": 0.01218324, "auxiliary_loss_mlp": 0.01035547, "balance_loss_clip": 1.06278837, "balance_loss_mlp": 1.02509177, "epoch": 0.1303432934527746, "flos": 21323616401280.0, "grad_norm": 2.812845786640293, "language_loss": 0.7944352, "learning_rate": 3.895409239200781e-06, "loss": 0.81697392, "num_input_tokens_seen": 22939985, "step": 1084, "time_per_iteration": 3.371481418609619 }, { "auxiliary_loss_clip": 0.01216099, "auxiliary_loss_mlp": 0.0103955, "balance_loss_clip": 1.06150091, "balance_loss_mlp": 1.02895808, "epoch": 0.1304635363434137, "flos": 20922494229600.0, "grad_norm": 2.9833668177326595, "language_loss": 0.91407645, "learning_rate": 3.895160487577673e-06, "loss": 0.93663287, "num_input_tokens_seen": 22957555, "step": 1085, "time_per_iteration": 2.5861222743988037 }, { "auxiliary_loss_clip": 0.01116523, "auxiliary_loss_mlp": 0.01003262, "balance_loss_clip": 1.02564478, "balance_loss_mlp": 1.00087786, "epoch": 0.1305837792340528, "flos": 63245669880000.0, "grad_norm": 0.7845969481347637, "language_loss": 0.60913652, "learning_rate": 3.894911448460226e-06, "loss": 0.63033432, "num_input_tokens_seen": 23016870, "step": 1086, "time_per_iteration": 3.0431196689605713 }, { "auxiliary_loss_clip": 0.01122168, "auxiliary_loss_mlp": 0.01032968, "balance_loss_clip": 1.04987764, "balance_loss_mlp": 1.02323413, "epoch": 0.13070402212469187, "flos": 26428856654880.0, "grad_norm": 1.950605435461352, "language_loss": 0.72676229, "learning_rate": 3.8946621218862195e-06, "loss": 0.74831367, "num_input_tokens_seen": 23037870, "step": 1087, "time_per_iteration": 3.718059539794922 }, { "auxiliary_loss_clip": 0.01187461, "auxiliary_loss_mlp": 0.01037688, "balance_loss_clip": 1.05923963, "balance_loss_mlp": 1.02753782, "epoch": 0.13082426501533098, "flos": 27673413534720.0, "grad_norm": 1.7284955338657393, "language_loss": 0.88850862, "learning_rate": 3.894412507893475e-06, "loss": 0.91076005, "num_input_tokens_seen": 23058150, "step": 1088, "time_per_iteration": 3.610281467437744 }, { "auxiliary_loss_clip": 0.01180162, "auxiliary_loss_mlp": 0.01039116, "balance_loss_clip": 1.05732441, "balance_loss_mlp": 1.0290724, "epoch": 0.13094450790597006, "flos": 24826774404480.0, "grad_norm": 3.66834425107297, "language_loss": 0.71939695, "learning_rate": 3.894162606519859e-06, "loss": 0.74158967, "num_input_tokens_seen": 23077100, "step": 1089, "time_per_iteration": 2.7391226291656494 }, { "auxiliary_loss_clip": 0.01175063, "auxiliary_loss_mlp": 0.01035495, "balance_loss_clip": 1.05869043, "balance_loss_mlp": 1.02542734, "epoch": 0.13106475079660915, "flos": 19062608841600.0, "grad_norm": 1.8071876267345275, "language_loss": 0.76899552, "learning_rate": 3.893912417803282e-06, "loss": 0.7911011, "num_input_tokens_seen": 23096815, "step": 1090, "time_per_iteration": 2.6997523307800293 }, { "auxiliary_loss_clip": 0.01171525, "auxiliary_loss_mlp": 0.01039311, "balance_loss_clip": 1.05178082, "balance_loss_mlp": 1.02926171, "epoch": 0.13118499368724823, "flos": 28913193458880.0, "grad_norm": 1.958355188412345, "language_loss": 0.76770604, "learning_rate": 3.8936619417816975e-06, "loss": 0.78981435, "num_input_tokens_seen": 23117145, "step": 1091, "time_per_iteration": 2.744126796722412 }, { "auxiliary_loss_clip": 0.01189097, "auxiliary_loss_mlp": 0.01042883, "balance_loss_clip": 1.0611968, "balance_loss_mlp": 1.03273237, "epoch": 0.13130523657788734, "flos": 14283401172960.0, "grad_norm": 1.7576413599593592, "language_loss": 0.71756554, "learning_rate": 3.8934111784931015e-06, "loss": 0.73988533, "num_input_tokens_seen": 23134595, "step": 1092, "time_per_iteration": 2.6496078968048096 }, { "auxiliary_loss_clip": 0.01104623, "auxiliary_loss_mlp": 0.0100428, "balance_loss_clip": 1.02254593, "balance_loss_mlp": 1.00196767, "epoch": 0.13142547946852642, "flos": 70174166610720.0, "grad_norm": 0.9258176637912348, "language_loss": 0.59155393, "learning_rate": 3.893160127975535e-06, "loss": 0.61264288, "num_input_tokens_seen": 23195285, "step": 1093, "time_per_iteration": 3.303156614303589 }, { "auxiliary_loss_clip": 0.01182517, "auxiliary_loss_mlp": 0.01042705, "balance_loss_clip": 1.05715632, "balance_loss_mlp": 1.03347802, "epoch": 0.1315457223591655, "flos": 45806004654240.0, "grad_norm": 2.280346660217006, "language_loss": 0.81282264, "learning_rate": 3.8929087902670826e-06, "loss": 0.83507478, "num_input_tokens_seen": 23216915, "step": 1094, "time_per_iteration": 2.9146225452423096 }, { "auxiliary_loss_clip": 0.01118966, "auxiliary_loss_mlp": 0.01002289, "balance_loss_clip": 1.02258825, "balance_loss_mlp": 0.99998802, "epoch": 0.13166596524980462, "flos": 62881176619200.0, "grad_norm": 0.9395586151355468, "language_loss": 0.60812622, "learning_rate": 3.8926571654058715e-06, "loss": 0.62933874, "num_input_tokens_seen": 23273560, "step": 1095, "time_per_iteration": 3.090149164199829 }, { "auxiliary_loss_clip": 0.01189019, "auxiliary_loss_mlp": 0.01034675, "balance_loss_clip": 1.06040573, "balance_loss_mlp": 1.02528095, "epoch": 0.1317862081404437, "flos": 23586527559840.0, "grad_norm": 2.117192135998399, "language_loss": 0.76870179, "learning_rate": 3.892405253430074e-06, "loss": 0.79093874, "num_input_tokens_seen": 23291080, "step": 1096, "time_per_iteration": 2.647686719894409 }, { "auxiliary_loss_clip": 0.01209358, "auxiliary_loss_mlp": 0.00765586, "balance_loss_clip": 1.0622772, "balance_loss_mlp": 1.00024831, "epoch": 0.13190645103108278, "flos": 20260767397440.0, "grad_norm": 1.9289554805451314, "language_loss": 0.82200718, "learning_rate": 3.892153054377904e-06, "loss": 0.84175658, "num_input_tokens_seen": 23308485, "step": 1097, "time_per_iteration": 2.6158697605133057 }, { "auxiliary_loss_clip": 0.01069622, "auxiliary_loss_mlp": 0.01043816, "balance_loss_clip": 1.03258288, "balance_loss_mlp": 1.04195595, "epoch": 0.13202669392172187, "flos": 53455448895840.0, "grad_norm": 0.9385863905471425, "language_loss": 0.59441268, "learning_rate": 3.891900568287619e-06, "loss": 0.61554706, "num_input_tokens_seen": 23360870, "step": 1098, "time_per_iteration": 3.0868000984191895 }, { "auxiliary_loss_clip": 0.01193477, "auxiliary_loss_mlp": 0.0104042, "balance_loss_clip": 1.05993915, "balance_loss_mlp": 1.02950692, "epoch": 0.13214693681236098, "flos": 15851296865760.0, "grad_norm": 2.4455173241437116, "language_loss": 0.72421002, "learning_rate": 3.891647795197523e-06, "loss": 0.74654901, "num_input_tokens_seen": 23376910, "step": 1099, "time_per_iteration": 2.6170380115509033 }, { "auxiliary_loss_clip": 0.01195599, "auxiliary_loss_mlp": 0.01041383, "balance_loss_clip": 1.05590034, "balance_loss_mlp": 1.02992105, "epoch": 0.13226717970300006, "flos": 19353844280640.0, "grad_norm": 1.9565912312997955, "language_loss": 0.68353438, "learning_rate": 3.8913947351459605e-06, "loss": 0.70590419, "num_input_tokens_seen": 23394450, "step": 1100, "time_per_iteration": 2.6312756538391113 }, { "auxiliary_loss_clip": 0.01236724, "auxiliary_loss_mlp": 0.01043874, "balance_loss_clip": 1.06633747, "balance_loss_mlp": 1.03415871, "epoch": 0.13238742259363914, "flos": 20698087476480.0, "grad_norm": 1.9396350515361234, "language_loss": 0.67656255, "learning_rate": 3.89114138817132e-06, "loss": 0.69936854, "num_input_tokens_seen": 23411115, "step": 1101, "time_per_iteration": 2.5438315868377686 }, { "auxiliary_loss_clip": 0.0122144, "auxiliary_loss_mlp": 0.0103894, "balance_loss_clip": 1.06573224, "balance_loss_mlp": 1.02982688, "epoch": 0.13250766548427825, "flos": 21032452796160.0, "grad_norm": 1.9237346141249962, "language_loss": 0.84457743, "learning_rate": 3.890887754312035e-06, "loss": 0.86718118, "num_input_tokens_seen": 23429360, "step": 1102, "time_per_iteration": 2.566643238067627 }, { "auxiliary_loss_clip": 0.01195369, "auxiliary_loss_mlp": 0.01039218, "balance_loss_clip": 1.05602396, "balance_loss_mlp": 1.02936506, "epoch": 0.13262790837491734, "flos": 22637875323840.0, "grad_norm": 1.747832183085858, "language_loss": 0.87925458, "learning_rate": 3.890633833606581e-06, "loss": 0.90160042, "num_input_tokens_seen": 23449050, "step": 1103, "time_per_iteration": 2.609272003173828 }, { "auxiliary_loss_clip": 0.0121655, "auxiliary_loss_mlp": 0.01033502, "balance_loss_clip": 1.06463814, "balance_loss_mlp": 1.02383983, "epoch": 0.13274815126555642, "flos": 19683145308960.0, "grad_norm": 1.7154459383246448, "language_loss": 0.69381881, "learning_rate": 3.890379626093477e-06, "loss": 0.71631932, "num_input_tokens_seen": 23468800, "step": 1104, "time_per_iteration": 2.5422122478485107 }, { "auxiliary_loss_clip": 0.01160076, "auxiliary_loss_mlp": 0.01034542, "balance_loss_clip": 1.0546031, "balance_loss_mlp": 1.0242002, "epoch": 0.1328683941561955, "flos": 21317690102880.0, "grad_norm": 2.2932241770757043, "language_loss": 0.92749727, "learning_rate": 3.890125131811287e-06, "loss": 0.9494434, "num_input_tokens_seen": 23486850, "step": 1105, "time_per_iteration": 2.6650638580322266 }, { "auxiliary_loss_clip": 0.01187928, "auxiliary_loss_mlp": 0.01035231, "balance_loss_clip": 1.05177569, "balance_loss_mlp": 1.02564085, "epoch": 0.1329886370468346, "flos": 13699134446880.0, "grad_norm": 2.2655792510268227, "language_loss": 0.75271881, "learning_rate": 3.889870350798618e-06, "loss": 0.77495044, "num_input_tokens_seen": 23504195, "step": 1106, "time_per_iteration": 2.56547212600708 }, { "auxiliary_loss_clip": 0.01232224, "auxiliary_loss_mlp": 0.01036625, "balance_loss_clip": 1.06182241, "balance_loss_mlp": 1.02763653, "epoch": 0.1331088799374737, "flos": 21032416879200.0, "grad_norm": 1.6502842169457421, "language_loss": 0.78763241, "learning_rate": 3.889615283094119e-06, "loss": 0.81032091, "num_input_tokens_seen": 23523385, "step": 1107, "time_per_iteration": 2.5534982681274414 }, { "auxiliary_loss_clip": 0.01237869, "auxiliary_loss_mlp": 0.01040442, "balance_loss_clip": 1.06382096, "balance_loss_mlp": 1.0299511, "epoch": 0.13322912282811278, "flos": 18260436332160.0, "grad_norm": 1.9340126427984856, "language_loss": 0.84768403, "learning_rate": 3.889359928736485e-06, "loss": 0.87046707, "num_input_tokens_seen": 23541330, "step": 1108, "time_per_iteration": 2.5094785690307617 }, { "auxiliary_loss_clip": 0.01198964, "auxiliary_loss_mlp": 0.00765669, "balance_loss_clip": 1.06084037, "balance_loss_mlp": 1.00030267, "epoch": 0.1333493657187519, "flos": 24460880382240.0, "grad_norm": 2.106438455995801, "language_loss": 0.91313291, "learning_rate": 3.889104287764451e-06, "loss": 0.93277919, "num_input_tokens_seen": 23561705, "step": 1109, "time_per_iteration": 3.3901479244232178 }, { "auxiliary_loss_clip": 0.0120129, "auxiliary_loss_mlp": 0.01037553, "balance_loss_clip": 1.06256783, "balance_loss_mlp": 1.02727163, "epoch": 0.13346960860939097, "flos": 22158933876480.0, "grad_norm": 2.3578141061688696, "language_loss": 0.90789545, "learning_rate": 3.888848360216798e-06, "loss": 0.9302839, "num_input_tokens_seen": 23579350, "step": 1110, "time_per_iteration": 3.386291980743408 }, { "auxiliary_loss_clip": 0.01108195, "auxiliary_loss_mlp": 0.01023219, "balance_loss_clip": 1.02196872, "balance_loss_mlp": 1.02075088, "epoch": 0.13358985150003005, "flos": 67931225281920.0, "grad_norm": 0.8076919692214833, "language_loss": 0.56645882, "learning_rate": 3.888592146132351e-06, "loss": 0.58777297, "num_input_tokens_seen": 23640620, "step": 1111, "time_per_iteration": 3.2499804496765137 }, { "auxiliary_loss_clip": 0.01219395, "auxiliary_loss_mlp": 0.01039132, "balance_loss_clip": 1.06429255, "balance_loss_mlp": 1.02949357, "epoch": 0.13371009439066917, "flos": 26834288861760.0, "grad_norm": 1.9368958800084097, "language_loss": 0.7847898, "learning_rate": 3.888335645549978e-06, "loss": 0.80737507, "num_input_tokens_seen": 23661040, "step": 1112, "time_per_iteration": 2.5789296627044678 }, { "auxiliary_loss_clip": 0.01237705, "auxiliary_loss_mlp": 0.01036116, "balance_loss_clip": 1.06667066, "balance_loss_mlp": 1.02591801, "epoch": 0.13383033728130825, "flos": 26322848953920.0, "grad_norm": 2.8286011358492695, "language_loss": 0.81699169, "learning_rate": 3.888078858508588e-06, "loss": 0.83972991, "num_input_tokens_seen": 23680900, "step": 1113, "time_per_iteration": 3.4789812564849854 }, { "auxiliary_loss_clip": 0.01201972, "auxiliary_loss_mlp": 0.01035139, "balance_loss_clip": 1.06430566, "balance_loss_mlp": 1.02537012, "epoch": 0.13395058017194733, "flos": 22563935079840.0, "grad_norm": 1.7968443914111292, "language_loss": 0.84352893, "learning_rate": 3.8878217850471365e-06, "loss": 0.86590004, "num_input_tokens_seen": 23700815, "step": 1114, "time_per_iteration": 2.5934200286865234 }, { "auxiliary_loss_clip": 0.01243528, "auxiliary_loss_mlp": 0.01028753, "balance_loss_clip": 1.06919491, "balance_loss_mlp": 1.01888251, "epoch": 0.13407082306258641, "flos": 25810942125600.0, "grad_norm": 1.733472028660315, "language_loss": 0.74074996, "learning_rate": 3.887564425204621e-06, "loss": 0.76347274, "num_input_tokens_seen": 23722500, "step": 1115, "time_per_iteration": 3.5741803646087646 }, { "auxiliary_loss_clip": 0.01087314, "auxiliary_loss_mlp": 0.01005066, "balance_loss_clip": 1.02241111, "balance_loss_mlp": 1.00264645, "epoch": 0.13419106595322552, "flos": 68338375114560.0, "grad_norm": 0.8391578211284328, "language_loss": 0.54711455, "learning_rate": 3.887306779020083e-06, "loss": 0.56803834, "num_input_tokens_seen": 23777155, "step": 1116, "time_per_iteration": 3.1509084701538086 }, { "auxiliary_loss_clip": 0.01224118, "auxiliary_loss_mlp": 0.01041692, "balance_loss_clip": 1.06489778, "balance_loss_mlp": 1.03251278, "epoch": 0.1343113088438646, "flos": 20449083994080.0, "grad_norm": 2.1010438258680106, "language_loss": 0.70259261, "learning_rate": 3.887048846532608e-06, "loss": 0.72525072, "num_input_tokens_seen": 23794130, "step": 1117, "time_per_iteration": 2.5625085830688477 }, { "auxiliary_loss_clip": 0.0108992, "auxiliary_loss_mlp": 0.01003033, "balance_loss_clip": 1.02078247, "balance_loss_mlp": 1.0007323, "epoch": 0.1344315517345037, "flos": 67389794712480.0, "grad_norm": 0.7600696293303641, "language_loss": 0.581258, "learning_rate": 3.8867906277813224e-06, "loss": 0.60218757, "num_input_tokens_seen": 23852285, "step": 1118, "time_per_iteration": 3.1078336238861084 }, { "auxiliary_loss_clip": 0.01223774, "auxiliary_loss_mlp": 0.00765173, "balance_loss_clip": 1.06404424, "balance_loss_mlp": 1.00028265, "epoch": 0.1345517946251428, "flos": 40734448120800.0, "grad_norm": 2.2221914631256348, "language_loss": 0.73743713, "learning_rate": 3.886532122805399e-06, "loss": 0.75732666, "num_input_tokens_seen": 23874765, "step": 1119, "time_per_iteration": 2.7376887798309326 }, { "auxiliary_loss_clip": 0.01144366, "auxiliary_loss_mlp": 0.01048101, "balance_loss_clip": 1.05187559, "balance_loss_mlp": 1.03779507, "epoch": 0.13467203751578188, "flos": 22816853510880.0, "grad_norm": 2.468642072678062, "language_loss": 0.89534271, "learning_rate": 3.886273331644053e-06, "loss": 0.91726732, "num_input_tokens_seen": 23893635, "step": 1120, "time_per_iteration": 2.7895686626434326 }, { "auxiliary_loss_clip": 0.01168459, "auxiliary_loss_mlp": 0.01044396, "balance_loss_clip": 1.05743325, "balance_loss_mlp": 1.03527021, "epoch": 0.13479228040642097, "flos": 17091579627360.0, "grad_norm": 2.0460948894518927, "language_loss": 0.82492602, "learning_rate": 3.886014254336542e-06, "loss": 0.8470546, "num_input_tokens_seen": 23910110, "step": 1121, "time_per_iteration": 2.6420915126800537 }, { "auxiliary_loss_clip": 0.01214374, "auxiliary_loss_mlp": 0.01039673, "balance_loss_clip": 1.06088746, "balance_loss_mlp": 1.03103662, "epoch": 0.13491252329706005, "flos": 23730529016160.0, "grad_norm": 1.673763740578209, "language_loss": 0.92253304, "learning_rate": 3.885754890922168e-06, "loss": 0.94507349, "num_input_tokens_seen": 23930440, "step": 1122, "time_per_iteration": 2.630983829498291 }, { "auxiliary_loss_clip": 0.0113343, "auxiliary_loss_mlp": 0.010473, "balance_loss_clip": 1.05183864, "balance_loss_mlp": 1.03779268, "epoch": 0.13503276618769916, "flos": 34127063351520.0, "grad_norm": 1.9265433897450437, "language_loss": 0.78621, "learning_rate": 3.885495241440277e-06, "loss": 0.80801731, "num_input_tokens_seen": 23954535, "step": 1123, "time_per_iteration": 2.8174562454223633 }, { "auxiliary_loss_clip": 0.0123627, "auxiliary_loss_mlp": 0.01041482, "balance_loss_clip": 1.06452155, "balance_loss_mlp": 1.0318315, "epoch": 0.13515300907833824, "flos": 17712331596480.0, "grad_norm": 1.7323632165378702, "language_loss": 0.74163544, "learning_rate": 3.885235305930257e-06, "loss": 0.76441294, "num_input_tokens_seen": 23972735, "step": 1124, "time_per_iteration": 2.546643018722534 }, { "auxiliary_loss_clip": 0.01183127, "auxiliary_loss_mlp": 0.01036869, "balance_loss_clip": 1.05936742, "balance_loss_mlp": 1.02676046, "epoch": 0.13527325196897733, "flos": 20260875148320.0, "grad_norm": 1.861359450547799, "language_loss": 0.85306728, "learning_rate": 3.884975084431539e-06, "loss": 0.87526721, "num_input_tokens_seen": 23987685, "step": 1125, "time_per_iteration": 2.6146321296691895 }, { "auxiliary_loss_clip": 0.01206139, "auxiliary_loss_mlp": 0.007647, "balance_loss_clip": 1.05836701, "balance_loss_mlp": 1.00022674, "epoch": 0.13539349485961644, "flos": 18186496088160.0, "grad_norm": 2.415590825669004, "language_loss": 0.92061478, "learning_rate": 3.8847145769836e-06, "loss": 0.94032317, "num_input_tokens_seen": 24004105, "step": 1126, "time_per_iteration": 2.568962812423706 }, { "auxiliary_loss_clip": 0.01239089, "auxiliary_loss_mlp": 0.01046963, "balance_loss_clip": 1.06606591, "balance_loss_mlp": 1.0367527, "epoch": 0.13551373775025552, "flos": 19317466788480.0, "grad_norm": 2.182583793659897, "language_loss": 0.66147459, "learning_rate": 3.884453783625959e-06, "loss": 0.68433511, "num_input_tokens_seen": 24021715, "step": 1127, "time_per_iteration": 2.5010452270507812 }, { "auxiliary_loss_clip": 0.01197608, "auxiliary_loss_mlp": 0.01035483, "balance_loss_clip": 1.06047332, "balance_loss_mlp": 1.02660823, "epoch": 0.1356339806408946, "flos": 20850816754080.0, "grad_norm": 3.2370223546998513, "language_loss": 0.84494579, "learning_rate": 3.884192704398176e-06, "loss": 0.86727673, "num_input_tokens_seen": 24038915, "step": 1128, "time_per_iteration": 2.6096723079681396 }, { "auxiliary_loss_clip": 0.01220151, "auxiliary_loss_mlp": 0.0103285, "balance_loss_clip": 1.06202364, "balance_loss_mlp": 1.02400446, "epoch": 0.13575422353153369, "flos": 50476043929440.0, "grad_norm": 2.5431525444555634, "language_loss": 0.74317795, "learning_rate": 3.883931339339858e-06, "loss": 0.76570797, "num_input_tokens_seen": 24063300, "step": 1129, "time_per_iteration": 2.807321786880493 }, { "auxiliary_loss_clip": 0.01227392, "auxiliary_loss_mlp": 0.01035634, "balance_loss_clip": 1.06439197, "balance_loss_mlp": 1.02487564, "epoch": 0.1358744664221728, "flos": 18150801018240.0, "grad_norm": 1.727858421335935, "language_loss": 0.78342152, "learning_rate": 3.883669688490654e-06, "loss": 0.80605173, "num_input_tokens_seen": 24081070, "step": 1130, "time_per_iteration": 2.5856540203094482 }, { "auxiliary_loss_clip": 0.01188488, "auxiliary_loss_mlp": 0.00764586, "balance_loss_clip": 1.05489182, "balance_loss_mlp": 1.00028133, "epoch": 0.13599470931281188, "flos": 18442970298240.0, "grad_norm": 1.906889937568082, "language_loss": 0.85848665, "learning_rate": 3.883407751890256e-06, "loss": 0.87801737, "num_input_tokens_seen": 24099675, "step": 1131, "time_per_iteration": 2.5764784812927246 }, { "auxiliary_loss_clip": 0.01184366, "auxiliary_loss_mlp": 0.01041009, "balance_loss_clip": 1.05520272, "balance_loss_mlp": 1.03089976, "epoch": 0.13611495220345096, "flos": 26680769411040.0, "grad_norm": 2.001747511484287, "language_loss": 0.85848039, "learning_rate": 3.8831455295783994e-06, "loss": 0.88073409, "num_input_tokens_seen": 24118925, "step": 1132, "time_per_iteration": 2.7231576442718506 }, { "auxiliary_loss_clip": 0.01195034, "auxiliary_loss_mlp": 0.01035555, "balance_loss_clip": 1.05797172, "balance_loss_mlp": 1.02496362, "epoch": 0.13623519509409007, "flos": 21686241980160.0, "grad_norm": 1.799196190699074, "language_loss": 0.74272072, "learning_rate": 3.882883021594864e-06, "loss": 0.76502669, "num_input_tokens_seen": 24137065, "step": 1133, "time_per_iteration": 2.608921766281128 }, { "auxiliary_loss_clip": 0.01180214, "auxiliary_loss_mlp": 0.01043009, "balance_loss_clip": 1.0579778, "balance_loss_mlp": 1.033687, "epoch": 0.13635543798472916, "flos": 14830392482880.0, "grad_norm": 2.0109164150216907, "language_loss": 0.86526287, "learning_rate": 3.8826202279794705e-06, "loss": 0.8874951, "num_input_tokens_seen": 24154125, "step": 1134, "time_per_iteration": 2.6391279697418213 }, { "auxiliary_loss_clip": 0.01240321, "auxiliary_loss_mlp": 0.01038937, "balance_loss_clip": 1.06924915, "balance_loss_mlp": 1.02882862, "epoch": 0.13647568087536824, "flos": 22890326834400.0, "grad_norm": 2.340028919718094, "language_loss": 0.70345747, "learning_rate": 3.882357148772085e-06, "loss": 0.72625005, "num_input_tokens_seen": 24171550, "step": 1135, "time_per_iteration": 3.308788537979126 }, { "auxiliary_loss_clip": 0.0116981, "auxiliary_loss_mlp": 0.01038678, "balance_loss_clip": 1.05486667, "balance_loss_mlp": 1.02861071, "epoch": 0.13659592376600732, "flos": 19937931421920.0, "grad_norm": 3.035788579405516, "language_loss": 0.84228462, "learning_rate": 3.882093784012617e-06, "loss": 0.86436951, "num_input_tokens_seen": 24190190, "step": 1136, "time_per_iteration": 3.3570754528045654 }, { "auxiliary_loss_clip": 0.01202705, "auxiliary_loss_mlp": 0.01043108, "balance_loss_clip": 1.06094182, "balance_loss_mlp": 1.03273654, "epoch": 0.13671616665664643, "flos": 21428582510400.0, "grad_norm": 2.0900629763681278, "language_loss": 0.84244674, "learning_rate": 3.881830133741019e-06, "loss": 0.86490488, "num_input_tokens_seen": 24209055, "step": 1137, "time_per_iteration": 2.610085964202881 }, { "auxiliary_loss_clip": 0.01187495, "auxiliary_loss_mlp": 0.01039981, "balance_loss_clip": 1.06192374, "balance_loss_mlp": 1.03032517, "epoch": 0.13683640954728551, "flos": 22778141416320.0, "grad_norm": 9.617729454973519, "language_loss": 0.76125127, "learning_rate": 3.881566197997285e-06, "loss": 0.78352606, "num_input_tokens_seen": 24225490, "step": 1138, "time_per_iteration": 2.612194776535034 }, { "auxiliary_loss_clip": 0.01199382, "auxiliary_loss_mlp": 0.01030414, "balance_loss_clip": 1.06294048, "balance_loss_mlp": 1.02169359, "epoch": 0.1369566524379246, "flos": 21725887915680.0, "grad_norm": 1.4728458117415923, "language_loss": 0.74657524, "learning_rate": 3.881301976821456e-06, "loss": 0.76887321, "num_input_tokens_seen": 24245520, "step": 1139, "time_per_iteration": 3.4944956302642822 }, { "auxiliary_loss_clip": 0.01218065, "auxiliary_loss_mlp": 0.01035084, "balance_loss_clip": 1.06477237, "balance_loss_mlp": 1.02543998, "epoch": 0.1370768953285637, "flos": 18624462672480.0, "grad_norm": 1.7937870158014337, "language_loss": 0.90879822, "learning_rate": 3.881037470253612e-06, "loss": 0.93132979, "num_input_tokens_seen": 24265035, "step": 1140, "time_per_iteration": 3.4103407859802246 }, { "auxiliary_loss_clip": 0.01175652, "auxiliary_loss_mlp": 0.01033206, "balance_loss_clip": 1.06051838, "balance_loss_mlp": 1.02350259, "epoch": 0.1371971382192028, "flos": 14939524959360.0, "grad_norm": 8.010936410953734, "language_loss": 0.7940942, "learning_rate": 3.88077267833388e-06, "loss": 0.81618273, "num_input_tokens_seen": 24281550, "step": 1141, "time_per_iteration": 2.603039026260376 }, { "auxiliary_loss_clip": 0.01165949, "auxiliary_loss_mlp": 0.01038807, "balance_loss_clip": 1.05647874, "balance_loss_mlp": 1.0291388, "epoch": 0.13731738110984187, "flos": 19023788996160.0, "grad_norm": 1.9091861841310764, "language_loss": 0.83849883, "learning_rate": 3.880507601102427e-06, "loss": 0.86054641, "num_input_tokens_seen": 24299485, "step": 1142, "time_per_iteration": 2.670757293701172 }, { "auxiliary_loss_clip": 0.01236346, "auxiliary_loss_mlp": 0.01044392, "balance_loss_clip": 1.06794035, "balance_loss_mlp": 1.03465819, "epoch": 0.13743762400048098, "flos": 18187465846080.0, "grad_norm": 1.7756986165958795, "language_loss": 0.82261717, "learning_rate": 3.880242238599467e-06, "loss": 0.84542453, "num_input_tokens_seen": 24316010, "step": 1143, "time_per_iteration": 2.5275802612304688 }, { "auxiliary_loss_clip": 0.01232543, "auxiliary_loss_mlp": 0.01032803, "balance_loss_clip": 1.06526268, "balance_loss_mlp": 1.02367187, "epoch": 0.13755786689112007, "flos": 21031985875680.0, "grad_norm": 1.7354314669611952, "language_loss": 0.83342302, "learning_rate": 3.879976590865254e-06, "loss": 0.85607648, "num_input_tokens_seen": 24335465, "step": 1144, "time_per_iteration": 2.5534584522247314 }, { "auxiliary_loss_clip": 0.01198408, "auxiliary_loss_mlp": 0.01038388, "balance_loss_clip": 1.06132555, "balance_loss_mlp": 1.02845788, "epoch": 0.13767810978175915, "flos": 21360640398720.0, "grad_norm": 1.9729234523254953, "language_loss": 0.87677944, "learning_rate": 3.879710657940087e-06, "loss": 0.89914739, "num_input_tokens_seen": 24354415, "step": 1145, "time_per_iteration": 2.6126179695129395 }, { "auxiliary_loss_clip": 0.01221192, "auxiliary_loss_mlp": 0.01043019, "balance_loss_clip": 1.06502008, "balance_loss_mlp": 1.03219509, "epoch": 0.13779835267239823, "flos": 30592089310080.0, "grad_norm": 2.114661781728496, "language_loss": 0.70191932, "learning_rate": 3.879444439864308e-06, "loss": 0.72456145, "num_input_tokens_seen": 24373990, "step": 1146, "time_per_iteration": 2.6408722400665283 }, { "auxiliary_loss_clip": 0.01217998, "auxiliary_loss_mlp": 0.00764956, "balance_loss_clip": 1.06229067, "balance_loss_mlp": 1.00041175, "epoch": 0.13791859556303734, "flos": 22669870946880.0, "grad_norm": 1.8075080405029726, "language_loss": 0.85820925, "learning_rate": 3.879177936678301e-06, "loss": 0.87803882, "num_input_tokens_seen": 24392995, "step": 1147, "time_per_iteration": 2.5934505462646484 }, { "auxiliary_loss_clip": 0.01211139, "auxiliary_loss_mlp": 0.01038552, "balance_loss_clip": 1.06382859, "balance_loss_mlp": 1.02891374, "epoch": 0.13803883845367643, "flos": 35224170746880.0, "grad_norm": 1.7683659419402826, "language_loss": 0.77229345, "learning_rate": 3.878911148422496e-06, "loss": 0.79479039, "num_input_tokens_seen": 24414470, "step": 1148, "time_per_iteration": 2.725125789642334 }, { "auxiliary_loss_clip": 0.01220729, "auxiliary_loss_mlp": 0.0103921, "balance_loss_clip": 1.0646894, "balance_loss_mlp": 1.02870131, "epoch": 0.1381590813443155, "flos": 32014546868160.0, "grad_norm": 2.1658162435966086, "language_loss": 0.70378321, "learning_rate": 3.878644075137364e-06, "loss": 0.72638261, "num_input_tokens_seen": 24435120, "step": 1149, "time_per_iteration": 2.6520836353302 }, { "auxiliary_loss_clip": 0.01164646, "auxiliary_loss_mlp": 0.01040926, "balance_loss_clip": 1.05285454, "balance_loss_mlp": 1.0314728, "epoch": 0.13827932423495462, "flos": 17821859159520.0, "grad_norm": 1.8382351523832487, "language_loss": 0.79398322, "learning_rate": 3.878376716863418e-06, "loss": 0.81603891, "num_input_tokens_seen": 24451420, "step": 1150, "time_per_iteration": 2.6048669815063477 }, { "auxiliary_loss_clip": 0.01199967, "auxiliary_loss_mlp": 0.01036334, "balance_loss_clip": 1.06008554, "balance_loss_mlp": 1.02740562, "epoch": 0.1383995671255937, "flos": 19427102102400.0, "grad_norm": 2.0620546057300824, "language_loss": 0.71430671, "learning_rate": 3.878109073641219e-06, "loss": 0.73666966, "num_input_tokens_seen": 24470450, "step": 1151, "time_per_iteration": 2.617219924926758 }, { "auxiliary_loss_clip": 0.0117227, "auxiliary_loss_mlp": 0.01041919, "balance_loss_clip": 1.05959702, "balance_loss_mlp": 1.03259075, "epoch": 0.13851981001623279, "flos": 28296607857120.0, "grad_norm": 1.5551431379745524, "language_loss": 0.81111473, "learning_rate": 3.877841145511366e-06, "loss": 0.8332566, "num_input_tokens_seen": 24493190, "step": 1152, "time_per_iteration": 2.7338738441467285 }, { "auxiliary_loss_clip": 0.0122422, "auxiliary_loss_mlp": 0.0104001, "balance_loss_clip": 1.06458771, "balance_loss_mlp": 1.02983522, "epoch": 0.13864005290687187, "flos": 21213083163360.0, "grad_norm": 1.7399561464802378, "language_loss": 0.82845807, "learning_rate": 3.8775729325145035e-06, "loss": 0.85110039, "num_input_tokens_seen": 24512425, "step": 1153, "time_per_iteration": 2.6001803874969482 }, { "auxiliary_loss_clip": 0.01081107, "auxiliary_loss_mlp": 0.01009153, "balance_loss_clip": 1.02781558, "balance_loss_mlp": 1.00704253, "epoch": 0.13876029579751098, "flos": 71653402494240.0, "grad_norm": 0.7874319973595921, "language_loss": 0.64776236, "learning_rate": 3.877304434691321e-06, "loss": 0.66866499, "num_input_tokens_seen": 24579275, "step": 1154, "time_per_iteration": 3.320204734802246 }, { "auxiliary_loss_clip": 0.01189259, "auxiliary_loss_mlp": 0.01035628, "balance_loss_clip": 1.06075144, "balance_loss_mlp": 1.02651501, "epoch": 0.13888053868815006, "flos": 21941351345760.0, "grad_norm": 2.1652707492641983, "language_loss": 0.79709315, "learning_rate": 3.877035652082548e-06, "loss": 0.81934202, "num_input_tokens_seen": 24598720, "step": 1155, "time_per_iteration": 2.654372453689575 }, { "auxiliary_loss_clip": 0.01199101, "auxiliary_loss_mlp": 0.0104383, "balance_loss_clip": 1.06342423, "balance_loss_mlp": 1.0331434, "epoch": 0.13900078157878915, "flos": 19608630393600.0, "grad_norm": 3.4030065147285047, "language_loss": 0.85209209, "learning_rate": 3.87676658472896e-06, "loss": 0.87452137, "num_input_tokens_seen": 24617530, "step": 1156, "time_per_iteration": 2.59190034866333 }, { "auxiliary_loss_clip": 0.01216734, "auxiliary_loss_mlp": 0.01041768, "balance_loss_clip": 1.06070638, "balance_loss_mlp": 1.03145075, "epoch": 0.13912102446942826, "flos": 22638054908640.0, "grad_norm": 6.449734440118809, "language_loss": 0.85308683, "learning_rate": 3.876497232671372e-06, "loss": 0.87567186, "num_input_tokens_seen": 24637485, "step": 1157, "time_per_iteration": 2.6226155757904053 }, { "auxiliary_loss_clip": 0.01175932, "auxiliary_loss_mlp": 0.01030864, "balance_loss_clip": 1.05706143, "balance_loss_mlp": 1.02052808, "epoch": 0.13924126736006734, "flos": 29643329323200.0, "grad_norm": 2.053505647874147, "language_loss": 0.84203762, "learning_rate": 3.876227595950647e-06, "loss": 0.86410558, "num_input_tokens_seen": 24656915, "step": 1158, "time_per_iteration": 2.72664475440979 }, { "auxiliary_loss_clip": 0.01233962, "auxiliary_loss_mlp": 0.01040432, "balance_loss_clip": 1.06619394, "balance_loss_mlp": 1.03018594, "epoch": 0.13936151025070642, "flos": 27417657663840.0, "grad_norm": 1.6573345061150067, "language_loss": 0.79183775, "learning_rate": 3.875957674607686e-06, "loss": 0.81458169, "num_input_tokens_seen": 24679190, "step": 1159, "time_per_iteration": 2.610462188720703 }, { "auxiliary_loss_clip": 0.01209251, "auxiliary_loss_mlp": 0.00765738, "balance_loss_clip": 1.05874813, "balance_loss_mlp": 1.00049627, "epoch": 0.1394817531413455, "flos": 16399329767520.0, "grad_norm": 1.7450853609093917, "language_loss": 0.87951845, "learning_rate": 3.8756874686834386e-06, "loss": 0.89926833, "num_input_tokens_seen": 24697405, "step": 1160, "time_per_iteration": 2.567777395248413 }, { "auxiliary_loss_clip": 0.01226724, "auxiliary_loss_mlp": 0.00765692, "balance_loss_clip": 1.06470466, "balance_loss_mlp": 1.00044417, "epoch": 0.13960199603198462, "flos": 30922324179360.0, "grad_norm": 1.8515914365731716, "language_loss": 0.80335963, "learning_rate": 3.875416978218893e-06, "loss": 0.82328379, "num_input_tokens_seen": 24720600, "step": 1161, "time_per_iteration": 3.415235996246338 }, { "auxiliary_loss_clip": 0.01198155, "auxiliary_loss_mlp": 0.0103618, "balance_loss_clip": 1.05690265, "balance_loss_mlp": 1.0264523, "epoch": 0.1397222389226237, "flos": 18113777020800.0, "grad_norm": 2.196258673346456, "language_loss": 0.82835418, "learning_rate": 3.8751462032550835e-06, "loss": 0.85069752, "num_input_tokens_seen": 24737605, "step": 1162, "time_per_iteration": 3.3722643852233887 }, { "auxiliary_loss_clip": 0.01202653, "auxiliary_loss_mlp": 0.01032384, "balance_loss_clip": 1.06593466, "balance_loss_mlp": 1.02282357, "epoch": 0.13984248181326278, "flos": 16872775920000.0, "grad_norm": 2.0928079698425273, "language_loss": 0.83014774, "learning_rate": 3.874875143833085e-06, "loss": 0.85249805, "num_input_tokens_seen": 24755845, "step": 1163, "time_per_iteration": 2.590696096420288 }, { "auxiliary_loss_clip": 0.01223433, "auxiliary_loss_mlp": 0.01035087, "balance_loss_clip": 1.06464922, "balance_loss_mlp": 1.02443588, "epoch": 0.1399627247039019, "flos": 54121407540960.0, "grad_norm": 1.723565979349869, "language_loss": 0.69054365, "learning_rate": 3.874603799994019e-06, "loss": 0.71312892, "num_input_tokens_seen": 24779380, "step": 1164, "time_per_iteration": 2.87278413772583 }, { "auxiliary_loss_clip": 0.01182348, "auxiliary_loss_mlp": 0.01040912, "balance_loss_clip": 1.05976605, "balance_loss_mlp": 1.03105927, "epoch": 0.14008296759454097, "flos": 11765524316640.0, "grad_norm": 1.8486901487395453, "language_loss": 0.86824048, "learning_rate": 3.874332171779046e-06, "loss": 0.89047313, "num_input_tokens_seen": 24794260, "step": 1165, "time_per_iteration": 3.556387424468994 }, { "auxiliary_loss_clip": 0.01185315, "auxiliary_loss_mlp": 0.01034733, "balance_loss_clip": 1.05818045, "balance_loss_mlp": 1.02484417, "epoch": 0.14020321048518006, "flos": 22017518441280.0, "grad_norm": 1.6543923734018395, "language_loss": 0.75757861, "learning_rate": 3.874060259229373e-06, "loss": 0.77977908, "num_input_tokens_seen": 24815835, "step": 1166, "time_per_iteration": 3.564964771270752 }, { "auxiliary_loss_clip": 0.01224976, "auxiliary_loss_mlp": 0.01035655, "balance_loss_clip": 1.06716871, "balance_loss_mlp": 1.02536178, "epoch": 0.14032345337581917, "flos": 23404316846400.0, "grad_norm": 2.328610142868375, "language_loss": 0.93725383, "learning_rate": 3.873788062386249e-06, "loss": 0.95986015, "num_input_tokens_seen": 24834095, "step": 1167, "time_per_iteration": 2.5773983001708984 }, { "auxiliary_loss_clip": 0.01195325, "auxiliary_loss_mlp": 0.01037925, "balance_loss_clip": 1.06345224, "balance_loss_mlp": 1.02761912, "epoch": 0.14044369626645825, "flos": 29645771676480.0, "grad_norm": 1.6911910579670246, "language_loss": 0.82033408, "learning_rate": 3.873515581290965e-06, "loss": 0.84266651, "num_input_tokens_seen": 24858900, "step": 1168, "time_per_iteration": 2.7253241539001465 }, { "auxiliary_loss_clip": 0.0118941, "auxiliary_loss_mlp": 0.01039061, "balance_loss_clip": 1.06271422, "balance_loss_mlp": 1.02987003, "epoch": 0.14056393915709733, "flos": 18332975814720.0, "grad_norm": 1.9874489178588823, "language_loss": 0.75440645, "learning_rate": 3.8732428159848575e-06, "loss": 0.77669114, "num_input_tokens_seen": 24877875, "step": 1169, "time_per_iteration": 2.6178839206695557 }, { "auxiliary_loss_clip": 0.01223017, "auxiliary_loss_mlp": 0.01040586, "balance_loss_clip": 1.06933594, "balance_loss_mlp": 1.03017306, "epoch": 0.14068418204773642, "flos": 26687521799520.0, "grad_norm": 1.8428580247457755, "language_loss": 0.78318381, "learning_rate": 3.872969766509304e-06, "loss": 0.80581987, "num_input_tokens_seen": 24898430, "step": 1170, "time_per_iteration": 2.616997480392456 }, { "auxiliary_loss_clip": 0.01079585, "auxiliary_loss_mlp": 0.01011002, "balance_loss_clip": 1.02045798, "balance_loss_mlp": 1.00880849, "epoch": 0.14080442493837553, "flos": 65259326137440.0, "grad_norm": 0.7664327612391667, "language_loss": 0.55667573, "learning_rate": 3.872696432905726e-06, "loss": 0.57758158, "num_input_tokens_seen": 24959250, "step": 1171, "time_per_iteration": 3.2102036476135254 }, { "auxiliary_loss_clip": 0.01224735, "auxiliary_loss_mlp": 0.01040478, "balance_loss_clip": 1.06277871, "balance_loss_mlp": 1.03007138, "epoch": 0.1409246678290146, "flos": 25776719651040.0, "grad_norm": 2.0852374033869028, "language_loss": 0.7177496, "learning_rate": 3.872422815215589e-06, "loss": 0.74040174, "num_input_tokens_seen": 24978330, "step": 1172, "time_per_iteration": 2.620666027069092 }, { "auxiliary_loss_clip": 0.01217552, "auxiliary_loss_mlp": 0.01038426, "balance_loss_clip": 1.06287432, "balance_loss_mlp": 1.02831078, "epoch": 0.1410449107196537, "flos": 21868524527520.0, "grad_norm": 1.7852675731485603, "language_loss": 0.7438426, "learning_rate": 3.8721489134803994e-06, "loss": 0.76640236, "num_input_tokens_seen": 24997120, "step": 1173, "time_per_iteration": 2.5742249488830566 }, { "auxiliary_loss_clip": 0.01217393, "auxiliary_loss_mlp": 0.01036319, "balance_loss_clip": 1.06213927, "balance_loss_mlp": 1.0261147, "epoch": 0.1411651536102928, "flos": 16684136070720.0, "grad_norm": 2.2559651645194294, "language_loss": 0.7209307, "learning_rate": 3.871874727741707e-06, "loss": 0.74346781, "num_input_tokens_seen": 25014350, "step": 1174, "time_per_iteration": 2.5420150756835938 }, { "auxiliary_loss_clip": 0.01218032, "auxiliary_loss_mlp": 0.01038742, "balance_loss_clip": 1.06636047, "balance_loss_mlp": 1.02918148, "epoch": 0.1412853965009319, "flos": 20992268106240.0, "grad_norm": 1.8069562100613534, "language_loss": 0.96516156, "learning_rate": 3.871600258041108e-06, "loss": 0.98772931, "num_input_tokens_seen": 25033875, "step": 1175, "time_per_iteration": 2.5903208255767822 }, { "auxiliary_loss_clip": 0.01199629, "auxiliary_loss_mlp": 0.01034069, "balance_loss_clip": 1.05891263, "balance_loss_mlp": 1.02440095, "epoch": 0.14140563939157097, "flos": 20335282312800.0, "grad_norm": 5.909468511237083, "language_loss": 0.86066389, "learning_rate": 3.871325504420238e-06, "loss": 0.88300085, "num_input_tokens_seen": 25052865, "step": 1176, "time_per_iteration": 2.609589099884033 }, { "auxiliary_loss_clip": 0.01237791, "auxiliary_loss_mlp": 0.01037799, "balance_loss_clip": 1.06707501, "balance_loss_mlp": 1.02764881, "epoch": 0.14152588228221005, "flos": 21068830288320.0, "grad_norm": 3.3758448690452076, "language_loss": 0.81622434, "learning_rate": 3.871050466920776e-06, "loss": 0.8389802, "num_input_tokens_seen": 25072770, "step": 1177, "time_per_iteration": 2.553762197494507 }, { "auxiliary_loss_clip": 0.01178358, "auxiliary_loss_mlp": 0.01034189, "balance_loss_clip": 1.05413699, "balance_loss_mlp": 1.02502739, "epoch": 0.14164612517284916, "flos": 18223160916000.0, "grad_norm": 1.7884557114726907, "language_loss": 0.7967723, "learning_rate": 3.870775145584447e-06, "loss": 0.81889772, "num_input_tokens_seen": 25090550, "step": 1178, "time_per_iteration": 2.6261444091796875 }, { "auxiliary_loss_clip": 0.01211541, "auxiliary_loss_mlp": 0.01037407, "balance_loss_clip": 1.06385708, "balance_loss_mlp": 1.02540231, "epoch": 0.14176636806348825, "flos": 22744457696160.0, "grad_norm": 4.104891574890078, "language_loss": 0.65053642, "learning_rate": 3.8704995404530145e-06, "loss": 0.67302585, "num_input_tokens_seen": 25106175, "step": 1179, "time_per_iteration": 2.596994400024414 }, { "auxiliary_loss_clip": 0.01238936, "auxiliary_loss_mlp": 0.01037643, "balance_loss_clip": 1.07075143, "balance_loss_mlp": 1.02790964, "epoch": 0.14188661095412733, "flos": 22091099515680.0, "grad_norm": 1.8913748538980095, "language_loss": 0.85045582, "learning_rate": 3.87022365156829e-06, "loss": 0.87322164, "num_input_tokens_seen": 25126890, "step": 1180, "time_per_iteration": 2.577859401702881 }, { "auxiliary_loss_clip": 0.01141449, "auxiliary_loss_mlp": 0.01033892, "balance_loss_clip": 1.05284214, "balance_loss_mlp": 1.02335429, "epoch": 0.14200685384476644, "flos": 24352394411040.0, "grad_norm": 1.915595933581461, "language_loss": 0.81258392, "learning_rate": 3.869947478972123e-06, "loss": 0.83433735, "num_input_tokens_seen": 25147915, "step": 1181, "time_per_iteration": 2.741725206375122 }, { "auxiliary_loss_clip": 0.01210651, "auxiliary_loss_mlp": 0.01038508, "balance_loss_clip": 1.06129074, "balance_loss_mlp": 1.02866721, "epoch": 0.14212709673540552, "flos": 24022051790880.0, "grad_norm": 1.9408165330680462, "language_loss": 0.82463819, "learning_rate": 3.869671022706412e-06, "loss": 0.84712976, "num_input_tokens_seen": 25166645, "step": 1182, "time_per_iteration": 2.6218717098236084 }, { "auxiliary_loss_clip": 0.01157422, "auxiliary_loss_mlp": 0.01040796, "balance_loss_clip": 1.05284917, "balance_loss_mlp": 1.03046656, "epoch": 0.1422473396260446, "flos": 26431801845600.0, "grad_norm": 3.812659106264438, "language_loss": 0.64715308, "learning_rate": 3.869394282813092e-06, "loss": 0.66913533, "num_input_tokens_seen": 25185845, "step": 1183, "time_per_iteration": 2.707285165786743 }, { "auxiliary_loss_clip": 0.01198656, "auxiliary_loss_mlp": 0.01036069, "balance_loss_clip": 1.06076503, "balance_loss_mlp": 1.02584124, "epoch": 0.1423675825166837, "flos": 17055309886080.0, "grad_norm": 2.945826574237492, "language_loss": 0.89386266, "learning_rate": 3.869117259334147e-06, "loss": 0.91620994, "num_input_tokens_seen": 25203770, "step": 1184, "time_per_iteration": 2.636664628982544 }, { "auxiliary_loss_clip": 0.01216261, "auxiliary_loss_mlp": 0.0104228, "balance_loss_clip": 1.06225705, "balance_loss_mlp": 1.03274894, "epoch": 0.1424878254073228, "flos": 17929483123680.0, "grad_norm": 2.111841816738139, "language_loss": 0.81794822, "learning_rate": 3.868839952311599e-06, "loss": 0.84053361, "num_input_tokens_seen": 25221725, "step": 1185, "time_per_iteration": 2.558450222015381 }, { "auxiliary_loss_clip": 0.0119881, "auxiliary_loss_mlp": 0.01036791, "balance_loss_clip": 1.06269515, "balance_loss_mlp": 1.02698636, "epoch": 0.14260806829796188, "flos": 20303609942400.0, "grad_norm": 2.2215118459381458, "language_loss": 0.80602252, "learning_rate": 3.868562361787516e-06, "loss": 0.82837856, "num_input_tokens_seen": 25240855, "step": 1186, "time_per_iteration": 2.60082745552063 }, { "auxiliary_loss_clip": 0.01134882, "auxiliary_loss_mlp": 0.0104074, "balance_loss_clip": 1.05044746, "balance_loss_mlp": 1.03097045, "epoch": 0.14272831118860096, "flos": 23185728640800.0, "grad_norm": 1.9561564923323498, "language_loss": 0.68885446, "learning_rate": 3.868284487804009e-06, "loss": 0.71061063, "num_input_tokens_seen": 25260085, "step": 1187, "time_per_iteration": 3.544635772705078 }, { "auxiliary_loss_clip": 0.01210017, "auxiliary_loss_mlp": 0.01038903, "balance_loss_clip": 1.06066346, "balance_loss_mlp": 1.02910399, "epoch": 0.14284855407924008, "flos": 27232214424000.0, "grad_norm": 1.563903729385867, "language_loss": 0.78071231, "learning_rate": 3.86800633040323e-06, "loss": 0.8032015, "num_input_tokens_seen": 25280675, "step": 1188, "time_per_iteration": 3.402433395385742 }, { "auxiliary_loss_clip": 0.01204328, "auxiliary_loss_mlp": 0.00765162, "balance_loss_clip": 1.06602275, "balance_loss_mlp": 1.00022662, "epoch": 0.14296879696987916, "flos": 28184206937280.0, "grad_norm": 2.1453029518423095, "language_loss": 0.78199893, "learning_rate": 3.867727889627376e-06, "loss": 0.8016938, "num_input_tokens_seen": 25300290, "step": 1189, "time_per_iteration": 2.657757520675659 }, { "auxiliary_loss_clip": 0.01179005, "auxiliary_loss_mlp": 0.01040308, "balance_loss_clip": 1.06131482, "balance_loss_mlp": 1.02965069, "epoch": 0.14308903986051824, "flos": 19390293606720.0, "grad_norm": 2.101468924034636, "language_loss": 0.78139091, "learning_rate": 3.867449165518687e-06, "loss": 0.80358404, "num_input_tokens_seen": 25316760, "step": 1190, "time_per_iteration": 2.605367422103882 }, { "auxiliary_loss_clip": 0.01236955, "auxiliary_loss_mlp": 0.00765592, "balance_loss_clip": 1.06494904, "balance_loss_mlp": 1.00032163, "epoch": 0.14320928275115732, "flos": 17457509566560.0, "grad_norm": 1.7247458326764484, "language_loss": 0.71098328, "learning_rate": 3.867170158119444e-06, "loss": 0.73100877, "num_input_tokens_seen": 25335760, "step": 1191, "time_per_iteration": 3.4385931491851807 }, { "auxiliary_loss_clip": 0.01236787, "auxiliary_loss_mlp": 0.01045906, "balance_loss_clip": 1.06533527, "balance_loss_mlp": 1.03638077, "epoch": 0.14332952564179643, "flos": 21466073428320.0, "grad_norm": 2.3648271110627186, "language_loss": 0.75297135, "learning_rate": 3.866890867471972e-06, "loss": 0.77579832, "num_input_tokens_seen": 25354230, "step": 1192, "time_per_iteration": 2.5612854957580566 }, { "auxiliary_loss_clip": 0.01198541, "auxiliary_loss_mlp": 0.01039054, "balance_loss_clip": 1.05713248, "balance_loss_mlp": 1.02848601, "epoch": 0.14344976853243552, "flos": 16396995165120.0, "grad_norm": 2.458116227914604, "language_loss": 0.89976758, "learning_rate": 3.86661129361864e-06, "loss": 0.92214346, "num_input_tokens_seen": 25368720, "step": 1193, "time_per_iteration": 3.3220436573028564 }, { "auxiliary_loss_clip": 0.01203435, "auxiliary_loss_mlp": 0.01040516, "balance_loss_clip": 1.06308413, "balance_loss_mlp": 1.02993023, "epoch": 0.1435700114230746, "flos": 18916739703360.0, "grad_norm": 1.9662935985352774, "language_loss": 0.86322427, "learning_rate": 3.866331436601859e-06, "loss": 0.88566381, "num_input_tokens_seen": 25386715, "step": 1194, "time_per_iteration": 2.5489866733551025 }, { "auxiliary_loss_clip": 0.01235139, "auxiliary_loss_mlp": 0.0103474, "balance_loss_clip": 1.06684613, "balance_loss_mlp": 1.02540624, "epoch": 0.1436902543137137, "flos": 19755397455840.0, "grad_norm": 2.122281804640451, "language_loss": 0.73278892, "learning_rate": 3.866051296464083e-06, "loss": 0.75548768, "num_input_tokens_seen": 25405550, "step": 1195, "time_per_iteration": 2.4691858291625977 }, { "auxiliary_loss_clip": 0.01240291, "auxiliary_loss_mlp": 0.00765447, "balance_loss_clip": 1.06806922, "balance_loss_mlp": 1.00028586, "epoch": 0.1438104972043528, "flos": 14684810680320.0, "grad_norm": 2.145462738024547, "language_loss": 0.85423517, "learning_rate": 3.86577087324781e-06, "loss": 0.87429261, "num_input_tokens_seen": 25422040, "step": 1196, "time_per_iteration": 2.5206120014190674 }, { "auxiliary_loss_clip": 0.01215893, "auxiliary_loss_mlp": 0.01029571, "balance_loss_clip": 1.06481695, "balance_loss_mlp": 1.01965868, "epoch": 0.14393074009499188, "flos": 17092333883520.0, "grad_norm": 1.8423641083948799, "language_loss": 0.77310175, "learning_rate": 3.865490166995578e-06, "loss": 0.79555637, "num_input_tokens_seen": 25440270, "step": 1197, "time_per_iteration": 2.5398685932159424 }, { "auxiliary_loss_clip": 0.01223831, "auxiliary_loss_mlp": 0.0103983, "balance_loss_clip": 1.06683147, "balance_loss_mlp": 1.02931547, "epoch": 0.144050982985631, "flos": 30476204445120.0, "grad_norm": 3.0112414543600456, "language_loss": 0.84275585, "learning_rate": 3.86520917774997e-06, "loss": 0.86539245, "num_input_tokens_seen": 25459705, "step": 1198, "time_per_iteration": 2.646131753921509 }, { "auxiliary_loss_clip": 0.01215745, "auxiliary_loss_mlp": 0.01035891, "balance_loss_clip": 1.0632385, "balance_loss_mlp": 1.02605653, "epoch": 0.14417122587627007, "flos": 17858488070400.0, "grad_norm": 2.0227491832459172, "language_loss": 0.75086629, "learning_rate": 3.864927905553614e-06, "loss": 0.77338266, "num_input_tokens_seen": 25477615, "step": 1199, "time_per_iteration": 2.5363903045654297 }, { "auxiliary_loss_clip": 0.01185751, "auxiliary_loss_mlp": 0.01039412, "balance_loss_clip": 1.06034803, "balance_loss_mlp": 1.03023291, "epoch": 0.14429146876690915, "flos": 21613917999360.0, "grad_norm": 1.5619295976014245, "language_loss": 0.88712776, "learning_rate": 3.8646463504491765e-06, "loss": 0.90937942, "num_input_tokens_seen": 25497750, "step": 1200, "time_per_iteration": 2.6355841159820557 }, { "auxiliary_loss_clip": 0.01222822, "auxiliary_loss_mlp": 0.01038758, "balance_loss_clip": 1.06807828, "balance_loss_mlp": 1.02859545, "epoch": 0.14441171165754824, "flos": 23258124455520.0, "grad_norm": 1.660003813749119, "language_loss": 0.83201659, "learning_rate": 3.8643645124793705e-06, "loss": 0.85463238, "num_input_tokens_seen": 25516650, "step": 1201, "time_per_iteration": 2.5666463375091553 }, { "auxiliary_loss_clip": 0.01219362, "auxiliary_loss_mlp": 0.01041174, "balance_loss_clip": 1.06452084, "balance_loss_mlp": 1.03137493, "epoch": 0.14453195454818735, "flos": 42854219830080.0, "grad_norm": 1.5648110411362897, "language_loss": 0.74863231, "learning_rate": 3.8640823916869515e-06, "loss": 0.77123773, "num_input_tokens_seen": 25540960, "step": 1202, "time_per_iteration": 2.7662501335144043 }, { "auxiliary_loss_clip": 0.01237687, "auxiliary_loss_mlp": 0.01046626, "balance_loss_clip": 1.06650257, "balance_loss_mlp": 1.03633869, "epoch": 0.14465219743882643, "flos": 27235877953920.0, "grad_norm": 1.506954697114476, "language_loss": 0.78425092, "learning_rate": 3.863799988114714e-06, "loss": 0.80709398, "num_input_tokens_seen": 25562990, "step": 1203, "time_per_iteration": 2.5733156204223633 }, { "auxiliary_loss_clip": 0.01237217, "auxiliary_loss_mlp": 0.01041866, "balance_loss_clip": 1.06562757, "balance_loss_mlp": 1.03079152, "epoch": 0.1447724403294655, "flos": 16690708874400.0, "grad_norm": 3.0985836271176175, "language_loss": 0.70223606, "learning_rate": 3.863517301805502e-06, "loss": 0.72502685, "num_input_tokens_seen": 25581380, "step": 1204, "time_per_iteration": 2.5285916328430176 }, { "auxiliary_loss_clip": 0.01195049, "auxiliary_loss_mlp": 0.01041206, "balance_loss_clip": 1.06512094, "balance_loss_mlp": 1.03011918, "epoch": 0.14489268322010462, "flos": 20073743811360.0, "grad_norm": 2.6071386486994097, "language_loss": 0.96961629, "learning_rate": 3.863234332802196e-06, "loss": 0.99197882, "num_input_tokens_seen": 25593585, "step": 1205, "time_per_iteration": 2.599982738494873 }, { "auxiliary_loss_clip": 0.01203749, "auxiliary_loss_mlp": 0.01038781, "balance_loss_clip": 1.06248236, "balance_loss_mlp": 1.02883863, "epoch": 0.1450129261107437, "flos": 27125632051680.0, "grad_norm": 1.9919773992364906, "language_loss": 0.73896873, "learning_rate": 3.862951081147723e-06, "loss": 0.76139402, "num_input_tokens_seen": 25613750, "step": 1206, "time_per_iteration": 2.6521799564361572 }, { "auxiliary_loss_clip": 0.01224361, "auxiliary_loss_mlp": 0.0103331, "balance_loss_clip": 1.06724727, "balance_loss_mlp": 1.02407682, "epoch": 0.1451331690013828, "flos": 25702348403520.0, "grad_norm": 2.0332356442157953, "language_loss": 0.78083491, "learning_rate": 3.862667546885053e-06, "loss": 0.80341166, "num_input_tokens_seen": 25632300, "step": 1207, "time_per_iteration": 2.5872957706451416 }, { "auxiliary_loss_clip": 0.01210524, "auxiliary_loss_mlp": 0.01039914, "balance_loss_clip": 1.06177807, "balance_loss_mlp": 1.02948308, "epoch": 0.14525341189202187, "flos": 25737396968160.0, "grad_norm": 1.8181802100193114, "language_loss": 0.73520195, "learning_rate": 3.8623837300571965e-06, "loss": 0.7577064, "num_input_tokens_seen": 25651285, "step": 1208, "time_per_iteration": 2.6316936016082764 }, { "auxiliary_loss_clip": 0.0123974, "auxiliary_loss_mlp": 0.01046753, "balance_loss_clip": 1.06800556, "balance_loss_mlp": 1.03613758, "epoch": 0.14537365478266098, "flos": 23073902392320.0, "grad_norm": 1.7690249075007445, "language_loss": 0.84167504, "learning_rate": 3.8620996307072085e-06, "loss": 0.86453998, "num_input_tokens_seen": 25671990, "step": 1209, "time_per_iteration": 2.5796327590942383 }, { "auxiliary_loss_clip": 0.01193335, "auxiliary_loss_mlp": 0.01043621, "balance_loss_clip": 1.05808234, "balance_loss_mlp": 1.0342871, "epoch": 0.14549389767330007, "flos": 20595024966240.0, "grad_norm": 1.586062091352703, "language_loss": 0.64409733, "learning_rate": 3.861815248878188e-06, "loss": 0.66646695, "num_input_tokens_seen": 25689475, "step": 1210, "time_per_iteration": 2.6293134689331055 }, { "auxiliary_loss_clip": 0.0120069, "auxiliary_loss_mlp": 0.01039007, "balance_loss_clip": 1.06318831, "balance_loss_mlp": 1.02876735, "epoch": 0.14561414056393915, "flos": 15121807506720.0, "grad_norm": 2.483129892329597, "language_loss": 0.79643083, "learning_rate": 3.861530584613274e-06, "loss": 0.81882775, "num_input_tokens_seen": 25707475, "step": 1211, "time_per_iteration": 2.591754913330078 }, { "auxiliary_loss_clip": 0.01225623, "auxiliary_loss_mlp": 0.0076565, "balance_loss_clip": 1.06766033, "balance_loss_mlp": 1.00021243, "epoch": 0.14573438345457826, "flos": 19427497188960.0, "grad_norm": 2.973591078660197, "language_loss": 0.82256866, "learning_rate": 3.86124563795565e-06, "loss": 0.84248137, "num_input_tokens_seen": 25726290, "step": 1212, "time_per_iteration": 2.5651495456695557 }, { "auxiliary_loss_clip": 0.01237665, "auxiliary_loss_mlp": 0.01042071, "balance_loss_clip": 1.06856644, "balance_loss_mlp": 1.03251028, "epoch": 0.14585462634521734, "flos": 24828426584640.0, "grad_norm": 1.6454455371531738, "language_loss": 0.70303512, "learning_rate": 3.860960408948543e-06, "loss": 0.72583246, "num_input_tokens_seen": 25748040, "step": 1213, "time_per_iteration": 3.337252140045166 }, { "auxiliary_loss_clip": 0.01214841, "auxiliary_loss_mlp": 0.01034806, "balance_loss_clip": 1.06478786, "balance_loss_mlp": 1.02438688, "epoch": 0.14597486923585642, "flos": 15448630264800.0, "grad_norm": 2.278162394492896, "language_loss": 0.89923525, "learning_rate": 3.860674897635222e-06, "loss": 0.92173171, "num_input_tokens_seen": 25764525, "step": 1214, "time_per_iteration": 3.264616012573242 }, { "auxiliary_loss_clip": 0.01227126, "auxiliary_loss_mlp": 0.01043998, "balance_loss_clip": 1.07101262, "balance_loss_mlp": 1.03443766, "epoch": 0.1460951121264955, "flos": 16655157472320.0, "grad_norm": 1.9669201497381157, "language_loss": 0.83454299, "learning_rate": 3.860389104058998e-06, "loss": 0.85725427, "num_input_tokens_seen": 25782755, "step": 1215, "time_per_iteration": 2.5644643306732178 }, { "auxiliary_loss_clip": 0.01207165, "auxiliary_loss_mlp": 0.01035003, "balance_loss_clip": 1.06196213, "balance_loss_mlp": 1.0246917, "epoch": 0.14621535501713462, "flos": 24863295564480.0, "grad_norm": 1.839095632071577, "language_loss": 0.72449756, "learning_rate": 3.860103028263227e-06, "loss": 0.74691916, "num_input_tokens_seen": 25805860, "step": 1216, "time_per_iteration": 2.643733501434326 }, { "auxiliary_loss_clip": 0.01170011, "auxiliary_loss_mlp": 0.01040452, "balance_loss_clip": 1.0553143, "balance_loss_mlp": 1.03022337, "epoch": 0.1463355979077737, "flos": 25228004327040.0, "grad_norm": 2.336154779636646, "language_loss": 0.70596725, "learning_rate": 3.859816670291304e-06, "loss": 0.72807187, "num_input_tokens_seen": 25824955, "step": 1217, "time_per_iteration": 3.6283535957336426 }, { "auxiliary_loss_clip": 0.01151823, "auxiliary_loss_mlp": 0.01039468, "balance_loss_clip": 1.05850101, "balance_loss_mlp": 1.02987766, "epoch": 0.14645584079841278, "flos": 22054147352160.0, "grad_norm": 2.2749868872606265, "language_loss": 0.89543146, "learning_rate": 3.859530030186672e-06, "loss": 0.91734439, "num_input_tokens_seen": 25841965, "step": 1218, "time_per_iteration": 2.693723201751709 }, { "auxiliary_loss_clip": 0.01211679, "auxiliary_loss_mlp": 0.01047367, "balance_loss_clip": 1.06553614, "balance_loss_mlp": 1.03677547, "epoch": 0.1465760836890519, "flos": 23623874809920.0, "grad_norm": 2.7130429791283572, "language_loss": 0.82947892, "learning_rate": 3.859243107992813e-06, "loss": 0.85206938, "num_input_tokens_seen": 25860770, "step": 1219, "time_per_iteration": 3.5174758434295654 }, { "auxiliary_loss_clip": 0.01194557, "auxiliary_loss_mlp": 0.01037742, "balance_loss_clip": 1.05600548, "balance_loss_mlp": 1.02728748, "epoch": 0.14669632657969098, "flos": 37407897785280.0, "grad_norm": 2.3214470082936525, "language_loss": 0.7799089, "learning_rate": 3.858955903753252e-06, "loss": 0.80223191, "num_input_tokens_seen": 25879410, "step": 1220, "time_per_iteration": 2.755621910095215 }, { "auxiliary_loss_clip": 0.01220112, "auxiliary_loss_mlp": 0.01043374, "balance_loss_clip": 1.06290483, "balance_loss_mlp": 1.03375411, "epoch": 0.14681656947033006, "flos": 28365914813280.0, "grad_norm": 1.5503443280559912, "language_loss": 0.83651227, "learning_rate": 3.858668417511559e-06, "loss": 0.85914707, "num_input_tokens_seen": 25902160, "step": 1221, "time_per_iteration": 2.641679525375366 }, { "auxiliary_loss_clip": 0.01207782, "auxiliary_loss_mlp": 0.01046736, "balance_loss_clip": 1.06341314, "balance_loss_mlp": 1.03673434, "epoch": 0.14693681236096917, "flos": 18479491458240.0, "grad_norm": 1.9964940596988863, "language_loss": 0.76615411, "learning_rate": 3.8583806493113445e-06, "loss": 0.78869927, "num_input_tokens_seen": 25920505, "step": 1222, "time_per_iteration": 2.5787408351898193 }, { "auxiliary_loss_clip": 0.01218164, "auxiliary_loss_mlp": 0.01034121, "balance_loss_clip": 1.06295598, "balance_loss_mlp": 1.02378523, "epoch": 0.14705705525160825, "flos": 20777810351040.0, "grad_norm": 2.391214743247772, "language_loss": 0.82330829, "learning_rate": 3.858092599196263e-06, "loss": 0.84583116, "num_input_tokens_seen": 25938460, "step": 1223, "time_per_iteration": 2.5657079219818115 }, { "auxiliary_loss_clip": 0.01220048, "auxiliary_loss_mlp": 0.01033202, "balance_loss_clip": 1.06519437, "balance_loss_mlp": 1.02267599, "epoch": 0.14717729814224734, "flos": 29932948499040.0, "grad_norm": 2.22824425511712, "language_loss": 0.82228076, "learning_rate": 3.857804267210012e-06, "loss": 0.84481329, "num_input_tokens_seen": 25957760, "step": 1224, "time_per_iteration": 2.6110177040100098 }, { "auxiliary_loss_clip": 0.0117656, "auxiliary_loss_mlp": 0.01035926, "balance_loss_clip": 1.0547787, "balance_loss_mlp": 1.02612674, "epoch": 0.14729754103288642, "flos": 20047494901920.0, "grad_norm": 2.5546112601301343, "language_loss": 0.88240486, "learning_rate": 3.857515653396331e-06, "loss": 0.90452969, "num_input_tokens_seen": 25974970, "step": 1225, "time_per_iteration": 2.6347572803497314 }, { "auxiliary_loss_clip": 0.01177937, "auxiliary_loss_mlp": 0.01039269, "balance_loss_clip": 1.05916119, "balance_loss_mlp": 1.02905858, "epoch": 0.14741778392352553, "flos": 19281520299840.0, "grad_norm": 2.453912569423712, "language_loss": 0.87143266, "learning_rate": 3.857226757799002e-06, "loss": 0.89360476, "num_input_tokens_seen": 25992525, "step": 1226, "time_per_iteration": 2.6310083866119385 }, { "auxiliary_loss_clip": 0.01205566, "auxiliary_loss_mlp": 0.01033691, "balance_loss_clip": 1.06183767, "balance_loss_mlp": 1.02371323, "epoch": 0.1475380268141646, "flos": 25411112964480.0, "grad_norm": 2.210985204137292, "language_loss": 0.74009973, "learning_rate": 3.85693758046185e-06, "loss": 0.7624923, "num_input_tokens_seen": 26010815, "step": 1227, "time_per_iteration": 2.652834415435791 }, { "auxiliary_loss_clip": 0.01239389, "auxiliary_loss_mlp": 0.01036913, "balance_loss_clip": 1.07062912, "balance_loss_mlp": 1.02753687, "epoch": 0.1476582697048037, "flos": 20847656061600.0, "grad_norm": 2.096642296966634, "language_loss": 0.82567513, "learning_rate": 3.8566481214287435e-06, "loss": 0.84843814, "num_input_tokens_seen": 26028935, "step": 1228, "time_per_iteration": 2.5396716594696045 }, { "auxiliary_loss_clip": 0.0118015, "auxiliary_loss_mlp": 0.01044508, "balance_loss_clip": 1.05609953, "balance_loss_mlp": 1.0355916, "epoch": 0.1477785125954428, "flos": 14028112222560.0, "grad_norm": 1.991496256205519, "language_loss": 0.91086638, "learning_rate": 3.8563583807435935e-06, "loss": 0.93311298, "num_input_tokens_seen": 26045080, "step": 1229, "time_per_iteration": 2.6383464336395264 }, { "auxiliary_loss_clip": 0.01222719, "auxiliary_loss_mlp": 0.0076569, "balance_loss_clip": 1.06520391, "balance_loss_mlp": 1.00017047, "epoch": 0.1478987554860819, "flos": 20516702853120.0, "grad_norm": 1.8499442967496833, "language_loss": 0.77655357, "learning_rate": 3.856068358450353e-06, "loss": 0.79643762, "num_input_tokens_seen": 26065030, "step": 1230, "time_per_iteration": 2.5867831707000732 }, { "auxiliary_loss_clip": 0.01204929, "auxiliary_loss_mlp": 0.01033195, "balance_loss_clip": 1.06732929, "balance_loss_mlp": 1.02324653, "epoch": 0.14801899837672097, "flos": 17857015475040.0, "grad_norm": 1.6664932816518299, "language_loss": 0.85693413, "learning_rate": 3.8557780545930186e-06, "loss": 0.87931538, "num_input_tokens_seen": 26083445, "step": 1231, "time_per_iteration": 2.5893197059631348 }, { "auxiliary_loss_clip": 0.01205473, "auxiliary_loss_mlp": 0.01036271, "balance_loss_clip": 1.06573248, "balance_loss_mlp": 1.02678227, "epoch": 0.14813924126736006, "flos": 20881411615680.0, "grad_norm": 1.6736370805045722, "language_loss": 0.7970342, "learning_rate": 3.855487469215628e-06, "loss": 0.81945163, "num_input_tokens_seen": 26102375, "step": 1232, "time_per_iteration": 2.607849359512329 }, { "auxiliary_loss_clip": 0.01186853, "auxiliary_loss_mlp": 0.01034582, "balance_loss_clip": 1.06041825, "balance_loss_mlp": 1.02499759, "epoch": 0.14825948415799917, "flos": 37414075502400.0, "grad_norm": 2.307261337001904, "language_loss": 0.72924829, "learning_rate": 3.855196602362264e-06, "loss": 0.75146264, "num_input_tokens_seen": 26125295, "step": 1233, "time_per_iteration": 2.7646069526672363 }, { "auxiliary_loss_clip": 0.01221071, "auxiliary_loss_mlp": 0.01036618, "balance_loss_clip": 1.0638926, "balance_loss_mlp": 1.02739775, "epoch": 0.14837972704863825, "flos": 22014645084480.0, "grad_norm": 2.16493790670649, "language_loss": 0.94197512, "learning_rate": 3.854905454077051e-06, "loss": 0.96455199, "num_input_tokens_seen": 26142905, "step": 1234, "time_per_iteration": 2.5491554737091064 }, { "auxiliary_loss_clip": 0.01137645, "auxiliary_loss_mlp": 0.01040736, "balance_loss_clip": 1.05215263, "balance_loss_mlp": 1.03037691, "epoch": 0.14849996993927733, "flos": 20996326722720.0, "grad_norm": 2.087493984294837, "language_loss": 0.88349462, "learning_rate": 3.854614024404155e-06, "loss": 0.90527844, "num_input_tokens_seen": 26161215, "step": 1235, "time_per_iteration": 2.7126965522766113 }, { "auxiliary_loss_clip": 0.01190172, "auxiliary_loss_mlp": 0.01044562, "balance_loss_clip": 1.05907416, "balance_loss_mlp": 1.03420842, "epoch": 0.14862021282991644, "flos": 20047997739360.0, "grad_norm": 1.8209499413468, "language_loss": 0.89314997, "learning_rate": 3.8543223133877865e-06, "loss": 0.9154973, "num_input_tokens_seen": 26179810, "step": 1236, "time_per_iteration": 2.583059072494507 }, { "auxiliary_loss_clip": 0.0118241, "auxiliary_loss_mlp": 0.01042687, "balance_loss_clip": 1.0551461, "balance_loss_mlp": 1.03204179, "epoch": 0.14874045572055553, "flos": 22712031069600.0, "grad_norm": 1.8676359729074214, "language_loss": 0.88260329, "learning_rate": 3.854030321072198e-06, "loss": 0.9048543, "num_input_tokens_seen": 26199715, "step": 1237, "time_per_iteration": 2.6276609897613525 }, { "auxiliary_loss_clip": 0.01193718, "auxiliary_loss_mlp": 0.01039479, "balance_loss_clip": 1.05976129, "balance_loss_mlp": 1.02985311, "epoch": 0.1488606986111946, "flos": 25411292549280.0, "grad_norm": 1.9857307602155645, "language_loss": 0.73628879, "learning_rate": 3.853738047501682e-06, "loss": 0.7586208, "num_input_tokens_seen": 26220275, "step": 1238, "time_per_iteration": 2.654232978820801 }, { "auxiliary_loss_clip": 0.01225287, "auxiliary_loss_mlp": 0.01035281, "balance_loss_clip": 1.06960154, "balance_loss_mlp": 1.02531528, "epoch": 0.1489809415018337, "flos": 17018752809120.0, "grad_norm": 2.5498850403603632, "language_loss": 0.77206546, "learning_rate": 3.85344549272058e-06, "loss": 0.79467106, "num_input_tokens_seen": 26238255, "step": 1239, "time_per_iteration": 3.269251585006714 }, { "auxiliary_loss_clip": 0.01215272, "auxiliary_loss_mlp": 0.01039216, "balance_loss_clip": 1.06332672, "balance_loss_mlp": 1.02907765, "epoch": 0.1491011843924728, "flos": 33659399829600.0, "grad_norm": 1.8200528004020156, "language_loss": 0.82668257, "learning_rate": 3.853152656773269e-06, "loss": 0.84922743, "num_input_tokens_seen": 26259690, "step": 1240, "time_per_iteration": 3.4414544105529785 }, { "auxiliary_loss_clip": 0.01199267, "auxiliary_loss_mlp": 0.01034635, "balance_loss_clip": 1.06224442, "balance_loss_mlp": 1.02501488, "epoch": 0.14922142728311188, "flos": 21179004356640.0, "grad_norm": 1.6384491747010905, "language_loss": 0.8474229, "learning_rate": 3.852859539704174e-06, "loss": 0.86976194, "num_input_tokens_seen": 26278990, "step": 1241, "time_per_iteration": 2.6067681312561035 }, { "auxiliary_loss_clip": 0.01174034, "auxiliary_loss_mlp": 0.01036372, "balance_loss_clip": 1.05820537, "balance_loss_mlp": 1.02727032, "epoch": 0.14934167017375097, "flos": 29860552684320.0, "grad_norm": 1.877431931727231, "language_loss": 0.76145613, "learning_rate": 3.85256614155776e-06, "loss": 0.78356028, "num_input_tokens_seen": 26299120, "step": 1242, "time_per_iteration": 2.7473814487457275 }, { "auxiliary_loss_clip": 0.01213063, "auxiliary_loss_mlp": 0.01038042, "balance_loss_clip": 1.06024337, "balance_loss_mlp": 1.02878582, "epoch": 0.14946191306439008, "flos": 17019219729600.0, "grad_norm": 1.826815387434482, "language_loss": 0.74374199, "learning_rate": 3.852272462378535e-06, "loss": 0.76625299, "num_input_tokens_seen": 26316995, "step": 1243, "time_per_iteration": 3.484161138534546 }, { "auxiliary_loss_clip": 0.01202097, "auxiliary_loss_mlp": 0.01030431, "balance_loss_clip": 1.06128311, "balance_loss_mlp": 1.02159739, "epoch": 0.14958215595502916, "flos": 15669050235360.0, "grad_norm": 2.0232718051593976, "language_loss": 0.77647614, "learning_rate": 3.85197850221105e-06, "loss": 0.79880142, "num_input_tokens_seen": 26333295, "step": 1244, "time_per_iteration": 3.4399819374084473 }, { "auxiliary_loss_clip": 0.01213012, "auxiliary_loss_mlp": 0.01036776, "balance_loss_clip": 1.06452167, "balance_loss_mlp": 1.02792454, "epoch": 0.14970239884566824, "flos": 33108565404960.0, "grad_norm": 2.697091201699044, "language_loss": 0.75638735, "learning_rate": 3.851684261099899e-06, "loss": 0.77888525, "num_input_tokens_seen": 26355035, "step": 1245, "time_per_iteration": 2.6769089698791504 }, { "auxiliary_loss_clip": 0.01195803, "auxiliary_loss_mlp": 0.0103672, "balance_loss_clip": 1.05816889, "balance_loss_mlp": 1.02684987, "epoch": 0.14982264173630733, "flos": 17821248571200.0, "grad_norm": 1.7906879914568117, "language_loss": 0.86726367, "learning_rate": 3.851389739089718e-06, "loss": 0.88958895, "num_input_tokens_seen": 26371655, "step": 1246, "time_per_iteration": 2.5789382457733154 }, { "auxiliary_loss_clip": 0.01218658, "auxiliary_loss_mlp": 0.0103785, "balance_loss_clip": 1.06713581, "balance_loss_mlp": 1.02851593, "epoch": 0.14994288462694644, "flos": 32409563156640.0, "grad_norm": 1.732242539190619, "language_loss": 0.80164373, "learning_rate": 3.851094936225186e-06, "loss": 0.82420874, "num_input_tokens_seen": 26392540, "step": 1247, "time_per_iteration": 2.6905267238616943 }, { "auxiliary_loss_clip": 0.01198585, "auxiliary_loss_mlp": 0.01038627, "balance_loss_clip": 1.06432581, "balance_loss_mlp": 1.02888203, "epoch": 0.15006312751758552, "flos": 31794665652000.0, "grad_norm": 1.6112059089628712, "language_loss": 0.76561642, "learning_rate": 3.850799852551024e-06, "loss": 0.78798854, "num_input_tokens_seen": 26414960, "step": 1248, "time_per_iteration": 2.680058002471924 }, { "auxiliary_loss_clip": 0.01206872, "auxiliary_loss_mlp": 0.01040237, "balance_loss_clip": 1.06041598, "balance_loss_mlp": 1.03162432, "epoch": 0.1501833704082246, "flos": 16618025724000.0, "grad_norm": 2.1516874032024673, "language_loss": 0.85837388, "learning_rate": 3.850504488111995e-06, "loss": 0.88084501, "num_input_tokens_seen": 26431635, "step": 1249, "time_per_iteration": 2.545436143875122 }, { "auxiliary_loss_clip": 0.01193127, "auxiliary_loss_mlp": 0.01036191, "balance_loss_clip": 1.05942011, "balance_loss_mlp": 1.02766728, "epoch": 0.15030361329886371, "flos": 23471181449280.0, "grad_norm": 3.944686170122494, "language_loss": 0.82809305, "learning_rate": 3.850208842952907e-06, "loss": 0.85038626, "num_input_tokens_seen": 26450440, "step": 1250, "time_per_iteration": 2.627532720565796 }, { "auxiliary_loss_clip": 0.0117916, "auxiliary_loss_mlp": 0.01029818, "balance_loss_clip": 1.05599785, "balance_loss_mlp": 1.02012014, "epoch": 0.1504238561895028, "flos": 25629413834400.0, "grad_norm": 2.210564664133399, "language_loss": 0.79348129, "learning_rate": 3.849912917118608e-06, "loss": 0.81557107, "num_input_tokens_seen": 26471480, "step": 1251, "time_per_iteration": 2.705655097961426 }, { "auxiliary_loss_clip": 0.01114433, "auxiliary_loss_mlp": 0.01001092, "balance_loss_clip": 1.02239633, "balance_loss_mlp": 0.99887496, "epoch": 0.15054409908014188, "flos": 52095157207200.0, "grad_norm": 0.881821476479684, "language_loss": 0.59313655, "learning_rate": 3.849616710653992e-06, "loss": 0.61429179, "num_input_tokens_seen": 26532950, "step": 1252, "time_per_iteration": 3.1207265853881836 }, { "auxiliary_loss_clip": 0.01218963, "auxiliary_loss_mlp": 0.01030942, "balance_loss_clip": 1.06472898, "balance_loss_mlp": 1.02159619, "epoch": 0.150664341970781, "flos": 18880254460320.0, "grad_norm": 2.7649274059428475, "language_loss": 0.75048006, "learning_rate": 3.84932022360399e-06, "loss": 0.77297914, "num_input_tokens_seen": 26551615, "step": 1253, "time_per_iteration": 2.5661821365356445 }, { "auxiliary_loss_clip": 0.0119744, "auxiliary_loss_mlp": 0.01032984, "balance_loss_clip": 1.06410837, "balance_loss_mlp": 1.02418017, "epoch": 0.15078458486142007, "flos": 22163243911680.0, "grad_norm": 2.7064254751193215, "language_loss": 0.84239268, "learning_rate": 3.849023456013581e-06, "loss": 0.86469698, "num_input_tokens_seen": 26569175, "step": 1254, "time_per_iteration": 2.6158158779144287 }, { "auxiliary_loss_clip": 0.01222211, "auxiliary_loss_mlp": 0.01041564, "balance_loss_clip": 1.06355095, "balance_loss_mlp": 1.03195596, "epoch": 0.15090482775205916, "flos": 26651898563520.0, "grad_norm": 2.080274181108472, "language_loss": 0.62212205, "learning_rate": 3.848726407927784e-06, "loss": 0.64475977, "num_input_tokens_seen": 26589560, "step": 1255, "time_per_iteration": 2.6083555221557617 }, { "auxiliary_loss_clip": 0.01203025, "auxiliary_loss_mlp": 0.01032149, "balance_loss_clip": 1.06383395, "balance_loss_mlp": 1.0211277, "epoch": 0.15102507064269824, "flos": 21799002069600.0, "grad_norm": 2.611848692939244, "language_loss": 0.86870766, "learning_rate": 3.84842907939166e-06, "loss": 0.8910594, "num_input_tokens_seen": 26608785, "step": 1256, "time_per_iteration": 2.606674909591675 }, { "auxiliary_loss_clip": 0.0118058, "auxiliary_loss_mlp": 0.0103402, "balance_loss_clip": 1.0600729, "balance_loss_mlp": 1.02532351, "epoch": 0.15114531353333735, "flos": 22820912127360.0, "grad_norm": 2.8649092566139442, "language_loss": 0.71584857, "learning_rate": 3.8481314704503146e-06, "loss": 0.73799455, "num_input_tokens_seen": 26628615, "step": 1257, "time_per_iteration": 2.617628335952759 }, { "auxiliary_loss_clip": 0.01213787, "auxiliary_loss_mlp": 0.01042571, "balance_loss_clip": 1.06648624, "balance_loss_mlp": 1.03295684, "epoch": 0.15126555642397643, "flos": 19682678388480.0, "grad_norm": 2.2183371809415657, "language_loss": 0.88156587, "learning_rate": 3.847833581148895e-06, "loss": 0.90412945, "num_input_tokens_seen": 26647525, "step": 1258, "time_per_iteration": 2.5770463943481445 }, { "auxiliary_loss_clip": 0.01230555, "auxiliary_loss_mlp": 0.01033672, "balance_loss_clip": 1.06407082, "balance_loss_mlp": 1.02393818, "epoch": 0.15138579931461552, "flos": 28726026204960.0, "grad_norm": 2.3704946482830556, "language_loss": 0.81082702, "learning_rate": 3.84753541153259e-06, "loss": 0.83346927, "num_input_tokens_seen": 26667095, "step": 1259, "time_per_iteration": 2.587491750717163 }, { "auxiliary_loss_clip": 0.01217218, "auxiliary_loss_mlp": 0.01036242, "balance_loss_clip": 1.06579208, "balance_loss_mlp": 1.0270865, "epoch": 0.15150604220525463, "flos": 22127010087360.0, "grad_norm": 1.7467027898859266, "language_loss": 0.83458495, "learning_rate": 3.847236961646633e-06, "loss": 0.85711956, "num_input_tokens_seen": 26686075, "step": 1260, "time_per_iteration": 2.6010000705718994 }, { "auxiliary_loss_clip": 0.01192456, "auxiliary_loss_mlp": 0.01034368, "balance_loss_clip": 1.05769908, "balance_loss_mlp": 1.02563584, "epoch": 0.1516262850958937, "flos": 12968711246880.0, "grad_norm": 3.24735901972944, "language_loss": 0.77770197, "learning_rate": 3.846938231536296e-06, "loss": 0.79997027, "num_input_tokens_seen": 26701695, "step": 1261, "time_per_iteration": 2.5776829719543457 }, { "auxiliary_loss_clip": 0.01225378, "auxiliary_loss_mlp": 0.0104587, "balance_loss_clip": 1.06849277, "balance_loss_mlp": 1.03647637, "epoch": 0.1517465279865328, "flos": 21797134387680.0, "grad_norm": 1.6979216113606468, "language_loss": 0.80709887, "learning_rate": 3.8466392212468995e-06, "loss": 0.82981133, "num_input_tokens_seen": 26721885, "step": 1262, "time_per_iteration": 2.598142623901367 }, { "auxiliary_loss_clip": 0.01092705, "auxiliary_loss_mlp": 0.01002476, "balance_loss_clip": 1.01916325, "balance_loss_mlp": 1.00035417, "epoch": 0.15186677087717187, "flos": 58174580267040.0, "grad_norm": 0.8193126755402029, "language_loss": 0.61983383, "learning_rate": 3.8463399308238e-06, "loss": 0.64078557, "num_input_tokens_seen": 26780990, "step": 1263, "time_per_iteration": 3.1464641094207764 }, { "auxiliary_loss_clip": 0.01214067, "auxiliary_loss_mlp": 0.0104289, "balance_loss_clip": 1.06417537, "balance_loss_mlp": 1.03377676, "epoch": 0.15198701376781099, "flos": 32669700896640.0, "grad_norm": 2.4305225668835186, "language_loss": 0.64194524, "learning_rate": 3.846040360312402e-06, "loss": 0.66451478, "num_input_tokens_seen": 26804250, "step": 1264, "time_per_iteration": 2.6720008850097656 }, { "auxiliary_loss_clip": 0.01230982, "auxiliary_loss_mlp": 0.01032255, "balance_loss_clip": 1.0640887, "balance_loss_mlp": 1.02259338, "epoch": 0.15210725665845007, "flos": 28402579641120.0, "grad_norm": 2.5424879748723086, "language_loss": 0.81148666, "learning_rate": 3.8457405097581485e-06, "loss": 0.83411908, "num_input_tokens_seen": 26823240, "step": 1265, "time_per_iteration": 3.3987841606140137 }, { "auxiliary_loss_clip": 0.0117322, "auxiliary_loss_mlp": 0.01034787, "balance_loss_clip": 1.05531859, "balance_loss_mlp": 1.02589965, "epoch": 0.15222749954908915, "flos": 19938182840640.0, "grad_norm": 1.843541403079935, "language_loss": 0.77758324, "learning_rate": 3.8454403792065275e-06, "loss": 0.79966331, "num_input_tokens_seen": 26842060, "step": 1266, "time_per_iteration": 3.4213664531707764 }, { "auxiliary_loss_clip": 0.01174495, "auxiliary_loss_mlp": 0.01032763, "balance_loss_clip": 1.05821002, "balance_loss_mlp": 1.02383387, "epoch": 0.15234774243972826, "flos": 21324226989600.0, "grad_norm": 2.3601118503564327, "language_loss": 0.8547653, "learning_rate": 3.845139968703068e-06, "loss": 0.87683785, "num_input_tokens_seen": 26859580, "step": 1267, "time_per_iteration": 2.6765120029449463 }, { "auxiliary_loss_clip": 0.0116951, "auxiliary_loss_mlp": 0.01044095, "balance_loss_clip": 1.05763841, "balance_loss_mlp": 1.03470731, "epoch": 0.15246798533036734, "flos": 25957816938720.0, "grad_norm": 1.8735993895910328, "language_loss": 0.8286376, "learning_rate": 3.844839278293342e-06, "loss": 0.85077363, "num_input_tokens_seen": 26880430, "step": 1268, "time_per_iteration": 2.711047410964966 }, { "auxiliary_loss_clip": 0.01234355, "auxiliary_loss_mlp": 0.01030581, "balance_loss_clip": 1.06753349, "balance_loss_mlp": 1.02064514, "epoch": 0.15258822822100643, "flos": 25811911883520.0, "grad_norm": 2.243444111718649, "language_loss": 0.76717794, "learning_rate": 3.8445383080229654e-06, "loss": 0.78982729, "num_input_tokens_seen": 26896445, "step": 1269, "time_per_iteration": 3.399487018585205 }, { "auxiliary_loss_clip": 0.01191463, "auxiliary_loss_mlp": 0.01030146, "balance_loss_clip": 1.0585835, "balance_loss_mlp": 1.02037072, "epoch": 0.1527084711116455, "flos": 25265459328000.0, "grad_norm": 1.9318314128005254, "language_loss": 0.73150098, "learning_rate": 3.844237057937593e-06, "loss": 0.75371701, "num_input_tokens_seen": 26915450, "step": 1270, "time_per_iteration": 2.6464009284973145 }, { "auxiliary_loss_clip": 0.01222578, "auxiliary_loss_mlp": 0.01041443, "balance_loss_clip": 1.06471252, "balance_loss_mlp": 1.03117299, "epoch": 0.15282871400228462, "flos": 29240231718720.0, "grad_norm": 2.931678616493101, "language_loss": 0.77900225, "learning_rate": 3.843935528082926e-06, "loss": 0.80164242, "num_input_tokens_seen": 26936475, "step": 1271, "time_per_iteration": 3.492061138153076 }, { "auxiliary_loss_clip": 0.01215165, "auxiliary_loss_mlp": 0.01036813, "balance_loss_clip": 1.06247175, "balance_loss_mlp": 1.02720463, "epoch": 0.1529489568929237, "flos": 20882956044960.0, "grad_norm": 1.824061332642185, "language_loss": 0.84761888, "learning_rate": 3.843633718504704e-06, "loss": 0.87013865, "num_input_tokens_seen": 26954920, "step": 1272, "time_per_iteration": 2.585784912109375 }, { "auxiliary_loss_clip": 0.0119035, "auxiliary_loss_mlp": 0.01032853, "balance_loss_clip": 1.06061673, "balance_loss_mlp": 1.0232327, "epoch": 0.1530691997835628, "flos": 20083836477120.0, "grad_norm": 2.6731232229698034, "language_loss": 0.89951426, "learning_rate": 3.843331629248715e-06, "loss": 0.92174625, "num_input_tokens_seen": 26972520, "step": 1273, "time_per_iteration": 2.613285541534424 }, { "auxiliary_loss_clip": 0.01232293, "auxiliary_loss_mlp": 0.01027433, "balance_loss_clip": 1.06828046, "balance_loss_mlp": 1.01839733, "epoch": 0.1531894426742019, "flos": 28759817676000.0, "grad_norm": 2.4668314745055637, "language_loss": 0.76405418, "learning_rate": 3.843029260360782e-06, "loss": 0.78665143, "num_input_tokens_seen": 26990890, "step": 1274, "time_per_iteration": 2.6063711643218994 }, { "auxiliary_loss_clip": 0.01214705, "auxiliary_loss_mlp": 0.01036012, "balance_loss_clip": 1.06568897, "balance_loss_mlp": 1.02677357, "epoch": 0.15330968556484098, "flos": 22236286231680.0, "grad_norm": 1.821789843890735, "language_loss": 0.7910437, "learning_rate": 3.8427266118867755e-06, "loss": 0.81355083, "num_input_tokens_seen": 27010640, "step": 1275, "time_per_iteration": 2.5835702419281006 }, { "auxiliary_loss_clip": 0.0119679, "auxiliary_loss_mlp": 0.01041792, "balance_loss_clip": 1.06115925, "balance_loss_mlp": 1.0317781, "epoch": 0.15342992845548006, "flos": 27527508479520.0, "grad_norm": 4.564608655645954, "language_loss": 0.82677841, "learning_rate": 3.842423683872608e-06, "loss": 0.84916425, "num_input_tokens_seen": 27031215, "step": 1276, "time_per_iteration": 2.6465413570404053 }, { "auxiliary_loss_clip": 0.01212635, "auxiliary_loss_mlp": 0.01032539, "balance_loss_clip": 1.06115866, "balance_loss_mlp": 1.02248406, "epoch": 0.15355017134611917, "flos": 19609600151520.0, "grad_norm": 7.343704365338558, "language_loss": 0.77942926, "learning_rate": 3.842120476364232e-06, "loss": 0.80188102, "num_input_tokens_seen": 27049665, "step": 1277, "time_per_iteration": 2.5290634632110596 }, { "auxiliary_loss_clip": 0.01219954, "auxiliary_loss_mlp": 0.01039588, "balance_loss_clip": 1.06215334, "balance_loss_mlp": 1.03039074, "epoch": 0.15367041423675826, "flos": 18478593534240.0, "grad_norm": 2.0818483698951495, "language_loss": 0.83390188, "learning_rate": 3.841816989407644e-06, "loss": 0.85649729, "num_input_tokens_seen": 27065155, "step": 1278, "time_per_iteration": 2.5412707328796387 }, { "auxiliary_loss_clip": 0.01177044, "auxiliary_loss_mlp": 0.01034319, "balance_loss_clip": 1.05673504, "balance_loss_mlp": 1.02568817, "epoch": 0.15379065712739734, "flos": 41427667738560.0, "grad_norm": 1.9570159442391664, "language_loss": 0.76973498, "learning_rate": 3.841513223048884e-06, "loss": 0.7918486, "num_input_tokens_seen": 27085840, "step": 1279, "time_per_iteration": 2.778535842895508 }, { "auxiliary_loss_clip": 0.01180648, "auxiliary_loss_mlp": 0.01031756, "balance_loss_clip": 1.05762661, "balance_loss_mlp": 1.0214324, "epoch": 0.15391090001803642, "flos": 22054219186080.0, "grad_norm": 1.9082531835891834, "language_loss": 0.78482747, "learning_rate": 3.841209177334031e-06, "loss": 0.80695152, "num_input_tokens_seen": 27104200, "step": 1280, "time_per_iteration": 2.6532976627349854 }, { "auxiliary_loss_clip": 0.01213349, "auxiliary_loss_mlp": 0.01039441, "balance_loss_clip": 1.06387782, "balance_loss_mlp": 1.03075111, "epoch": 0.15403114290867553, "flos": 15450354278880.0, "grad_norm": 1.7266638126112377, "language_loss": 0.74273187, "learning_rate": 3.84090485230921e-06, "loss": 0.76525974, "num_input_tokens_seen": 27122440, "step": 1281, "time_per_iteration": 2.579768657684326 }, { "auxiliary_loss_clip": 0.01227767, "auxiliary_loss_mlp": 0.01039774, "balance_loss_clip": 1.06373, "balance_loss_mlp": 1.02943814, "epoch": 0.15415138579931462, "flos": 17929159871040.0, "grad_norm": 5.458325480869878, "language_loss": 0.7640723, "learning_rate": 3.840600248020588e-06, "loss": 0.78674775, "num_input_tokens_seen": 27139380, "step": 1282, "time_per_iteration": 2.514479398727417 }, { "auxiliary_loss_clip": 0.01208207, "auxiliary_loss_mlp": 0.01033315, "balance_loss_clip": 1.0608449, "balance_loss_mlp": 1.02306855, "epoch": 0.1542716286899537, "flos": 11429327232000.0, "grad_norm": 2.0301374935218126, "language_loss": 0.80189955, "learning_rate": 3.840295364514371e-06, "loss": 0.82431477, "num_input_tokens_seen": 27156760, "step": 1283, "time_per_iteration": 2.6120800971984863 }, { "auxiliary_loss_clip": 0.01203689, "auxiliary_loss_mlp": 0.01035675, "balance_loss_clip": 1.06240678, "balance_loss_mlp": 1.02586377, "epoch": 0.1543918715805928, "flos": 17420341901280.0, "grad_norm": 2.13779507569265, "language_loss": 0.78213727, "learning_rate": 3.83999020183681e-06, "loss": 0.80453092, "num_input_tokens_seen": 27175455, "step": 1284, "time_per_iteration": 2.580746650695801 }, { "auxiliary_loss_clip": 0.01150078, "auxiliary_loss_mlp": 0.01038137, "balance_loss_clip": 1.0556252, "balance_loss_mlp": 1.02870202, "epoch": 0.1545121144712319, "flos": 17786379591360.0, "grad_norm": 3.0457049814749384, "language_loss": 0.78420067, "learning_rate": 3.839684760034199e-06, "loss": 0.80608284, "num_input_tokens_seen": 27193660, "step": 1285, "time_per_iteration": 2.6992030143737793 }, { "auxiliary_loss_clip": 0.01177686, "auxiliary_loss_mlp": 0.01034239, "balance_loss_clip": 1.0593307, "balance_loss_mlp": 1.02458918, "epoch": 0.15463235736187098, "flos": 28220189342880.0, "grad_norm": 2.8880649711437214, "language_loss": 0.6517024, "learning_rate": 3.8393790391528716e-06, "loss": 0.67382169, "num_input_tokens_seen": 27214355, "step": 1286, "time_per_iteration": 2.677201747894287 }, { "auxiliary_loss_clip": 0.01195647, "auxiliary_loss_mlp": 0.0103584, "balance_loss_clip": 1.0582962, "balance_loss_mlp": 1.02721512, "epoch": 0.15475260025251006, "flos": 22856894532960.0, "grad_norm": 3.189388870741018, "language_loss": 0.89035428, "learning_rate": 3.8390730392392075e-06, "loss": 0.91266906, "num_input_tokens_seen": 27234335, "step": 1287, "time_per_iteration": 2.614237070083618 }, { "auxiliary_loss_clip": 0.01233803, "auxiliary_loss_mlp": 0.01036147, "balance_loss_clip": 1.06639361, "balance_loss_mlp": 1.02682447, "epoch": 0.15487284314314917, "flos": 17602875867360.0, "grad_norm": 2.2511419365690712, "language_loss": 0.79031622, "learning_rate": 3.838766760339626e-06, "loss": 0.8130157, "num_input_tokens_seen": 27252860, "step": 1288, "time_per_iteration": 2.5184366703033447 }, { "auxiliary_loss_clip": 0.01159753, "auxiliary_loss_mlp": 0.01030515, "balance_loss_clip": 1.05261445, "balance_loss_mlp": 1.02121675, "epoch": 0.15499308603378825, "flos": 20082040629120.0, "grad_norm": 2.6619139618732475, "language_loss": 0.79260755, "learning_rate": 3.838460202500587e-06, "loss": 0.81451029, "num_input_tokens_seen": 27268650, "step": 1289, "time_per_iteration": 2.636092185974121 }, { "auxiliary_loss_clip": 0.01178918, "auxiliary_loss_mlp": 0.01044521, "balance_loss_clip": 1.05936325, "balance_loss_mlp": 1.03442407, "epoch": 0.15511332892442733, "flos": 15918053717760.0, "grad_norm": 2.055800529036839, "language_loss": 0.73953605, "learning_rate": 3.838153365768599e-06, "loss": 0.76177043, "num_input_tokens_seen": 27285160, "step": 1290, "time_per_iteration": 2.5920209884643555 }, { "auxiliary_loss_clip": 0.01184214, "auxiliary_loss_mlp": 0.010523, "balance_loss_clip": 1.06484389, "balance_loss_mlp": 1.04276872, "epoch": 0.15523357181506645, "flos": 41282481022560.0, "grad_norm": 2.4582701607309825, "language_loss": 0.75353533, "learning_rate": 3.837846250190206e-06, "loss": 0.77590048, "num_input_tokens_seen": 27308025, "step": 1291, "time_per_iteration": 3.547860622406006 }, { "auxiliary_loss_clip": 0.01163948, "auxiliary_loss_mlp": 0.00763785, "balance_loss_clip": 1.05698419, "balance_loss_mlp": 1.00024271, "epoch": 0.15535381470570553, "flos": 18478773119040.0, "grad_norm": 2.0530859534475763, "language_loss": 0.76774806, "learning_rate": 3.837538855811998e-06, "loss": 0.78702533, "num_input_tokens_seen": 27326200, "step": 1292, "time_per_iteration": 3.3593554496765137 }, { "auxiliary_loss_clip": 0.01204198, "auxiliary_loss_mlp": 0.01034639, "balance_loss_clip": 1.06241155, "balance_loss_mlp": 1.02534723, "epoch": 0.1554740575963446, "flos": 13918153656000.0, "grad_norm": 1.895148631615623, "language_loss": 0.71251655, "learning_rate": 3.837231182680606e-06, "loss": 0.73490489, "num_input_tokens_seen": 27344165, "step": 1293, "time_per_iteration": 2.5877010822296143 }, { "auxiliary_loss_clip": 0.01216239, "auxiliary_loss_mlp": 0.01043748, "balance_loss_clip": 1.06221938, "balance_loss_mlp": 1.03509295, "epoch": 0.1555943004869837, "flos": 20847081390240.0, "grad_norm": 1.7471985225922122, "language_loss": 0.76200831, "learning_rate": 3.836923230842706e-06, "loss": 0.78460813, "num_input_tokens_seen": 27363280, "step": 1294, "time_per_iteration": 2.611440658569336 }, { "auxiliary_loss_clip": 0.01172548, "auxiliary_loss_mlp": 0.0104172, "balance_loss_clip": 1.05547762, "balance_loss_mlp": 1.03191531, "epoch": 0.1557145433776228, "flos": 22085891556480.0, "grad_norm": 2.2556448034768075, "language_loss": 0.80407453, "learning_rate": 3.836615000345011e-06, "loss": 0.82621717, "num_input_tokens_seen": 27381460, "step": 1295, "time_per_iteration": 3.5469956398010254 }, { "auxiliary_loss_clip": 0.01226441, "auxiliary_loss_mlp": 0.01032938, "balance_loss_clip": 1.06295681, "balance_loss_mlp": 1.02402091, "epoch": 0.1558347862682619, "flos": 19791990449760.0, "grad_norm": 2.3199187306443605, "language_loss": 0.78534365, "learning_rate": 3.836306491234282e-06, "loss": 0.80793738, "num_input_tokens_seen": 27399310, "step": 1296, "time_per_iteration": 2.5454583168029785 }, { "auxiliary_loss_clip": 0.01190971, "auxiliary_loss_mlp": 0.01030685, "balance_loss_clip": 1.06054282, "balance_loss_mlp": 1.02152419, "epoch": 0.15595502915890097, "flos": 17237089596000.0, "grad_norm": 2.083309297062416, "language_loss": 0.75497532, "learning_rate": 3.835997703557317e-06, "loss": 0.77719182, "num_input_tokens_seen": 27416050, "step": 1297, "time_per_iteration": 3.4473650455474854 }, { "auxiliary_loss_clip": 0.01166426, "auxiliary_loss_mlp": 0.01036205, "balance_loss_clip": 1.05193496, "balance_loss_mlp": 1.02831972, "epoch": 0.15607527204954008, "flos": 19719522801120.0, "grad_norm": 1.6574575875975393, "language_loss": 0.80269051, "learning_rate": 3.83568863736096e-06, "loss": 0.82471681, "num_input_tokens_seen": 27434920, "step": 1298, "time_per_iteration": 2.6607019901275635 }, { "auxiliary_loss_clip": 0.0118223, "auxiliary_loss_mlp": 0.01034437, "balance_loss_clip": 1.05465913, "balance_loss_mlp": 1.025002, "epoch": 0.15619551494017916, "flos": 18515653448640.0, "grad_norm": 2.62821442590206, "language_loss": 0.89347041, "learning_rate": 3.8353792926920975e-06, "loss": 0.91563708, "num_input_tokens_seen": 27453570, "step": 1299, "time_per_iteration": 2.6327130794525146 }, { "auxiliary_loss_clip": 0.01219681, "auxiliary_loss_mlp": 0.01031964, "balance_loss_clip": 1.06406617, "balance_loss_mlp": 1.02271307, "epoch": 0.15631575783081825, "flos": 19902128601120.0, "grad_norm": 2.1009641352020405, "language_loss": 0.81918108, "learning_rate": 3.835069669597655e-06, "loss": 0.84169751, "num_input_tokens_seen": 27471960, "step": 1300, "time_per_iteration": 2.5659165382385254 }, { "auxiliary_loss_clip": 0.01218527, "auxiliary_loss_mlp": 0.00764169, "balance_loss_clip": 1.06306314, "balance_loss_mlp": 1.0002501, "epoch": 0.15643600072145733, "flos": 20777666683200.0, "grad_norm": 1.9903683692042264, "language_loss": 0.79662013, "learning_rate": 3.834759768124603e-06, "loss": 0.81644702, "num_input_tokens_seen": 27490835, "step": 1301, "time_per_iteration": 2.5858335494995117 }, { "auxiliary_loss_clip": 0.01189669, "auxiliary_loss_mlp": 0.0103444, "balance_loss_clip": 1.06348336, "balance_loss_mlp": 1.02517116, "epoch": 0.15655624361209644, "flos": 18546391978080.0, "grad_norm": 2.1654440500227667, "language_loss": 0.76116067, "learning_rate": 3.834449588319953e-06, "loss": 0.78340179, "num_input_tokens_seen": 27508870, "step": 1302, "time_per_iteration": 2.610802412033081 }, { "auxiliary_loss_clip": 0.01207891, "auxiliary_loss_mlp": 0.01032674, "balance_loss_clip": 1.06405342, "balance_loss_mlp": 1.02428794, "epoch": 0.15667648650273552, "flos": 25229548756320.0, "grad_norm": 1.6964575595426947, "language_loss": 0.85113704, "learning_rate": 3.834139130230758e-06, "loss": 0.87354273, "num_input_tokens_seen": 27528175, "step": 1303, "time_per_iteration": 2.6086950302124023 }, { "auxiliary_loss_clip": 0.01203221, "auxiliary_loss_mlp": 0.01032274, "balance_loss_clip": 1.06062806, "balance_loss_mlp": 1.02325583, "epoch": 0.1567967293933746, "flos": 24827097657120.0, "grad_norm": 1.7045377854438373, "language_loss": 0.81060266, "learning_rate": 3.833828393904117e-06, "loss": 0.83295763, "num_input_tokens_seen": 27548455, "step": 1304, "time_per_iteration": 2.65641450881958 }, { "auxiliary_loss_clip": 0.01163217, "auxiliary_loss_mlp": 0.01032359, "balance_loss_clip": 1.05386257, "balance_loss_mlp": 1.02321553, "epoch": 0.15691697228401372, "flos": 19164557926080.0, "grad_norm": 2.03817025117133, "language_loss": 0.77323115, "learning_rate": 3.833517379387165e-06, "loss": 0.79518688, "num_input_tokens_seen": 27564910, "step": 1305, "time_per_iteration": 2.683288812637329 }, { "auxiliary_loss_clip": 0.0121415, "auxiliary_loss_mlp": 0.010382, "balance_loss_clip": 1.05948567, "balance_loss_mlp": 1.02940786, "epoch": 0.1570372151746528, "flos": 24790935666720.0, "grad_norm": 1.9613601778462704, "language_loss": 0.88729405, "learning_rate": 3.833206086727085e-06, "loss": 0.90981758, "num_input_tokens_seen": 27584260, "step": 1306, "time_per_iteration": 2.604839563369751 }, { "auxiliary_loss_clip": 0.01190301, "auxiliary_loss_mlp": 0.01032575, "balance_loss_clip": 1.05815887, "balance_loss_mlp": 1.02353859, "epoch": 0.15715745806529188, "flos": 24863654734080.0, "grad_norm": 2.1454382969437438, "language_loss": 0.70832682, "learning_rate": 3.8328945159710994e-06, "loss": 0.73055553, "num_input_tokens_seen": 27604440, "step": 1307, "time_per_iteration": 2.6808152198791504 }, { "auxiliary_loss_clip": 0.01225866, "auxiliary_loss_mlp": 0.00763831, "balance_loss_clip": 1.06883717, "balance_loss_mlp": 1.00016856, "epoch": 0.157277700955931, "flos": 21872152140480.0, "grad_norm": 2.2104303037843698, "language_loss": 0.8883425, "learning_rate": 3.832582667166473e-06, "loss": 0.90823936, "num_input_tokens_seen": 27624250, "step": 1308, "time_per_iteration": 2.594386577606201 }, { "auxiliary_loss_clip": 0.01198419, "auxiliary_loss_mlp": 0.01034342, "balance_loss_clip": 1.05887115, "balance_loss_mlp": 1.02420318, "epoch": 0.15739794384657008, "flos": 24533348030880.0, "grad_norm": 1.7136688727178149, "language_loss": 0.81777298, "learning_rate": 3.8322705403605125e-06, "loss": 0.84010053, "num_input_tokens_seen": 27644595, "step": 1309, "time_per_iteration": 2.6286754608154297 }, { "auxiliary_loss_clip": 0.01192032, "auxiliary_loss_mlp": 0.0102981, "balance_loss_clip": 1.06041884, "balance_loss_mlp": 1.02089357, "epoch": 0.15751818673720916, "flos": 17745332894400.0, "grad_norm": 4.705248057835542, "language_loss": 0.81135798, "learning_rate": 3.831958135600568e-06, "loss": 0.83357644, "num_input_tokens_seen": 27662145, "step": 1310, "time_per_iteration": 2.57487154006958 }, { "auxiliary_loss_clip": 0.01215604, "auxiliary_loss_mlp": 0.01036433, "balance_loss_clip": 1.0627892, "balance_loss_mlp": 1.02743888, "epoch": 0.15763842962784824, "flos": 17858523987360.0, "grad_norm": 1.9681424005591683, "language_loss": 0.79194009, "learning_rate": 3.831645452934032e-06, "loss": 0.8144604, "num_input_tokens_seen": 27680575, "step": 1311, "time_per_iteration": 2.5574448108673096 }, { "auxiliary_loss_clip": 0.01231239, "auxiliary_loss_mlp": 0.01047422, "balance_loss_clip": 1.06637728, "balance_loss_mlp": 1.03776598, "epoch": 0.15775867251848735, "flos": 26980912256160.0, "grad_norm": 2.1956590273858567, "language_loss": 0.79869103, "learning_rate": 3.831332492408336e-06, "loss": 0.82147759, "num_input_tokens_seen": 27701985, "step": 1312, "time_per_iteration": 2.5942084789276123 }, { "auxiliary_loss_clip": 0.01192695, "auxiliary_loss_mlp": 0.01038268, "balance_loss_clip": 1.05726349, "balance_loss_mlp": 1.02917862, "epoch": 0.15787891540912644, "flos": 19240401768960.0, "grad_norm": 1.8593989488436617, "language_loss": 0.68935579, "learning_rate": 3.831019254070957e-06, "loss": 0.71166539, "num_input_tokens_seen": 27719770, "step": 1313, "time_per_iteration": 2.608309507369995 }, { "auxiliary_loss_clip": 0.01172013, "auxiliary_loss_mlp": 0.01031052, "balance_loss_clip": 1.05661476, "balance_loss_mlp": 1.02270699, "epoch": 0.15799915829976552, "flos": 27271106103360.0, "grad_norm": 2.475261036654272, "language_loss": 0.95315278, "learning_rate": 3.8307057379694135e-06, "loss": 0.97518337, "num_input_tokens_seen": 27739105, "step": 1314, "time_per_iteration": 2.7018775939941406 }, { "auxiliary_loss_clip": 0.01232591, "auxiliary_loss_mlp": 0.01035558, "balance_loss_clip": 1.06487322, "balance_loss_mlp": 1.0269748, "epoch": 0.15811940119040463, "flos": 20405523109920.0, "grad_norm": 4.2825295039894185, "language_loss": 0.82483268, "learning_rate": 3.830391944151264e-06, "loss": 0.84751415, "num_input_tokens_seen": 27754985, "step": 1315, "time_per_iteration": 2.5693092346191406 }, { "auxiliary_loss_clip": 0.01200638, "auxiliary_loss_mlp": 0.01037925, "balance_loss_clip": 1.05934656, "balance_loss_mlp": 1.02929401, "epoch": 0.1582396440810437, "flos": 32599352348640.0, "grad_norm": 1.952795063437523, "language_loss": 0.67292178, "learning_rate": 3.830077872664114e-06, "loss": 0.69530737, "num_input_tokens_seen": 27776110, "step": 1316, "time_per_iteration": 2.7115936279296875 }, { "auxiliary_loss_clip": 0.01155898, "auxiliary_loss_mlp": 0.01036987, "balance_loss_clip": 1.05256665, "balance_loss_mlp": 1.02819538, "epoch": 0.1583598869716828, "flos": 33800563846080.0, "grad_norm": 1.7518107055925787, "language_loss": 0.72880816, "learning_rate": 3.829763523555604e-06, "loss": 0.75073701, "num_input_tokens_seen": 27796510, "step": 1317, "time_per_iteration": 4.2968909740448 }, { "auxiliary_loss_clip": 0.01207344, "auxiliary_loss_mlp": 0.01033686, "balance_loss_clip": 1.06269896, "balance_loss_mlp": 1.0249778, "epoch": 0.15848012986232188, "flos": 24681336269760.0, "grad_norm": 2.205473494213095, "language_loss": 0.77947366, "learning_rate": 3.829448896873423e-06, "loss": 0.80188394, "num_input_tokens_seen": 27815610, "step": 1318, "time_per_iteration": 2.6071414947509766 }, { "auxiliary_loss_clip": 0.01156901, "auxiliary_loss_mlp": 0.00763641, "balance_loss_clip": 1.05616176, "balance_loss_mlp": 1.00017858, "epoch": 0.158600372752961, "flos": 22602072503040.0, "grad_norm": 2.2115342951097525, "language_loss": 0.79263192, "learning_rate": 3.829133992665299e-06, "loss": 0.81183732, "num_input_tokens_seen": 27834735, "step": 1319, "time_per_iteration": 2.6510567665100098 }, { "auxiliary_loss_clip": 0.01199683, "auxiliary_loss_mlp": 0.01037162, "balance_loss_clip": 1.05755579, "balance_loss_mlp": 1.0281204, "epoch": 0.15872061564360007, "flos": 27927948228960.0, "grad_norm": 7.262715055081856, "language_loss": 0.89080775, "learning_rate": 3.828818810979002e-06, "loss": 0.91317618, "num_input_tokens_seen": 27853065, "step": 1320, "time_per_iteration": 2.6245334148406982 }, { "auxiliary_loss_clip": 0.01227041, "auxiliary_loss_mlp": 0.01035804, "balance_loss_clip": 1.06370687, "balance_loss_mlp": 1.02637422, "epoch": 0.15884085853423915, "flos": 23696809379040.0, "grad_norm": 2.1042593761927804, "language_loss": 0.80852973, "learning_rate": 3.8285033518623454e-06, "loss": 0.83115816, "num_input_tokens_seen": 27873315, "step": 1321, "time_per_iteration": 3.4914283752441406 }, { "auxiliary_loss_clip": 0.01220938, "auxiliary_loss_mlp": 0.01037783, "balance_loss_clip": 1.06583405, "balance_loss_mlp": 1.02781177, "epoch": 0.15896110142487826, "flos": 23112363068160.0, "grad_norm": 3.343785079849156, "language_loss": 0.81132662, "learning_rate": 3.8281876153631845e-06, "loss": 0.83391392, "num_input_tokens_seen": 27890070, "step": 1322, "time_per_iteration": 2.5542423725128174 }, { "auxiliary_loss_clip": 0.01163389, "auxiliary_loss_mlp": 0.01040853, "balance_loss_clip": 1.05393279, "balance_loss_mlp": 1.03182876, "epoch": 0.15908134431551735, "flos": 14685241683840.0, "grad_norm": 1.8057483637926524, "language_loss": 0.64634442, "learning_rate": 3.827871601529416e-06, "loss": 0.66838682, "num_input_tokens_seen": 27908590, "step": 1323, "time_per_iteration": 3.594022512435913 }, { "auxiliary_loss_clip": 0.01178183, "auxiliary_loss_mlp": 0.01038566, "balance_loss_clip": 1.05586457, "balance_loss_mlp": 1.02984571, "epoch": 0.15920158720615643, "flos": 20193615458880.0, "grad_norm": 5.6585708716312935, "language_loss": 0.8071897, "learning_rate": 3.827555310408979e-06, "loss": 0.82935715, "num_input_tokens_seen": 27927985, "step": 1324, "time_per_iteration": 2.6266071796417236 }, { "auxiliary_loss_clip": 0.01179266, "auxiliary_loss_mlp": 0.01029612, "balance_loss_clip": 1.06017852, "balance_loss_mlp": 1.02003336, "epoch": 0.1593218300967955, "flos": 24826630736640.0, "grad_norm": 1.5833707872068365, "language_loss": 0.82721448, "learning_rate": 3.827238742049854e-06, "loss": 0.84930325, "num_input_tokens_seen": 27948280, "step": 1325, "time_per_iteration": 2.673368453979492 }, { "auxiliary_loss_clip": 0.01228572, "auxiliary_loss_mlp": 0.0103357, "balance_loss_clip": 1.06280553, "balance_loss_mlp": 1.02363408, "epoch": 0.15944207298743462, "flos": 28328711231040.0, "grad_norm": 1.9078952651521066, "language_loss": 0.51850212, "learning_rate": 3.826921896500066e-06, "loss": 0.54112351, "num_input_tokens_seen": 27969565, "step": 1326, "time_per_iteration": 2.594592809677124 }, { "auxiliary_loss_clip": 0.01187951, "auxiliary_loss_mlp": 0.0103687, "balance_loss_clip": 1.05807161, "balance_loss_mlp": 1.02710116, "epoch": 0.1595623158780737, "flos": 22964841749760.0, "grad_norm": 1.8440595125738617, "language_loss": 0.77973652, "learning_rate": 3.826604773807678e-06, "loss": 0.80198479, "num_input_tokens_seen": 27987540, "step": 1327, "time_per_iteration": 2.6436638832092285 }, { "auxiliary_loss_clip": 0.01198792, "auxiliary_loss_mlp": 0.01037729, "balance_loss_clip": 1.05718219, "balance_loss_mlp": 1.02701259, "epoch": 0.1596825587687128, "flos": 19710543561120.0, "grad_norm": 3.3565743799850107, "language_loss": 0.73262, "learning_rate": 3.826287374020798e-06, "loss": 0.75498521, "num_input_tokens_seen": 28002345, "step": 1328, "time_per_iteration": 2.5948455333709717 }, { "auxiliary_loss_clip": 0.01230665, "auxiliary_loss_mlp": 0.01035903, "balance_loss_clip": 1.06533098, "balance_loss_mlp": 1.02682495, "epoch": 0.1598028016593519, "flos": 22637731656000.0, "grad_norm": 2.0917207123925614, "language_loss": 0.81596661, "learning_rate": 3.825969697187575e-06, "loss": 0.83863229, "num_input_tokens_seen": 28021675, "step": 1329, "time_per_iteration": 2.5439863204956055 }, { "auxiliary_loss_clip": 0.01177546, "auxiliary_loss_mlp": 0.01033752, "balance_loss_clip": 1.05358911, "balance_loss_mlp": 1.02502012, "epoch": 0.15992304454999098, "flos": 20482911382080.0, "grad_norm": 1.8802608819023814, "language_loss": 0.69701707, "learning_rate": 3.8256517433562015e-06, "loss": 0.71913004, "num_input_tokens_seen": 28039615, "step": 1330, "time_per_iteration": 2.644887924194336 }, { "auxiliary_loss_clip": 0.01226128, "auxiliary_loss_mlp": 0.01037139, "balance_loss_clip": 1.06306493, "balance_loss_mlp": 1.02755451, "epoch": 0.16004328744063007, "flos": 17676097772160.0, "grad_norm": 2.260815553416861, "language_loss": 0.91897494, "learning_rate": 3.82533351257491e-06, "loss": 0.94160759, "num_input_tokens_seen": 28057565, "step": 1331, "time_per_iteration": 2.5179457664489746 }, { "auxiliary_loss_clip": 0.01209279, "auxiliary_loss_mlp": 0.0103915, "balance_loss_clip": 1.06239963, "balance_loss_mlp": 1.02997136, "epoch": 0.16016353033126918, "flos": 24098721723840.0, "grad_norm": 1.9738218751026007, "language_loss": 0.88606805, "learning_rate": 3.825015004891975e-06, "loss": 0.90855235, "num_input_tokens_seen": 28076305, "step": 1332, "time_per_iteration": 2.624457836151123 }, { "auxiliary_loss_clip": 0.01208384, "auxiliary_loss_mlp": 0.01033988, "balance_loss_clip": 1.05981147, "balance_loss_mlp": 1.02521992, "epoch": 0.16028377322190826, "flos": 27634845108000.0, "grad_norm": 1.8012699152940115, "language_loss": 0.75445116, "learning_rate": 3.824696220355716e-06, "loss": 0.77687484, "num_input_tokens_seen": 28097895, "step": 1333, "time_per_iteration": 2.6106245517730713 }, { "auxiliary_loss_clip": 0.01194266, "auxiliary_loss_mlp": 0.01034325, "balance_loss_clip": 1.06107044, "balance_loss_mlp": 1.02522373, "epoch": 0.16040401611254734, "flos": 20961206324160.0, "grad_norm": 1.5836321548466838, "language_loss": 0.78647792, "learning_rate": 3.824377159014491e-06, "loss": 0.8087638, "num_input_tokens_seen": 28118790, "step": 1334, "time_per_iteration": 2.6082141399383545 }, { "auxiliary_loss_clip": 0.01210743, "auxiliary_loss_mlp": 0.01036328, "balance_loss_clip": 1.06195354, "balance_loss_mlp": 1.02712464, "epoch": 0.16052425900318643, "flos": 21247054219200.0, "grad_norm": 1.7119004395608766, "language_loss": 0.84660733, "learning_rate": 3.824057820916702e-06, "loss": 0.86907804, "num_input_tokens_seen": 28135995, "step": 1335, "time_per_iteration": 2.5390281677246094 }, { "auxiliary_loss_clip": 0.0120145, "auxiliary_loss_mlp": 0.01037762, "balance_loss_clip": 1.06191349, "balance_loss_mlp": 1.0282011, "epoch": 0.16064450189382554, "flos": 15524007187200.0, "grad_norm": 2.0484416050430734, "language_loss": 0.71405971, "learning_rate": 3.8237382061107904e-06, "loss": 0.73645186, "num_input_tokens_seen": 28152715, "step": 1336, "time_per_iteration": 2.5959978103637695 }, { "auxiliary_loss_clip": 0.01128237, "auxiliary_loss_mlp": 0.01039749, "balance_loss_clip": 1.04840446, "balance_loss_mlp": 1.03043914, "epoch": 0.16076474478446462, "flos": 21178501519200.0, "grad_norm": 1.9433975561205634, "language_loss": 0.78701782, "learning_rate": 3.823418314645243e-06, "loss": 0.8086977, "num_input_tokens_seen": 28171590, "step": 1337, "time_per_iteration": 2.722250461578369 }, { "auxiliary_loss_clip": 0.01149204, "auxiliary_loss_mlp": 0.01037026, "balance_loss_clip": 1.05372119, "balance_loss_mlp": 1.02777553, "epoch": 0.1608849876751037, "flos": 18366479950080.0, "grad_norm": 1.8087996810472005, "language_loss": 0.75411648, "learning_rate": 3.823098146568588e-06, "loss": 0.7759788, "num_input_tokens_seen": 28191295, "step": 1338, "time_per_iteration": 2.675149440765381 }, { "auxiliary_loss_clip": 0.01211714, "auxiliary_loss_mlp": 0.0103576, "balance_loss_clip": 1.06160343, "balance_loss_mlp": 1.0269084, "epoch": 0.1610052305657428, "flos": 29497029181440.0, "grad_norm": 2.3294657632434816, "language_loss": 0.71131086, "learning_rate": 3.822777701929394e-06, "loss": 0.73378557, "num_input_tokens_seen": 28213120, "step": 1339, "time_per_iteration": 2.6388752460479736 }, { "auxiliary_loss_clip": 0.01200337, "auxiliary_loss_mlp": 0.01028843, "balance_loss_clip": 1.05756152, "balance_loss_mlp": 1.02022374, "epoch": 0.1611254734563819, "flos": 26797875452640.0, "grad_norm": 1.8999557516895818, "language_loss": 0.73310995, "learning_rate": 3.8224569807762714e-06, "loss": 0.75540173, "num_input_tokens_seen": 28232440, "step": 1340, "time_per_iteration": 2.6255085468292236 }, { "auxiliary_loss_clip": 0.01146732, "auxiliary_loss_mlp": 0.01034997, "balance_loss_clip": 1.04810238, "balance_loss_mlp": 1.02577019, "epoch": 0.16124571634702098, "flos": 22419574453920.0, "grad_norm": 2.829484243452968, "language_loss": 0.76088428, "learning_rate": 3.822135983157873e-06, "loss": 0.78270161, "num_input_tokens_seen": 28251715, "step": 1341, "time_per_iteration": 2.664353609085083 }, { "auxiliary_loss_clip": 0.01226733, "auxiliary_loss_mlp": 0.00764125, "balance_loss_clip": 1.06384706, "balance_loss_mlp": 1.00010324, "epoch": 0.16136595923766006, "flos": 10999118711040.0, "grad_norm": 2.1060633670421596, "language_loss": 0.84236121, "learning_rate": 3.821814709122896e-06, "loss": 0.86226976, "num_input_tokens_seen": 28269765, "step": 1342, "time_per_iteration": 2.5558762550354004 }, { "auxiliary_loss_clip": 0.01196166, "auxiliary_loss_mlp": 0.0103399, "balance_loss_clip": 1.06077373, "balance_loss_mlp": 1.02531779, "epoch": 0.16148620212829917, "flos": 21214986762240.0, "grad_norm": 2.2100330799275025, "language_loss": 0.85041237, "learning_rate": 3.821493158720076e-06, "loss": 0.87271392, "num_input_tokens_seen": 28288870, "step": 1343, "time_per_iteration": 4.157044172286987 }, { "auxiliary_loss_clip": 0.01184406, "auxiliary_loss_mlp": 0.01036516, "balance_loss_clip": 1.0574187, "balance_loss_mlp": 1.02681875, "epoch": 0.16160644501893826, "flos": 16758471401280.0, "grad_norm": 3.4949309891553084, "language_loss": 0.73451161, "learning_rate": 3.821171331998191e-06, "loss": 0.75672084, "num_input_tokens_seen": 28305400, "step": 1344, "time_per_iteration": 2.593078851699829 }, { "auxiliary_loss_clip": 0.01086462, "auxiliary_loss_mlp": 0.01004241, "balance_loss_clip": 1.02109981, "balance_loss_mlp": 1.00223875, "epoch": 0.16172668790957734, "flos": 64444977778560.0, "grad_norm": 0.7125450657021546, "language_loss": 0.54536712, "learning_rate": 3.820849229006064e-06, "loss": 0.56627417, "num_input_tokens_seen": 28373150, "step": 1345, "time_per_iteration": 3.343367338180542 }, { "auxiliary_loss_clip": 0.01227552, "auxiliary_loss_mlp": 0.0103257, "balance_loss_clip": 1.06356645, "balance_loss_mlp": 1.02292609, "epoch": 0.16184693080021645, "flos": 23257693452000.0, "grad_norm": 1.84178811144188, "language_loss": 0.70498663, "learning_rate": 3.8205268497925564e-06, "loss": 0.72758782, "num_input_tokens_seen": 28393620, "step": 1346, "time_per_iteration": 2.558145046234131 }, { "auxiliary_loss_clip": 0.01230096, "auxiliary_loss_mlp": 0.01034049, "balance_loss_clip": 1.06552231, "balance_loss_mlp": 1.02468479, "epoch": 0.16196717369085553, "flos": 17451116347680.0, "grad_norm": 2.2817484237399683, "language_loss": 0.7855643, "learning_rate": 3.8202041944065725e-06, "loss": 0.80820584, "num_input_tokens_seen": 28409440, "step": 1347, "time_per_iteration": 3.442725419998169 }, { "auxiliary_loss_clip": 0.01229974, "auxiliary_loss_mlp": 0.01031853, "balance_loss_clip": 1.06544971, "balance_loss_mlp": 1.02318084, "epoch": 0.16208741658149461, "flos": 23873381129760.0, "grad_norm": 2.0526225396761086, "language_loss": 0.74010313, "learning_rate": 3.819881262897061e-06, "loss": 0.76272142, "num_input_tokens_seen": 28427575, "step": 1348, "time_per_iteration": 2.5682311058044434 }, { "auxiliary_loss_clip": 0.01187496, "auxiliary_loss_mlp": 0.01039502, "balance_loss_clip": 1.0636754, "balance_loss_mlp": 1.02998281, "epoch": 0.1622076594721337, "flos": 25884810535680.0, "grad_norm": 2.0998033508453626, "language_loss": 0.73446047, "learning_rate": 3.819558055313008e-06, "loss": 0.7567305, "num_input_tokens_seen": 28448260, "step": 1349, "time_per_iteration": 3.566743850708008 }, { "auxiliary_loss_clip": 0.01220938, "auxiliary_loss_mlp": 0.01042135, "balance_loss_clip": 1.06454909, "balance_loss_mlp": 1.03291416, "epoch": 0.1623279023627728, "flos": 21539762253600.0, "grad_norm": 2.388264877123052, "language_loss": 0.77328455, "learning_rate": 3.819234571703444e-06, "loss": 0.79591525, "num_input_tokens_seen": 28467085, "step": 1350, "time_per_iteration": 2.5935263633728027 }, { "auxiliary_loss_clip": 0.01208635, "auxiliary_loss_mlp": 0.01042683, "balance_loss_clip": 1.06114483, "balance_loss_mlp": 1.03327775, "epoch": 0.1624481452534119, "flos": 22085424636000.0, "grad_norm": 1.9013845854589464, "language_loss": 0.85285902, "learning_rate": 3.8189108121174435e-06, "loss": 0.87537223, "num_input_tokens_seen": 28486850, "step": 1351, "time_per_iteration": 2.5752053260803223 }, { "auxiliary_loss_clip": 0.01179052, "auxiliary_loss_mlp": 0.01038123, "balance_loss_clip": 1.06083465, "balance_loss_mlp": 1.02906919, "epoch": 0.16256838814405097, "flos": 27087494628480.0, "grad_norm": 1.8534950262979928, "language_loss": 0.83752394, "learning_rate": 3.818586776604118e-06, "loss": 0.85969567, "num_input_tokens_seen": 28507490, "step": 1352, "time_per_iteration": 2.662947177886963 }, { "auxiliary_loss_clip": 0.01195125, "auxiliary_loss_mlp": 0.01036397, "balance_loss_clip": 1.05998981, "balance_loss_mlp": 1.02771294, "epoch": 0.16268863103469008, "flos": 20120357637120.0, "grad_norm": 2.9460326516094146, "language_loss": 0.61436415, "learning_rate": 3.818262465212625e-06, "loss": 0.63667941, "num_input_tokens_seen": 28527615, "step": 1353, "time_per_iteration": 2.5962438583374023 }, { "auxiliary_loss_clip": 0.01203982, "auxiliary_loss_mlp": 0.01036928, "balance_loss_clip": 1.06119895, "balance_loss_mlp": 1.02731419, "epoch": 0.16280887392532917, "flos": 18332796229920.0, "grad_norm": 1.9438440577128033, "language_loss": 0.77008557, "learning_rate": 3.817937877992161e-06, "loss": 0.79249465, "num_input_tokens_seen": 28544910, "step": 1354, "time_per_iteration": 2.5544798374176025 }, { "auxiliary_loss_clip": 0.01182179, "auxiliary_loss_mlp": 0.00764999, "balance_loss_clip": 1.05523348, "balance_loss_mlp": 0.99999809, "epoch": 0.16292911681596825, "flos": 11874333540480.0, "grad_norm": 2.39061373953041, "language_loss": 0.85304767, "learning_rate": 3.817613014991967e-06, "loss": 0.87251943, "num_input_tokens_seen": 28561050, "step": 1355, "time_per_iteration": 2.606281280517578 }, { "auxiliary_loss_clip": 0.01172128, "auxiliary_loss_mlp": 0.01031751, "balance_loss_clip": 1.05436265, "balance_loss_mlp": 1.02216077, "epoch": 0.16304935970660733, "flos": 26103470575200.0, "grad_norm": 2.385936538512896, "language_loss": 0.76573902, "learning_rate": 3.817287876261323e-06, "loss": 0.78777778, "num_input_tokens_seen": 28581385, "step": 1356, "time_per_iteration": 2.6586153507232666 }, { "auxiliary_loss_clip": 0.01195826, "auxiliary_loss_mlp": 0.01037512, "balance_loss_clip": 1.06269574, "balance_loss_mlp": 1.02829075, "epoch": 0.16316960259724644, "flos": 29351950216320.0, "grad_norm": 1.918153923099531, "language_loss": 0.80354756, "learning_rate": 3.816962461849553e-06, "loss": 0.82588094, "num_input_tokens_seen": 28603255, "step": 1357, "time_per_iteration": 2.644925832748413 }, { "auxiliary_loss_clip": 0.01190303, "auxiliary_loss_mlp": 0.01038513, "balance_loss_clip": 1.06082547, "balance_loss_mlp": 1.02940512, "epoch": 0.16328984548788553, "flos": 20886763242720.0, "grad_norm": 1.8601285801759122, "language_loss": 0.8459965, "learning_rate": 3.8166367718060235e-06, "loss": 0.8682847, "num_input_tokens_seen": 28623145, "step": 1358, "time_per_iteration": 2.5948925018310547 }, { "auxiliary_loss_clip": 0.01206602, "auxiliary_loss_mlp": 0.01037461, "balance_loss_clip": 1.05856097, "balance_loss_mlp": 1.02745318, "epoch": 0.1634100883785246, "flos": 18041093870400.0, "grad_norm": 2.265655051208668, "language_loss": 0.76597053, "learning_rate": 3.816310806180139e-06, "loss": 0.78841114, "num_input_tokens_seen": 28641555, "step": 1359, "time_per_iteration": 2.6101884841918945 }, { "auxiliary_loss_clip": 0.01198796, "auxiliary_loss_mlp": 0.01039807, "balance_loss_clip": 1.063079, "balance_loss_mlp": 1.03000855, "epoch": 0.16353033126916372, "flos": 24572131959360.0, "grad_norm": 1.5681992737815103, "language_loss": 0.80646592, "learning_rate": 3.81598456502135e-06, "loss": 0.82885194, "num_input_tokens_seen": 28661575, "step": 1360, "time_per_iteration": 2.5813300609588623 }, { "auxiliary_loss_clip": 0.01192076, "auxiliary_loss_mlp": 0.01037116, "balance_loss_clip": 1.06073821, "balance_loss_mlp": 1.02703047, "epoch": 0.1636505741598028, "flos": 19892897942400.0, "grad_norm": 2.788553512767834, "language_loss": 0.87007171, "learning_rate": 3.8156580483791455e-06, "loss": 0.89236367, "num_input_tokens_seen": 28676765, "step": 1361, "time_per_iteration": 2.5188615322113037 }, { "auxiliary_loss_clip": 0.0122873, "auxiliary_loss_mlp": 0.0103404, "balance_loss_clip": 1.0649873, "balance_loss_mlp": 1.02552819, "epoch": 0.16377081705044189, "flos": 28402615558080.0, "grad_norm": 2.2558899104415406, "language_loss": 0.76788461, "learning_rate": 3.815331256303059e-06, "loss": 0.79051232, "num_input_tokens_seen": 28696795, "step": 1362, "time_per_iteration": 2.5535895824432373 }, { "auxiliary_loss_clip": 0.01180098, "auxiliary_loss_mlp": 0.01041019, "balance_loss_clip": 1.06252646, "balance_loss_mlp": 1.03194678, "epoch": 0.163891059941081, "flos": 21908062712160.0, "grad_norm": 1.9792060668323508, "language_loss": 0.77431214, "learning_rate": 3.815004188842665e-06, "loss": 0.79652327, "num_input_tokens_seen": 28714835, "step": 1363, "time_per_iteration": 2.5784053802490234 }, { "auxiliary_loss_clip": 0.01191761, "auxiliary_loss_mlp": 0.01029968, "balance_loss_clip": 1.05675232, "balance_loss_mlp": 1.02111065, "epoch": 0.16401130283172008, "flos": 26797624033920.0, "grad_norm": 1.5752524126585665, "language_loss": 0.7928282, "learning_rate": 3.814676846047578e-06, "loss": 0.81504548, "num_input_tokens_seen": 28735710, "step": 1364, "time_per_iteration": 2.6252095699310303 }, { "auxiliary_loss_clip": 0.0120892, "auxiliary_loss_mlp": 0.01044338, "balance_loss_clip": 1.0610888, "balance_loss_mlp": 1.03527856, "epoch": 0.16413154572235916, "flos": 32997421578720.0, "grad_norm": 1.883365165519676, "language_loss": 0.69871867, "learning_rate": 3.8143492279674565e-06, "loss": 0.72125125, "num_input_tokens_seen": 28758405, "step": 1365, "time_per_iteration": 2.6689677238464355 }, { "auxiliary_loss_clip": 0.01081335, "auxiliary_loss_mlp": 0.01012813, "balance_loss_clip": 1.0184561, "balance_loss_mlp": 1.01071501, "epoch": 0.16425178861299825, "flos": 40113630706080.0, "grad_norm": 0.8473942550611999, "language_loss": 0.5845902, "learning_rate": 3.8140213346519997e-06, "loss": 0.60553169, "num_input_tokens_seen": 28809000, "step": 1366, "time_per_iteration": 2.9447059631347656 }, { "auxiliary_loss_clip": 0.01172725, "auxiliary_loss_mlp": 0.01032531, "balance_loss_clip": 1.05716062, "balance_loss_mlp": 1.02244592, "epoch": 0.16437203150363736, "flos": 25447490456640.0, "grad_norm": 1.58160420583701, "language_loss": 0.77049077, "learning_rate": 3.813693166150948e-06, "loss": 0.79254335, "num_input_tokens_seen": 28829210, "step": 1367, "time_per_iteration": 2.680232048034668 }, { "auxiliary_loss_clip": 0.01172235, "auxiliary_loss_mlp": 0.01037864, "balance_loss_clip": 1.05710578, "balance_loss_mlp": 1.02780318, "epoch": 0.16449227439427644, "flos": 23476892245920.0, "grad_norm": 6.748239907771768, "language_loss": 0.85461563, "learning_rate": 3.813364722514086e-06, "loss": 0.87671661, "num_input_tokens_seen": 28847545, "step": 1368, "time_per_iteration": 3.399406909942627 }, { "auxiliary_loss_clip": 0.01210718, "auxiliary_loss_mlp": 0.01033828, "balance_loss_clip": 1.06177306, "balance_loss_mlp": 1.02504265, "epoch": 0.16461251728491552, "flos": 13545219909600.0, "grad_norm": 1.9828228779814077, "language_loss": 0.80541182, "learning_rate": 3.8130360037912368e-06, "loss": 0.82785726, "num_input_tokens_seen": 28863990, "step": 1369, "time_per_iteration": 3.2330892086029053 }, { "auxiliary_loss_clip": 0.01208183, "auxiliary_loss_mlp": 0.01033839, "balance_loss_clip": 1.05952692, "balance_loss_mlp": 1.02435637, "epoch": 0.16473276017555463, "flos": 23003302425600.0, "grad_norm": 1.8768079625658047, "language_loss": 0.81829679, "learning_rate": 3.812707010032268e-06, "loss": 0.84071702, "num_input_tokens_seen": 28883045, "step": 1370, "time_per_iteration": 2.5812506675720215 }, { "auxiliary_loss_clip": 0.01217234, "auxiliary_loss_mlp": 0.01037531, "balance_loss_clip": 1.06554914, "balance_loss_mlp": 1.02878737, "epoch": 0.16485300306619372, "flos": 24790684248000.0, "grad_norm": 1.687986098896729, "language_loss": 0.79069585, "learning_rate": 3.8123777412870863e-06, "loss": 0.81324351, "num_input_tokens_seen": 28902545, "step": 1371, "time_per_iteration": 2.58115291595459 }, { "auxiliary_loss_clip": 0.01196383, "auxiliary_loss_mlp": 0.01038077, "balance_loss_clip": 1.05737972, "balance_loss_mlp": 1.02895737, "epoch": 0.1649732459568328, "flos": 21106500791040.0, "grad_norm": 1.7953875412866112, "language_loss": 0.78094578, "learning_rate": 3.812048197605643e-06, "loss": 0.80329037, "num_input_tokens_seen": 28921440, "step": 1372, "time_per_iteration": 2.5998384952545166 }, { "auxiliary_loss_clip": 0.01209024, "auxiliary_loss_mlp": 0.01029411, "balance_loss_clip": 1.06116319, "balance_loss_mlp": 1.02007055, "epoch": 0.16509348884747188, "flos": 20266729612800.0, "grad_norm": 1.8185857484455217, "language_loss": 0.81214678, "learning_rate": 3.8117183790379277e-06, "loss": 0.83453113, "num_input_tokens_seen": 28939890, "step": 1373, "time_per_iteration": 3.4728801250457764 }, { "auxiliary_loss_clip": 0.01222075, "auxiliary_loss_mlp": 0.01033311, "balance_loss_clip": 1.05957139, "balance_loss_mlp": 1.02450728, "epoch": 0.165213731738111, "flos": 11035496203200.0, "grad_norm": 2.9169063218529665, "language_loss": 0.93981075, "learning_rate": 3.811388285633976e-06, "loss": 0.96236467, "num_input_tokens_seen": 28955875, "step": 1374, "time_per_iteration": 2.514012575149536 }, { "auxiliary_loss_clip": 0.0117091, "auxiliary_loss_mlp": 0.0103462, "balance_loss_clip": 1.05679464, "balance_loss_mlp": 1.02485132, "epoch": 0.16533397462875007, "flos": 29972055680160.0, "grad_norm": 2.6009510874132524, "language_loss": 0.6224997, "learning_rate": 3.811057917443861e-06, "loss": 0.64455503, "num_input_tokens_seen": 28975140, "step": 1375, "time_per_iteration": 2.700580358505249 }, { "auxiliary_loss_clip": 0.01097627, "auxiliary_loss_mlp": 0.01007752, "balance_loss_clip": 1.01883984, "balance_loss_mlp": 1.00584495, "epoch": 0.16545421751938916, "flos": 65556775210560.0, "grad_norm": 0.8707658508740795, "language_loss": 0.68397075, "learning_rate": 3.8107272745177e-06, "loss": 0.70502454, "num_input_tokens_seen": 29047470, "step": 1376, "time_per_iteration": 4.18358850479126 }, { "auxiliary_loss_clip": 0.0118275, "auxiliary_loss_mlp": 0.01030786, "balance_loss_clip": 1.05995154, "balance_loss_mlp": 1.02200592, "epoch": 0.16557446041002827, "flos": 22492365355200.0, "grad_norm": 1.979047325754292, "language_loss": 0.7836957, "learning_rate": 3.8103963569056513e-06, "loss": 0.80583102, "num_input_tokens_seen": 29066605, "step": 1377, "time_per_iteration": 2.6513078212738037 }, { "auxiliary_loss_clip": 0.01184263, "auxiliary_loss_mlp": 0.01036288, "balance_loss_clip": 1.05252695, "balance_loss_mlp": 1.02736509, "epoch": 0.16569470330066735, "flos": 24602726820960.0, "grad_norm": 1.521548347110829, "language_loss": 0.8789742, "learning_rate": 3.8100651646579146e-06, "loss": 0.90117967, "num_input_tokens_seen": 29085815, "step": 1378, "time_per_iteration": 2.626216173171997 }, { "auxiliary_loss_clip": 0.0118682, "auxiliary_loss_mlp": 0.01034866, "balance_loss_clip": 1.05470419, "balance_loss_mlp": 1.02602625, "epoch": 0.16581494619130643, "flos": 15006209977440.0, "grad_norm": 1.9540696658297372, "language_loss": 0.92890632, "learning_rate": 3.8097336978247317e-06, "loss": 0.95112318, "num_input_tokens_seen": 29102520, "step": 1379, "time_per_iteration": 2.5946412086486816 }, { "auxiliary_loss_clip": 0.01181269, "auxiliary_loss_mlp": 0.01037502, "balance_loss_clip": 1.05680037, "balance_loss_mlp": 1.02782226, "epoch": 0.16593518908194552, "flos": 17420341901280.0, "grad_norm": 2.043866472350977, "language_loss": 0.89194399, "learning_rate": 3.8094019564563854e-06, "loss": 0.91413164, "num_input_tokens_seen": 29119450, "step": 1380, "time_per_iteration": 2.56095290184021 }, { "auxiliary_loss_clip": 0.01225234, "auxiliary_loss_mlp": 0.00763843, "balance_loss_clip": 1.06298685, "balance_loss_mlp": 1.00011683, "epoch": 0.16605543197258463, "flos": 20412634668000.0, "grad_norm": 2.5782608895351116, "language_loss": 0.75187778, "learning_rate": 3.809069940603201e-06, "loss": 0.77176857, "num_input_tokens_seen": 29137405, "step": 1381, "time_per_iteration": 2.5392415523529053 }, { "auxiliary_loss_clip": 0.01185215, "auxiliary_loss_mlp": 0.0103212, "balance_loss_clip": 1.05940223, "balance_loss_mlp": 1.02330422, "epoch": 0.1661756748632237, "flos": 14209748264640.0, "grad_norm": 2.096279901926579, "language_loss": 0.78176665, "learning_rate": 3.8087376503155452e-06, "loss": 0.80393994, "num_input_tokens_seen": 29154890, "step": 1382, "time_per_iteration": 2.572848320007324 }, { "auxiliary_loss_clip": 0.01089713, "auxiliary_loss_mlp": 0.01002631, "balance_loss_clip": 1.01768303, "balance_loss_mlp": 1.0006876, "epoch": 0.1662959177538628, "flos": 66080887416960.0, "grad_norm": 0.9316787598443166, "language_loss": 0.56286395, "learning_rate": 3.808405085643826e-06, "loss": 0.58378744, "num_input_tokens_seen": 29219770, "step": 1383, "time_per_iteration": 3.21012806892395 }, { "auxiliary_loss_clip": 0.01223352, "auxiliary_loss_mlp": 0.00764414, "balance_loss_clip": 1.06284857, "balance_loss_mlp": 1.00004363, "epoch": 0.1664161606445019, "flos": 20740570851840.0, "grad_norm": 2.1843971450702653, "language_loss": 0.88587737, "learning_rate": 3.8080722466384925e-06, "loss": 0.90575504, "num_input_tokens_seen": 29237620, "step": 1384, "time_per_iteration": 2.522209644317627 }, { "auxiliary_loss_clip": 0.01227114, "auxiliary_loss_mlp": 0.01033094, "balance_loss_clip": 1.06243253, "balance_loss_mlp": 1.02392673, "epoch": 0.166536403535141, "flos": 25260933791040.0, "grad_norm": 2.0222042421982818, "language_loss": 0.70591533, "learning_rate": 3.8077391333500376e-06, "loss": 0.72851741, "num_input_tokens_seen": 29256760, "step": 1385, "time_per_iteration": 2.555199146270752 }, { "auxiliary_loss_clip": 0.01195256, "auxiliary_loss_mlp": 0.01034395, "balance_loss_clip": 1.06099522, "balance_loss_mlp": 1.02564502, "epoch": 0.16665664642578007, "flos": 25447454539680.0, "grad_norm": 2.087034687090771, "language_loss": 0.76535547, "learning_rate": 3.8074057458289934e-06, "loss": 0.78765202, "num_input_tokens_seen": 29277450, "step": 1386, "time_per_iteration": 2.6289877891540527 }, { "auxiliary_loss_clip": 0.01197119, "auxiliary_loss_mlp": 0.01037025, "balance_loss_clip": 1.05923724, "balance_loss_mlp": 1.02767348, "epoch": 0.16677688931641918, "flos": 22200770746560.0, "grad_norm": 2.004430061433992, "language_loss": 0.82359463, "learning_rate": 3.807072084125934e-06, "loss": 0.84593606, "num_input_tokens_seen": 29299300, "step": 1387, "time_per_iteration": 2.605699062347412 }, { "auxiliary_loss_clip": 0.01192798, "auxiliary_loss_mlp": 0.01035757, "balance_loss_clip": 1.06070042, "balance_loss_mlp": 1.02687597, "epoch": 0.16689713220705826, "flos": 16945961907840.0, "grad_norm": 2.434679226774566, "language_loss": 0.8022517, "learning_rate": 3.806738148291477e-06, "loss": 0.82453728, "num_input_tokens_seen": 29316125, "step": 1388, "time_per_iteration": 2.581627130508423 }, { "auxiliary_loss_clip": 0.01150057, "auxiliary_loss_mlp": 0.01041888, "balance_loss_clip": 1.0519762, "balance_loss_mlp": 1.03174973, "epoch": 0.16701737509769735, "flos": 36244428624480.0, "grad_norm": 2.0385467989747594, "language_loss": 0.71680319, "learning_rate": 3.8064039383762793e-06, "loss": 0.73872268, "num_input_tokens_seen": 29338490, "step": 1389, "time_per_iteration": 2.7802159786224365 }, { "auxiliary_loss_clip": 0.01209764, "auxiliary_loss_mlp": 0.01033529, "balance_loss_clip": 1.06396198, "balance_loss_mlp": 1.02480924, "epoch": 0.16713761798833643, "flos": 23258663209920.0, "grad_norm": 2.0717512460107157, "language_loss": 0.77128422, "learning_rate": 3.8060694544310396e-06, "loss": 0.79371715, "num_input_tokens_seen": 29357000, "step": 1390, "time_per_iteration": 2.5588490962982178 }, { "auxiliary_loss_clip": 0.01226456, "auxiliary_loss_mlp": 0.01038362, "balance_loss_clip": 1.06388474, "balance_loss_mlp": 1.02820504, "epoch": 0.16725786087897554, "flos": 25302519242400.0, "grad_norm": 1.7819268449630816, "language_loss": 0.78563952, "learning_rate": 3.8057346965065006e-06, "loss": 0.80828774, "num_input_tokens_seen": 29378230, "step": 1391, "time_per_iteration": 2.5603580474853516 }, { "auxiliary_loss_clip": 0.01193542, "auxiliary_loss_mlp": 0.01035352, "balance_loss_clip": 1.06192076, "balance_loss_mlp": 1.02655435, "epoch": 0.16737810376961462, "flos": 31831545981600.0, "grad_norm": 1.6183621710941163, "language_loss": 0.84508681, "learning_rate": 3.805399664653443e-06, "loss": 0.86737573, "num_input_tokens_seen": 29400370, "step": 1392, "time_per_iteration": 2.6712965965270996 }, { "auxiliary_loss_clip": 0.01226651, "auxiliary_loss_mlp": 0.01030224, "balance_loss_clip": 1.06324959, "balance_loss_mlp": 1.02069354, "epoch": 0.1674983466602537, "flos": 27961847450880.0, "grad_norm": 2.238817579782576, "language_loss": 0.73955107, "learning_rate": 3.805064358922692e-06, "loss": 0.76211977, "num_input_tokens_seen": 29418660, "step": 1393, "time_per_iteration": 2.5581414699554443 }, { "auxiliary_loss_clip": 0.01213301, "auxiliary_loss_mlp": 0.01031235, "balance_loss_clip": 1.06162441, "balance_loss_mlp": 1.02153158, "epoch": 0.16761858955089282, "flos": 21762660494400.0, "grad_norm": 2.064063287773361, "language_loss": 0.81417364, "learning_rate": 3.8047287793651136e-06, "loss": 0.83661902, "num_input_tokens_seen": 29440105, "step": 1394, "time_per_iteration": 2.5804684162139893 }, { "auxiliary_loss_clip": 0.01183573, "auxiliary_loss_mlp": 0.0103347, "balance_loss_clip": 1.05925107, "balance_loss_mlp": 1.02479196, "epoch": 0.1677388324415319, "flos": 23805510852000.0, "grad_norm": 1.8415675087187657, "language_loss": 0.88768339, "learning_rate": 3.8043929260316137e-06, "loss": 0.90985382, "num_input_tokens_seen": 29458260, "step": 1395, "time_per_iteration": 4.131240367889404 }, { "auxiliary_loss_clip": 0.01197983, "auxiliary_loss_mlp": 0.0103409, "balance_loss_clip": 1.06483817, "balance_loss_mlp": 1.02500033, "epoch": 0.16785907533217098, "flos": 20558862975840.0, "grad_norm": 2.0119147989881228, "language_loss": 0.83532739, "learning_rate": 3.8040567989731417e-06, "loss": 0.85764813, "num_input_tokens_seen": 29476205, "step": 1396, "time_per_iteration": 2.5769526958465576 }, { "auxiliary_loss_clip": 0.01206103, "auxiliary_loss_mlp": 0.01031448, "balance_loss_clip": 1.06253362, "balance_loss_mlp": 1.02251363, "epoch": 0.16797931822281006, "flos": 15669660823680.0, "grad_norm": 1.919153902604182, "language_loss": 0.79797179, "learning_rate": 3.8037203982406876e-06, "loss": 0.82034731, "num_input_tokens_seen": 29494370, "step": 1397, "time_per_iteration": 2.561969518661499 }, { "auxiliary_loss_clip": 0.01226582, "auxiliary_loss_mlp": 0.01033854, "balance_loss_clip": 1.06513929, "balance_loss_mlp": 1.02365518, "epoch": 0.16809956111344918, "flos": 16541104372320.0, "grad_norm": 1.828353861660635, "language_loss": 0.73348111, "learning_rate": 3.8033837238852835e-06, "loss": 0.7560854, "num_input_tokens_seen": 29511070, "step": 1398, "time_per_iteration": 2.5044283866882324 }, { "auxiliary_loss_clip": 0.01183703, "auxiliary_loss_mlp": 0.01035255, "balance_loss_clip": 1.05648398, "balance_loss_mlp": 1.0266006, "epoch": 0.16821980400408826, "flos": 23258088538560.0, "grad_norm": 2.1306981500569955, "language_loss": 0.69736457, "learning_rate": 3.8030467759580017e-06, "loss": 0.71955413, "num_input_tokens_seen": 29531990, "step": 1399, "time_per_iteration": 3.4977941513061523 }, { "auxiliary_loss_clip": 0.0121204, "auxiliary_loss_mlp": 0.0103207, "balance_loss_clip": 1.05996108, "balance_loss_mlp": 1.02267075, "epoch": 0.16834004689472734, "flos": 20774757409440.0, "grad_norm": 1.9708620157997891, "language_loss": 0.86913633, "learning_rate": 3.802709554509958e-06, "loss": 0.89157736, "num_input_tokens_seen": 29549790, "step": 1400, "time_per_iteration": 2.547978639602661 }, { "auxiliary_loss_clip": 0.01192568, "auxiliary_loss_mlp": 0.01030718, "balance_loss_clip": 1.05761909, "balance_loss_mlp": 1.02212918, "epoch": 0.16846028978536645, "flos": 26687306297760.0, "grad_norm": 1.6916386909754026, "language_loss": 0.79358691, "learning_rate": 3.8023720595923083e-06, "loss": 0.81581974, "num_input_tokens_seen": 29569045, "step": 1401, "time_per_iteration": 3.4757964611053467 }, { "auxiliary_loss_clip": 0.01162268, "auxiliary_loss_mlp": 0.01034699, "balance_loss_clip": 1.05111837, "balance_loss_mlp": 1.02503693, "epoch": 0.16858053267600553, "flos": 18843302296800.0, "grad_norm": 8.011028820080327, "language_loss": 0.8723861, "learning_rate": 3.80203429125625e-06, "loss": 0.89435577, "num_input_tokens_seen": 29587220, "step": 1402, "time_per_iteration": 2.644697427749634 }, { "auxiliary_loss_clip": 0.01141312, "auxiliary_loss_mlp": 0.01033363, "balance_loss_clip": 1.05239856, "balance_loss_mlp": 1.02478004, "epoch": 0.16870077556664462, "flos": 27744552255840.0, "grad_norm": 2.2232410389925397, "language_loss": 0.70374668, "learning_rate": 3.8016962495530225e-06, "loss": 0.72549343, "num_input_tokens_seen": 29606410, "step": 1403, "time_per_iteration": 2.729588508605957 }, { "auxiliary_loss_clip": 0.01226447, "auxiliary_loss_mlp": 0.0103538, "balance_loss_clip": 1.06307352, "balance_loss_mlp": 1.02620125, "epoch": 0.1688210184572837, "flos": 13730770900320.0, "grad_norm": 2.2347759490412638, "language_loss": 0.77011478, "learning_rate": 3.8013579345339063e-06, "loss": 0.79273307, "num_input_tokens_seen": 29621275, "step": 1404, "time_per_iteration": 2.512690544128418 }, { "auxiliary_loss_clip": 0.0118408, "auxiliary_loss_mlp": 0.01034902, "balance_loss_clip": 1.05926085, "balance_loss_mlp": 1.02551436, "epoch": 0.1689412613479228, "flos": 26468789926080.0, "grad_norm": 2.378726181716378, "language_loss": 0.69516033, "learning_rate": 3.801019346250224e-06, "loss": 0.71735018, "num_input_tokens_seen": 29641420, "step": 1405, "time_per_iteration": 2.6545963287353516 }, { "auxiliary_loss_clip": 0.01206217, "auxiliary_loss_mlp": 0.0103332, "balance_loss_clip": 1.06012356, "balance_loss_mlp": 1.02436757, "epoch": 0.1690615042385619, "flos": 21138855583680.0, "grad_norm": 3.0495051160292816, "language_loss": 0.83797574, "learning_rate": 3.8006804847533395e-06, "loss": 0.86037111, "num_input_tokens_seen": 29660935, "step": 1406, "time_per_iteration": 2.5846786499023438 }, { "auxiliary_loss_clip": 0.0122316, "auxiliary_loss_mlp": 0.01036036, "balance_loss_clip": 1.06116724, "balance_loss_mlp": 1.02687502, "epoch": 0.16918174712920098, "flos": 20849344158720.0, "grad_norm": 2.84709828356114, "language_loss": 0.85469025, "learning_rate": 3.8003413500946556e-06, "loss": 0.8772822, "num_input_tokens_seen": 29681045, "step": 1407, "time_per_iteration": 2.5265753269195557 }, { "auxiliary_loss_clip": 0.01197449, "auxiliary_loss_mlp": 0.0103674, "balance_loss_clip": 1.0597322, "balance_loss_mlp": 1.02774024, "epoch": 0.1693019900198401, "flos": 16983273240960.0, "grad_norm": 2.4815488080555252, "language_loss": 0.83805239, "learning_rate": 3.8000019423256216e-06, "loss": 0.8603943, "num_input_tokens_seen": 29698810, "step": 1408, "time_per_iteration": 2.578875780105591 }, { "auxiliary_loss_clip": 0.01181295, "auxiliary_loss_mlp": 0.01037088, "balance_loss_clip": 1.05698657, "balance_loss_mlp": 1.02694345, "epoch": 0.16942223291047917, "flos": 26796905694720.0, "grad_norm": 1.5842160031593866, "language_loss": 0.88037896, "learning_rate": 3.7996622614977234e-06, "loss": 0.90256274, "num_input_tokens_seen": 29720000, "step": 1409, "time_per_iteration": 2.711575984954834 }, { "auxiliary_loss_clip": 0.01192008, "auxiliary_loss_mlp": 0.01038354, "balance_loss_clip": 1.05962348, "balance_loss_mlp": 1.02899587, "epoch": 0.16954247580111825, "flos": 18583703311200.0, "grad_norm": 1.964000367111886, "language_loss": 0.79040039, "learning_rate": 3.799322307662492e-06, "loss": 0.81270397, "num_input_tokens_seen": 29737820, "step": 1410, "time_per_iteration": 2.580667018890381 }, { "auxiliary_loss_clip": 0.01165235, "auxiliary_loss_mlp": 0.01034651, "balance_loss_clip": 1.05395555, "balance_loss_mlp": 1.02548373, "epoch": 0.16966271869175734, "flos": 13983653414400.0, "grad_norm": 3.0256204722720965, "language_loss": 0.83861804, "learning_rate": 3.798982080871496e-06, "loss": 0.86061692, "num_input_tokens_seen": 29752960, "step": 1411, "time_per_iteration": 2.6157524585723877 }, { "auxiliary_loss_clip": 0.01223773, "auxiliary_loss_mlp": 0.01033775, "balance_loss_clip": 1.06114841, "balance_loss_mlp": 1.02432203, "epoch": 0.16978296158239645, "flos": 37487333324160.0, "grad_norm": 2.263928650609002, "language_loss": 0.67941499, "learning_rate": 3.798641581176349e-06, "loss": 0.70199049, "num_input_tokens_seen": 29775240, "step": 1412, "time_per_iteration": 2.6757960319519043 }, { "auxiliary_loss_clip": 0.01196201, "auxiliary_loss_mlp": 0.01035588, "balance_loss_clip": 1.05834866, "balance_loss_mlp": 1.02643907, "epoch": 0.16990320447303553, "flos": 28328962649760.0, "grad_norm": 1.7274219585905466, "language_loss": 0.7453959, "learning_rate": 3.7983008086287044e-06, "loss": 0.76771379, "num_input_tokens_seen": 29796560, "step": 1413, "time_per_iteration": 2.639925479888916 }, { "auxiliary_loss_clip": 0.01192886, "auxiliary_loss_mlp": 0.01034635, "balance_loss_clip": 1.05827785, "balance_loss_mlp": 1.02529526, "epoch": 0.1700234473636746, "flos": 20188191997920.0, "grad_norm": 1.9263669551346003, "language_loss": 0.7955302, "learning_rate": 3.797959763280257e-06, "loss": 0.81780541, "num_input_tokens_seen": 29815245, "step": 1414, "time_per_iteration": 2.599946975708008 }, { "auxiliary_loss_clip": 0.01212603, "auxiliary_loss_mlp": 0.01040961, "balance_loss_clip": 1.06159067, "balance_loss_mlp": 1.03176403, "epoch": 0.17014369025431372, "flos": 24858662276640.0, "grad_norm": 1.952509633560162, "language_loss": 0.79191476, "learning_rate": 3.797618445182743e-06, "loss": 0.81445038, "num_input_tokens_seen": 29836640, "step": 1415, "time_per_iteration": 2.588313102722168 }, { "auxiliary_loss_clip": 0.01160348, "auxiliary_loss_mlp": 0.01031916, "balance_loss_clip": 1.0534277, "balance_loss_mlp": 1.02324939, "epoch": 0.1702639331449528, "flos": 16467235962240.0, "grad_norm": 2.000109283975879, "language_loss": 0.8471036, "learning_rate": 3.79727685438794e-06, "loss": 0.86902618, "num_input_tokens_seen": 29850830, "step": 1416, "time_per_iteration": 2.6477925777435303 }, { "auxiliary_loss_clip": 0.01108281, "auxiliary_loss_mlp": 0.0100124, "balance_loss_clip": 1.01801765, "balance_loss_mlp": 0.99947613, "epoch": 0.1703841760355919, "flos": 52508879843520.0, "grad_norm": 0.8324769897056371, "language_loss": 0.61646402, "learning_rate": 3.796934990947667e-06, "loss": 0.63755924, "num_input_tokens_seen": 29912515, "step": 1417, "time_per_iteration": 3.148560047149658 }, { "auxiliary_loss_clip": 0.01105801, "auxiliary_loss_mlp": 0.01001225, "balance_loss_clip": 1.01704073, "balance_loss_mlp": 0.99953222, "epoch": 0.170504418926231, "flos": 49370646104640.0, "grad_norm": 0.8748023780189997, "language_loss": 0.62434244, "learning_rate": 3.7965928549137854e-06, "loss": 0.64541268, "num_input_tokens_seen": 29969330, "step": 1418, "time_per_iteration": 3.0832979679107666 }, { "auxiliary_loss_clip": 0.0118665, "auxiliary_loss_mlp": 0.01031916, "balance_loss_clip": 1.05533159, "balance_loss_mlp": 1.0216639, "epoch": 0.17062466181687008, "flos": 25849223216640.0, "grad_norm": 2.6789246262038846, "language_loss": 0.77186275, "learning_rate": 3.7962504463381953e-06, "loss": 0.79404837, "num_input_tokens_seen": 29990820, "step": 1419, "time_per_iteration": 2.6892402172088623 }, { "auxiliary_loss_clip": 0.0118615, "auxiliary_loss_mlp": 0.00765101, "balance_loss_clip": 1.05820727, "balance_loss_mlp": 1.00023973, "epoch": 0.17074490470750917, "flos": 20960416151040.0, "grad_norm": 1.9196514772851272, "language_loss": 0.78668368, "learning_rate": 3.7959077652728412e-06, "loss": 0.80619615, "num_input_tokens_seen": 30009275, "step": 1420, "time_per_iteration": 2.59586763381958 }, { "auxiliary_loss_clip": 0.01196813, "auxiliary_loss_mlp": 0.01037289, "balance_loss_clip": 1.05883861, "balance_loss_mlp": 1.02746606, "epoch": 0.17086514759814825, "flos": 20959769645760.0, "grad_norm": 2.3475085916571636, "language_loss": 0.77629626, "learning_rate": 3.795564811769707e-06, "loss": 0.79863727, "num_input_tokens_seen": 30027630, "step": 1421, "time_per_iteration": 3.4092564582824707 }, { "auxiliary_loss_clip": 0.01195753, "auxiliary_loss_mlp": 0.01037027, "balance_loss_clip": 1.06313789, "balance_loss_mlp": 1.02728176, "epoch": 0.17098539048878736, "flos": 28474077531840.0, "grad_norm": 1.9238060096812968, "language_loss": 0.78112292, "learning_rate": 3.795221585880818e-06, "loss": 0.8034507, "num_input_tokens_seen": 30048310, "step": 1422, "time_per_iteration": 2.634558916091919 }, { "auxiliary_loss_clip": 0.01180822, "auxiliary_loss_mlp": 0.01043592, "balance_loss_clip": 1.06314301, "balance_loss_mlp": 1.03530741, "epoch": 0.17110563337942644, "flos": 16290017706240.0, "grad_norm": 2.0270011663684455, "language_loss": 0.91364539, "learning_rate": 3.794878087658242e-06, "loss": 0.93588954, "num_input_tokens_seen": 30066080, "step": 1423, "time_per_iteration": 2.6054611206054688 }, { "auxiliary_loss_clip": 0.01211876, "auxiliary_loss_mlp": 0.01031952, "balance_loss_clip": 1.05995345, "balance_loss_mlp": 1.02288055, "epoch": 0.17122587627006552, "flos": 29674211520480.0, "grad_norm": 2.189062471489937, "language_loss": 0.78470117, "learning_rate": 3.7945343171540873e-06, "loss": 0.8071394, "num_input_tokens_seen": 30086955, "step": 1424, "time_per_iteration": 2.6208229064941406 }, { "auxiliary_loss_clip": 0.0122767, "auxiliary_loss_mlp": 0.01043019, "balance_loss_clip": 1.06389773, "balance_loss_mlp": 1.03373826, "epoch": 0.17134611916070464, "flos": 25338393897120.0, "grad_norm": 1.8408197856888353, "language_loss": 0.78690004, "learning_rate": 3.7941902744205033e-06, "loss": 0.80960691, "num_input_tokens_seen": 30107990, "step": 1425, "time_per_iteration": 3.479473829269409 }, { "auxiliary_loss_clip": 0.0119914, "auxiliary_loss_mlp": 0.01030661, "balance_loss_clip": 1.05922318, "balance_loss_mlp": 1.02100468, "epoch": 0.17146636205134372, "flos": 13953848725920.0, "grad_norm": 2.7205827264567355, "language_loss": 0.83417916, "learning_rate": 3.7938459595096817e-06, "loss": 0.85647714, "num_input_tokens_seen": 30126535, "step": 1426, "time_per_iteration": 2.6069509983062744 }, { "auxiliary_loss_clip": 0.01217823, "auxiliary_loss_mlp": 0.01039029, "balance_loss_clip": 1.06145036, "balance_loss_mlp": 1.02939057, "epoch": 0.1715866049419828, "flos": 23915217999840.0, "grad_norm": 1.773495429534251, "language_loss": 0.86126167, "learning_rate": 3.7935013724738545e-06, "loss": 0.88383019, "num_input_tokens_seen": 30147035, "step": 1427, "time_per_iteration": 3.4760265350341797 }, { "auxiliary_loss_clip": 0.01207535, "auxiliary_loss_mlp": 0.01033161, "balance_loss_clip": 1.06293774, "balance_loss_mlp": 1.0233798, "epoch": 0.17170684783262188, "flos": 22709373214560.0, "grad_norm": 1.9631781608455556, "language_loss": 0.77958512, "learning_rate": 3.7931565133652945e-06, "loss": 0.80199212, "num_input_tokens_seen": 30167110, "step": 1428, "time_per_iteration": 2.5860865116119385 }, { "auxiliary_loss_clip": 0.01223613, "auxiliary_loss_mlp": 0.01040702, "balance_loss_clip": 1.06063247, "balance_loss_mlp": 1.03147507, "epoch": 0.171827090723261, "flos": 26613294219840.0, "grad_norm": 2.081438571405592, "language_loss": 0.67450047, "learning_rate": 3.792811382236317e-06, "loss": 0.69714355, "num_input_tokens_seen": 30185620, "step": 1429, "time_per_iteration": 2.550386905670166 }, { "auxiliary_loss_clip": 0.01215059, "auxiliary_loss_mlp": 0.01033071, "balance_loss_clip": 1.06168866, "balance_loss_mlp": 1.02340281, "epoch": 0.17194733361390008, "flos": 28148511867360.0, "grad_norm": 1.737976402719043, "language_loss": 0.78431028, "learning_rate": 3.792465979139279e-06, "loss": 0.80679154, "num_input_tokens_seen": 30208225, "step": 1430, "time_per_iteration": 2.6175649166107178 }, { "auxiliary_loss_clip": 0.01079763, "auxiliary_loss_mlp": 0.01007209, "balance_loss_clip": 1.01666069, "balance_loss_mlp": 1.00537324, "epoch": 0.17206757650453916, "flos": 65530699497600.0, "grad_norm": 0.9449949400173457, "language_loss": 0.65644276, "learning_rate": 3.792120304126576e-06, "loss": 0.67731249, "num_input_tokens_seen": 30271600, "step": 1431, "time_per_iteration": 3.233307361602783 }, { "auxiliary_loss_clip": 0.01140874, "auxiliary_loss_mlp": 0.01032824, "balance_loss_clip": 1.05374289, "balance_loss_mlp": 1.02301359, "epoch": 0.17218781939517827, "flos": 22273489813920.0, "grad_norm": 2.6906429878844684, "language_loss": 0.83494419, "learning_rate": 3.791774357250649e-06, "loss": 0.85668111, "num_input_tokens_seen": 30290430, "step": 1432, "time_per_iteration": 2.7236664295196533 }, { "auxiliary_loss_clip": 0.01193248, "auxiliary_loss_mlp": 0.01033687, "balance_loss_clip": 1.05941606, "balance_loss_mlp": 1.02493751, "epoch": 0.17230806228581735, "flos": 14137316532960.0, "grad_norm": 2.1175091207598116, "language_loss": 0.78823966, "learning_rate": 3.7914281385639757e-06, "loss": 0.81050897, "num_input_tokens_seen": 30308305, "step": 1433, "time_per_iteration": 2.5952463150024414 }, { "auxiliary_loss_clip": 0.01210917, "auxiliary_loss_mlp": 0.0103337, "balance_loss_clip": 1.05930734, "balance_loss_mlp": 1.02451873, "epoch": 0.17242830517645644, "flos": 20704839864960.0, "grad_norm": 1.836685035101151, "language_loss": 0.79582185, "learning_rate": 3.7910816481190784e-06, "loss": 0.81826472, "num_input_tokens_seen": 30328120, "step": 1434, "time_per_iteration": 2.5645391941070557 }, { "auxiliary_loss_clip": 0.01184085, "auxiliary_loss_mlp": 0.01033913, "balance_loss_clip": 1.05470562, "balance_loss_mlp": 1.02438188, "epoch": 0.17254854806709552, "flos": 30774587359200.0, "grad_norm": 1.90536965404805, "language_loss": 0.74895626, "learning_rate": 3.7907348859685193e-06, "loss": 0.77113622, "num_input_tokens_seen": 30349825, "step": 1435, "time_per_iteration": 2.667623519897461 }, { "auxiliary_loss_clip": 0.01205762, "auxiliary_loss_mlp": 0.01039888, "balance_loss_clip": 1.06144571, "balance_loss_mlp": 1.03120959, "epoch": 0.17266879095773463, "flos": 26614730898240.0, "grad_norm": 1.8711073385160302, "language_loss": 0.80150235, "learning_rate": 3.790387852164902e-06, "loss": 0.82395887, "num_input_tokens_seen": 30370555, "step": 1436, "time_per_iteration": 2.5909180641174316 }, { "auxiliary_loss_clip": 0.01213185, "auxiliary_loss_mlp": 0.01042724, "balance_loss_clip": 1.06267166, "balance_loss_mlp": 1.03341973, "epoch": 0.1727890338483737, "flos": 20266298609280.0, "grad_norm": 1.9112673073804327, "language_loss": 0.76633918, "learning_rate": 3.7900405467608707e-06, "loss": 0.78889823, "num_input_tokens_seen": 30390100, "step": 1437, "time_per_iteration": 2.551964521408081 }, { "auxiliary_loss_clip": 0.01150286, "auxiliary_loss_mlp": 0.01043021, "balance_loss_clip": 1.04910862, "balance_loss_mlp": 1.03306723, "epoch": 0.1729092767390128, "flos": 18179815533600.0, "grad_norm": 2.748210477994246, "language_loss": 0.79830563, "learning_rate": 3.7896929698091114e-06, "loss": 0.82023871, "num_input_tokens_seen": 30402915, "step": 1438, "time_per_iteration": 2.6160194873809814 }, { "auxiliary_loss_clip": 0.01229505, "auxiliary_loss_mlp": 0.01039763, "balance_loss_clip": 1.06573844, "balance_loss_mlp": 1.03052402, "epoch": 0.1730295196296519, "flos": 26759522527680.0, "grad_norm": 2.716863581828206, "language_loss": 0.68006086, "learning_rate": 3.7893451213623518e-06, "loss": 0.70275354, "num_input_tokens_seen": 30420145, "step": 1439, "time_per_iteration": 2.55991792678833 }, { "auxiliary_loss_clip": 0.01210424, "auxiliary_loss_mlp": 0.00764589, "balance_loss_clip": 1.06300509, "balance_loss_mlp": 1.00022781, "epoch": 0.173149762520291, "flos": 23842534849440.0, "grad_norm": 2.779727577004632, "language_loss": 0.82202911, "learning_rate": 3.7889970014733606e-06, "loss": 0.84177923, "num_input_tokens_seen": 30439250, "step": 1440, "time_per_iteration": 2.5731492042541504 }, { "auxiliary_loss_clip": 0.01145401, "auxiliary_loss_mlp": 0.01033179, "balance_loss_clip": 1.04959774, "balance_loss_mlp": 1.023803, "epoch": 0.17327000541093007, "flos": 23368190772960.0, "grad_norm": 1.7451138749696593, "language_loss": 0.78268331, "learning_rate": 3.7886486101949463e-06, "loss": 0.80446911, "num_input_tokens_seen": 30460430, "step": 1441, "time_per_iteration": 2.692586898803711 }, { "auxiliary_loss_clip": 0.01150659, "auxiliary_loss_mlp": 0.01045607, "balance_loss_clip": 1.04981852, "balance_loss_mlp": 1.03639841, "epoch": 0.17339024830156918, "flos": 18221293234080.0, "grad_norm": 1.9104580074060808, "language_loss": 0.87708044, "learning_rate": 3.7882999475799594e-06, "loss": 0.89904314, "num_input_tokens_seen": 30478465, "step": 1442, "time_per_iteration": 2.637375593185425 }, { "auxiliary_loss_clip": 0.01146417, "auxiliary_loss_mlp": 0.01039035, "balance_loss_clip": 1.05407763, "balance_loss_mlp": 1.02943301, "epoch": 0.17351049119220827, "flos": 23332028782560.0, "grad_norm": 1.7993968161404779, "language_loss": 0.81399786, "learning_rate": 3.787951013681293e-06, "loss": 0.83585238, "num_input_tokens_seen": 30496510, "step": 1443, "time_per_iteration": 2.679546594619751 }, { "auxiliary_loss_clip": 0.01208389, "auxiliary_loss_mlp": 0.01034803, "balance_loss_clip": 1.05934703, "balance_loss_mlp": 1.02453947, "epoch": 0.17363073408284735, "flos": 23803499502240.0, "grad_norm": 2.0066805138278325, "language_loss": 0.77686304, "learning_rate": 3.787601808551879e-06, "loss": 0.79929495, "num_input_tokens_seen": 30516325, "step": 1444, "time_per_iteration": 2.5572516918182373 }, { "auxiliary_loss_clip": 0.01179402, "auxiliary_loss_mlp": 0.01035236, "balance_loss_clip": 1.05695689, "balance_loss_mlp": 1.02532971, "epoch": 0.17375097697348643, "flos": 18515294279040.0, "grad_norm": 2.1423708543531577, "language_loss": 0.8414219, "learning_rate": 3.7872523322446926e-06, "loss": 0.86356831, "num_input_tokens_seen": 30535210, "step": 1445, "time_per_iteration": 2.561971426010132 }, { "auxiliary_loss_clip": 0.01170655, "auxiliary_loss_mlp": 0.01034706, "balance_loss_clip": 1.0524323, "balance_loss_mlp": 1.02582526, "epoch": 0.17387121986412554, "flos": 38877903010080.0, "grad_norm": 1.7971810616679469, "language_loss": 0.60091782, "learning_rate": 3.7869025848127478e-06, "loss": 0.62297142, "num_input_tokens_seen": 30559405, "step": 1446, "time_per_iteration": 2.790411949157715 }, { "auxiliary_loss_clip": 0.01212632, "auxiliary_loss_mlp": 0.01032233, "balance_loss_clip": 1.06160212, "balance_loss_mlp": 1.02316141, "epoch": 0.17399146275476463, "flos": 20375718421440.0, "grad_norm": 2.682753217744724, "language_loss": 0.80039018, "learning_rate": 3.786552566309102e-06, "loss": 0.82283878, "num_input_tokens_seen": 30577615, "step": 1447, "time_per_iteration": 4.0085227489471436 }, { "auxiliary_loss_clip": 0.0119318, "auxiliary_loss_mlp": 0.00764532, "balance_loss_clip": 1.06270385, "balance_loss_mlp": 1.00018358, "epoch": 0.1741117056454037, "flos": 19164342424320.0, "grad_norm": 2.057001573970727, "language_loss": 0.86341596, "learning_rate": 3.7862022767868517e-06, "loss": 0.88299298, "num_input_tokens_seen": 30595205, "step": 1448, "time_per_iteration": 2.595681667327881 }, { "auxiliary_loss_clip": 0.0117791, "auxiliary_loss_mlp": 0.01034274, "balance_loss_clip": 1.06234944, "balance_loss_mlp": 1.02561975, "epoch": 0.17423194853604282, "flos": 25374304468800.0, "grad_norm": 1.8890074221059563, "language_loss": 0.84534979, "learning_rate": 3.7858517162991367e-06, "loss": 0.86747158, "num_input_tokens_seen": 30615280, "step": 1449, "time_per_iteration": 2.6367363929748535 }, { "auxiliary_loss_clip": 0.01179347, "auxiliary_loss_mlp": 0.01035269, "balance_loss_clip": 1.05458128, "balance_loss_mlp": 1.0259707, "epoch": 0.1743521914266819, "flos": 25191878253600.0, "grad_norm": 2.55150854755962, "language_loss": 0.60744756, "learning_rate": 3.7855008848991363e-06, "loss": 0.62959373, "num_input_tokens_seen": 30633485, "step": 1450, "time_per_iteration": 2.676560878753662 }, { "auxiliary_loss_clip": 0.01190266, "auxiliary_loss_mlp": 0.01032634, "balance_loss_clip": 1.05940163, "balance_loss_mlp": 1.02372861, "epoch": 0.17447243431732098, "flos": 25666581499680.0, "grad_norm": 1.8464242503509543, "language_loss": 0.77553999, "learning_rate": 3.7851497826400714e-06, "loss": 0.79776895, "num_input_tokens_seen": 30653625, "step": 1451, "time_per_iteration": 3.4768550395965576 }, { "auxiliary_loss_clip": 0.01232323, "auxiliary_loss_mlp": 0.0103548, "balance_loss_clip": 1.06799126, "balance_loss_mlp": 1.0259074, "epoch": 0.17459267720796007, "flos": 36281955459360.0, "grad_norm": 1.8083596702494387, "language_loss": 0.7583428, "learning_rate": 3.7847984095752034e-06, "loss": 0.78102082, "num_input_tokens_seen": 30677080, "step": 1452, "time_per_iteration": 2.675966739654541 }, { "auxiliary_loss_clip": 0.01224055, "auxiliary_loss_mlp": 0.0103063, "balance_loss_clip": 1.06264043, "balance_loss_mlp": 1.0216893, "epoch": 0.17471292009859918, "flos": 20011117409760.0, "grad_norm": 2.1451710874307026, "language_loss": 0.80346024, "learning_rate": 3.784446765757836e-06, "loss": 0.82600713, "num_input_tokens_seen": 30695725, "step": 1453, "time_per_iteration": 3.4664316177368164 }, { "auxiliary_loss_clip": 0.01162228, "auxiliary_loss_mlp": 0.01037277, "balance_loss_clip": 1.05218363, "balance_loss_mlp": 1.02769887, "epoch": 0.17483316298923826, "flos": 27819246756000.0, "grad_norm": 2.767343559482085, "language_loss": 0.77880728, "learning_rate": 3.7840948512413133e-06, "loss": 0.80080235, "num_input_tokens_seen": 30713310, "step": 1454, "time_per_iteration": 2.7142646312713623 }, { "auxiliary_loss_clip": 0.01175608, "auxiliary_loss_mlp": 0.01036264, "balance_loss_clip": 1.05897045, "balance_loss_mlp": 1.02710867, "epoch": 0.17495340587987734, "flos": 44017940409600.0, "grad_norm": 2.218084170443049, "language_loss": 0.78799796, "learning_rate": 3.7837426660790196e-06, "loss": 0.81011665, "num_input_tokens_seen": 30734725, "step": 1455, "time_per_iteration": 2.815418004989624 }, { "auxiliary_loss_clip": 0.01226037, "auxiliary_loss_mlp": 0.01043862, "balance_loss_clip": 1.06483889, "balance_loss_mlp": 1.03496909, "epoch": 0.17507364877051645, "flos": 20885829401760.0, "grad_norm": 1.994250263840304, "language_loss": 0.82258677, "learning_rate": 3.783390210324382e-06, "loss": 0.84528577, "num_input_tokens_seen": 30754450, "step": 1456, "time_per_iteration": 2.5432167053222656 }, { "auxiliary_loss_clip": 0.01180452, "auxiliary_loss_mlp": 0.01033928, "balance_loss_clip": 1.06119609, "balance_loss_mlp": 1.02505934, "epoch": 0.17519389166115554, "flos": 24717606011040.0, "grad_norm": 4.575203266181416, "language_loss": 0.72768247, "learning_rate": 3.7830374840308676e-06, "loss": 0.74982631, "num_input_tokens_seen": 30774605, "step": 1457, "time_per_iteration": 2.662407159805298 }, { "auxiliary_loss_clip": 0.01211754, "auxiliary_loss_mlp": 0.01040762, "balance_loss_clip": 1.06442904, "balance_loss_mlp": 1.03134441, "epoch": 0.17531413455179462, "flos": 23798147875200.0, "grad_norm": 2.2492539539062335, "language_loss": 0.82401037, "learning_rate": 3.7826844872519842e-06, "loss": 0.8465355, "num_input_tokens_seen": 30792460, "step": 1458, "time_per_iteration": 2.5797441005706787 }, { "auxiliary_loss_clip": 0.011921, "auxiliary_loss_mlp": 0.01037079, "balance_loss_clip": 1.06253886, "balance_loss_mlp": 1.02869868, "epoch": 0.1754343774424337, "flos": 24572383378080.0, "grad_norm": 2.294631665662738, "language_loss": 0.72681534, "learning_rate": 3.782331220041282e-06, "loss": 0.74910718, "num_input_tokens_seen": 30812525, "step": 1459, "time_per_iteration": 2.642641067504883 }, { "auxiliary_loss_clip": 0.01191475, "auxiliary_loss_mlp": 0.01033945, "balance_loss_clip": 1.06016874, "balance_loss_mlp": 1.02502823, "epoch": 0.17555462033307281, "flos": 18114603110880.0, "grad_norm": 2.0944706574104903, "language_loss": 0.82867813, "learning_rate": 3.7819776824523504e-06, "loss": 0.85093236, "num_input_tokens_seen": 30830390, "step": 1460, "time_per_iteration": 2.6242690086364746 }, { "auxiliary_loss_clip": 0.01204808, "auxiliary_loss_mlp": 0.01043411, "balance_loss_clip": 1.06121683, "balance_loss_mlp": 1.03315949, "epoch": 0.1756748632237119, "flos": 28366022564160.0, "grad_norm": 1.8285708522629425, "language_loss": 0.83931786, "learning_rate": 3.7816238745388213e-06, "loss": 0.86180007, "num_input_tokens_seen": 30849935, "step": 1461, "time_per_iteration": 2.6427977085113525 }, { "auxiliary_loss_clip": 0.01198969, "auxiliary_loss_mlp": 0.01035933, "balance_loss_clip": 1.05815208, "balance_loss_mlp": 1.02675986, "epoch": 0.17579510611435098, "flos": 25732943265120.0, "grad_norm": 2.0673099433841253, "language_loss": 0.86939299, "learning_rate": 3.781269796354367e-06, "loss": 0.89174205, "num_input_tokens_seen": 30869555, "step": 1462, "time_per_iteration": 2.6293461322784424 }, { "auxiliary_loss_clip": 0.01198908, "auxiliary_loss_mlp": 0.01038457, "balance_loss_clip": 1.06255925, "balance_loss_mlp": 1.02963519, "epoch": 0.1759153490049901, "flos": 18588085180320.0, "grad_norm": 1.8307500481897507, "language_loss": 0.86247456, "learning_rate": 3.7809154479527006e-06, "loss": 0.88484824, "num_input_tokens_seen": 30888760, "step": 1463, "time_per_iteration": 2.581185817718506 }, { "auxiliary_loss_clip": 0.01171669, "auxiliary_loss_mlp": 0.01036728, "balance_loss_clip": 1.05785429, "balance_loss_mlp": 1.02758455, "epoch": 0.17603559189562917, "flos": 18619326547200.0, "grad_norm": 2.617791189161395, "language_loss": 0.84491861, "learning_rate": 3.780560829387577e-06, "loss": 0.86700255, "num_input_tokens_seen": 30907260, "step": 1464, "time_per_iteration": 2.6294476985931396 }, { "auxiliary_loss_clip": 0.01110943, "auxiliary_loss_mlp": 0.01007342, "balance_loss_clip": 1.02307773, "balance_loss_mlp": 1.00548232, "epoch": 0.17615583478626826, "flos": 60530784522720.0, "grad_norm": 0.8607584309865203, "language_loss": 0.57957458, "learning_rate": 3.7802059407127915e-06, "loss": 0.60075742, "num_input_tokens_seen": 30965810, "step": 1465, "time_per_iteration": 3.117621660232544 }, { "auxiliary_loss_clip": 0.01189227, "auxiliary_loss_mlp": 0.010408, "balance_loss_clip": 1.05782628, "balance_loss_mlp": 1.03209758, "epoch": 0.17627607767690734, "flos": 23616224497440.0, "grad_norm": 2.1088810614426188, "language_loss": 0.86356771, "learning_rate": 3.7798507819821797e-06, "loss": 0.88586795, "num_input_tokens_seen": 30982935, "step": 1466, "time_per_iteration": 2.6440913677215576 }, { "auxiliary_loss_clip": 0.01169286, "auxiliary_loss_mlp": 0.01037518, "balance_loss_clip": 1.05739188, "balance_loss_mlp": 1.02822566, "epoch": 0.17639632056754645, "flos": 17639073774720.0, "grad_norm": 2.500331500562406, "language_loss": 0.79067248, "learning_rate": 3.7794953532496197e-06, "loss": 0.8127405, "num_input_tokens_seen": 30998840, "step": 1467, "time_per_iteration": 2.5926806926727295 }, { "auxiliary_loss_clip": 0.01060117, "auxiliary_loss_mlp": 0.00755306, "balance_loss_clip": 1.02360868, "balance_loss_mlp": 0.99998516, "epoch": 0.17651656345818553, "flos": 57932616508800.0, "grad_norm": 0.8595382258375677, "language_loss": 0.5801419, "learning_rate": 3.7791396545690295e-06, "loss": 0.59829617, "num_input_tokens_seen": 31060075, "step": 1468, "time_per_iteration": 3.3159966468811035 }, { "auxiliary_loss_clip": 0.01211946, "auxiliary_loss_mlp": 0.01032813, "balance_loss_clip": 1.06424928, "balance_loss_mlp": 1.0234437, "epoch": 0.17663680634882462, "flos": 22929505849440.0, "grad_norm": 2.071262749911418, "language_loss": 0.80332565, "learning_rate": 3.7787836859943685e-06, "loss": 0.82577324, "num_input_tokens_seen": 31078800, "step": 1469, "time_per_iteration": 2.8039610385894775 }, { "auxiliary_loss_clip": 0.01212881, "auxiliary_loss_mlp": 0.01047272, "balance_loss_clip": 1.06269336, "balance_loss_mlp": 1.03778934, "epoch": 0.17675704923946373, "flos": 22637983074720.0, "grad_norm": 2.249344531504773, "language_loss": 0.78642845, "learning_rate": 3.7784274475796363e-06, "loss": 0.80902994, "num_input_tokens_seen": 31097430, "step": 1470, "time_per_iteration": 2.6169958114624023 }, { "auxiliary_loss_clip": 0.01178768, "auxiliary_loss_mlp": 0.01035897, "balance_loss_clip": 1.05621207, "balance_loss_mlp": 1.02631879, "epoch": 0.1768772921301028, "flos": 27126529975680.0, "grad_norm": 1.9919145228139084, "language_loss": 0.76085716, "learning_rate": 3.7780709393788745e-06, "loss": 0.78300381, "num_input_tokens_seen": 31117905, "step": 1471, "time_per_iteration": 2.6914453506469727 }, { "auxiliary_loss_clip": 0.0122294, "auxiliary_loss_mlp": 0.01031023, "balance_loss_clip": 1.06213486, "balance_loss_mlp": 1.02211189, "epoch": 0.1769975350207419, "flos": 19172136404640.0, "grad_norm": 2.3047985185983957, "language_loss": 0.75490975, "learning_rate": 3.777714161446165e-06, "loss": 0.77744937, "num_input_tokens_seen": 31137610, "step": 1472, "time_per_iteration": 2.5241734981536865 }, { "auxiliary_loss_clip": 0.01210566, "auxiliary_loss_mlp": 0.01029981, "balance_loss_clip": 1.06271648, "balance_loss_mlp": 1.02110028, "epoch": 0.177117777911381, "flos": 36134936978400.0, "grad_norm": 2.228783941371118, "language_loss": 0.69453335, "learning_rate": 3.7773571138356304e-06, "loss": 0.71693885, "num_input_tokens_seen": 31157780, "step": 1473, "time_per_iteration": 4.227472543716431 }, { "auxiliary_loss_clip": 0.01153356, "auxiliary_loss_mlp": 0.01028946, "balance_loss_clip": 1.05626845, "balance_loss_mlp": 1.02080989, "epoch": 0.17723802080202009, "flos": 22090596678240.0, "grad_norm": 2.280173165046279, "language_loss": 0.89325601, "learning_rate": 3.776999796601435e-06, "loss": 0.915079, "num_input_tokens_seen": 31176540, "step": 1474, "time_per_iteration": 2.6343045234680176 }, { "auxiliary_loss_clip": 0.01215881, "auxiliary_loss_mlp": 0.01038795, "balance_loss_clip": 1.06363368, "balance_loss_mlp": 1.02976501, "epoch": 0.17735826369265917, "flos": 30222675425760.0, "grad_norm": 1.7131132902186772, "language_loss": 0.72934878, "learning_rate": 3.776642209797783e-06, "loss": 0.75189555, "num_input_tokens_seen": 31198370, "step": 1475, "time_per_iteration": 2.651430606842041 }, { "auxiliary_loss_clip": 0.0120363, "auxiliary_loss_mlp": 0.01035793, "balance_loss_clip": 1.05968988, "balance_loss_mlp": 1.02692366, "epoch": 0.17747850658329825, "flos": 21397592562240.0, "grad_norm": 1.781813474853257, "language_loss": 0.78175724, "learning_rate": 3.7762843534789205e-06, "loss": 0.80415154, "num_input_tokens_seen": 31217120, "step": 1476, "time_per_iteration": 2.615093469619751 }, { "auxiliary_loss_clip": 0.01199643, "auxiliary_loss_mlp": 0.01029892, "balance_loss_clip": 1.05845106, "balance_loss_mlp": 1.02052236, "epoch": 0.17759874947393736, "flos": 16983345074880.0, "grad_norm": 2.019602787506744, "language_loss": 0.88284433, "learning_rate": 3.7759262276991343e-06, "loss": 0.90513968, "num_input_tokens_seen": 31234730, "step": 1477, "time_per_iteration": 3.4539339542388916 }, { "auxiliary_loss_clip": 0.01201463, "auxiliary_loss_mlp": 0.01036599, "balance_loss_clip": 1.06160951, "balance_loss_mlp": 1.0273366, "epoch": 0.17771899236457644, "flos": 11546110020960.0, "grad_norm": 2.2031056011283683, "language_loss": 0.80668068, "learning_rate": 3.7755678325127506e-06, "loss": 0.82906133, "num_input_tokens_seen": 31252410, "step": 1478, "time_per_iteration": 2.595885753631592 }, { "auxiliary_loss_clip": 0.01155227, "auxiliary_loss_mlp": 0.0103258, "balance_loss_clip": 1.05669987, "balance_loss_mlp": 1.02362776, "epoch": 0.17783923525521553, "flos": 18807750894720.0, "grad_norm": 1.6576155602512983, "language_loss": 0.75671864, "learning_rate": 3.7752091679741393e-06, "loss": 0.77859664, "num_input_tokens_seen": 31270200, "step": 1479, "time_per_iteration": 3.5147180557250977 }, { "auxiliary_loss_clip": 0.01207299, "auxiliary_loss_mlp": 0.0103225, "balance_loss_clip": 1.06103778, "balance_loss_mlp": 1.02231956, "epoch": 0.17795947814585464, "flos": 30408370084320.0, "grad_norm": 2.441379792565668, "language_loss": 0.77629089, "learning_rate": 3.774850234137708e-06, "loss": 0.79868633, "num_input_tokens_seen": 31287495, "step": 1480, "time_per_iteration": 2.596088171005249 }, { "auxiliary_loss_clip": 0.01209015, "auxiliary_loss_mlp": 0.01032655, "balance_loss_clip": 1.05940318, "balance_loss_mlp": 1.02311206, "epoch": 0.17807972103649372, "flos": 24389059238880.0, "grad_norm": 2.1944968118965424, "language_loss": 0.82409358, "learning_rate": 3.7744910310579076e-06, "loss": 0.84651029, "num_input_tokens_seen": 31306420, "step": 1481, "time_per_iteration": 2.5837676525115967 }, { "auxiliary_loss_clip": 0.01221095, "auxiliary_loss_mlp": 0.01033653, "balance_loss_clip": 1.06123722, "balance_loss_mlp": 1.02440238, "epoch": 0.1781999639271328, "flos": 20301562675680.0, "grad_norm": 1.8442104869212856, "language_loss": 0.85079229, "learning_rate": 3.774131558789229e-06, "loss": 0.87333977, "num_input_tokens_seen": 31325750, "step": 1482, "time_per_iteration": 2.533200740814209 }, { "auxiliary_loss_clip": 0.01227105, "auxiliary_loss_mlp": 0.00763913, "balance_loss_clip": 1.06516254, "balance_loss_mlp": 1.00045037, "epoch": 0.1783202068177719, "flos": 15924482853600.0, "grad_norm": 2.150710972958005, "language_loss": 0.69896102, "learning_rate": 3.773771817386203e-06, "loss": 0.71887124, "num_input_tokens_seen": 31343080, "step": 1483, "time_per_iteration": 2.515456199645996 }, { "auxiliary_loss_clip": 0.01193566, "auxiliary_loss_mlp": 0.01033259, "balance_loss_clip": 1.06005907, "balance_loss_mlp": 1.02459884, "epoch": 0.178440449708411, "flos": 20631761628000.0, "grad_norm": 1.8548250298801086, "language_loss": 0.7948842, "learning_rate": 3.773411806903403e-06, "loss": 0.81715244, "num_input_tokens_seen": 31362160, "step": 1484, "time_per_iteration": 2.603437900543213 }, { "auxiliary_loss_clip": 0.0115345, "auxiliary_loss_mlp": 0.01036666, "balance_loss_clip": 1.05401134, "balance_loss_mlp": 1.02759445, "epoch": 0.17856069259905008, "flos": 21686062395360.0, "grad_norm": 1.9364072612674024, "language_loss": 0.94509554, "learning_rate": 3.7730515273954415e-06, "loss": 0.96699673, "num_input_tokens_seen": 31380770, "step": 1485, "time_per_iteration": 2.6794826984405518 }, { "auxiliary_loss_clip": 0.0122359, "auxiliary_loss_mlp": 0.01032285, "balance_loss_clip": 1.06378829, "balance_loss_mlp": 1.02313602, "epoch": 0.17868093548968916, "flos": 26572965862080.0, "grad_norm": 1.7462995536830626, "language_loss": 0.85304487, "learning_rate": 3.772690978916973e-06, "loss": 0.87560368, "num_input_tokens_seen": 31400525, "step": 1486, "time_per_iteration": 2.5780670642852783 }, { "auxiliary_loss_clip": 0.01209757, "auxiliary_loss_mlp": 0.01032554, "balance_loss_clip": 1.06374884, "balance_loss_mlp": 1.02364278, "epoch": 0.17880117838032827, "flos": 18581009539200.0, "grad_norm": 2.757836412396436, "language_loss": 0.86984396, "learning_rate": 3.772330161522693e-06, "loss": 0.89226711, "num_input_tokens_seen": 31418435, "step": 1487, "time_per_iteration": 2.552363872528076 }, { "auxiliary_loss_clip": 0.01189314, "auxiliary_loss_mlp": 0.01027237, "balance_loss_clip": 1.06180751, "balance_loss_mlp": 1.01752162, "epoch": 0.17892142127096736, "flos": 26541221657760.0, "grad_norm": 2.004827559239685, "language_loss": 0.79699898, "learning_rate": 3.7719690752673365e-06, "loss": 0.81916451, "num_input_tokens_seen": 31439230, "step": 1488, "time_per_iteration": 2.643162727355957 }, { "auxiliary_loss_clip": 0.01179467, "auxiliary_loss_mlp": 0.01033121, "balance_loss_clip": 1.05760837, "balance_loss_mlp": 1.02410245, "epoch": 0.17904166416160644, "flos": 23872626873600.0, "grad_norm": 2.0646974876053394, "language_loss": 0.78334618, "learning_rate": 3.7716077202056796e-06, "loss": 0.80547202, "num_input_tokens_seen": 31457705, "step": 1489, "time_per_iteration": 2.6605255603790283 }, { "auxiliary_loss_clip": 0.01180825, "auxiliary_loss_mlp": 0.01033708, "balance_loss_clip": 1.05544949, "balance_loss_mlp": 1.02497029, "epoch": 0.17916190705224552, "flos": 19134429984960.0, "grad_norm": 2.186155066624866, "language_loss": 0.93796456, "learning_rate": 3.7712460963925404e-06, "loss": 0.96010983, "num_input_tokens_seen": 31473645, "step": 1490, "time_per_iteration": 2.5601799488067627 }, { "auxiliary_loss_clip": 0.01183777, "auxiliary_loss_mlp": 0.01034426, "balance_loss_clip": 1.0541749, "balance_loss_mlp": 1.02557433, "epoch": 0.17928214994288463, "flos": 25152124567200.0, "grad_norm": 1.74411675952628, "language_loss": 0.75833374, "learning_rate": 3.7708842038827775e-06, "loss": 0.78051579, "num_input_tokens_seen": 31492605, "step": 1491, "time_per_iteration": 2.6403133869171143 }, { "auxiliary_loss_clip": 0.01206404, "auxiliary_loss_mlp": 0.01028521, "balance_loss_clip": 1.05891335, "balance_loss_mlp": 1.01949072, "epoch": 0.17940239283352372, "flos": 22384633640160.0, "grad_norm": 1.6703157494169452, "language_loss": 0.85728854, "learning_rate": 3.770522042731288e-06, "loss": 0.87963784, "num_input_tokens_seen": 31514500, "step": 1492, "time_per_iteration": 2.5940115451812744 }, { "auxiliary_loss_clip": 0.01154304, "auxiliary_loss_mlp": 0.01033619, "balance_loss_clip": 1.05270648, "balance_loss_mlp": 1.02418399, "epoch": 0.1795226357241628, "flos": 23178688916640.0, "grad_norm": 1.7835790406538732, "language_loss": 0.87865925, "learning_rate": 3.7701596129930122e-06, "loss": 0.90053844, "num_input_tokens_seen": 31533225, "step": 1493, "time_per_iteration": 2.6819093227386475 }, { "auxiliary_loss_clip": 0.01184606, "auxiliary_loss_mlp": 0.01032485, "balance_loss_clip": 1.05744016, "balance_loss_mlp": 1.02275193, "epoch": 0.1796428786148019, "flos": 22090417093440.0, "grad_norm": 1.941437295265416, "language_loss": 0.73015475, "learning_rate": 3.7697969147229315e-06, "loss": 0.75232565, "num_input_tokens_seen": 31551385, "step": 1494, "time_per_iteration": 2.6180007457733154 }, { "auxiliary_loss_clip": 0.01205956, "auxiliary_loss_mlp": 0.010327, "balance_loss_clip": 1.05987835, "balance_loss_mlp": 1.02415311, "epoch": 0.179763121505441, "flos": 21324622076160.0, "grad_norm": 1.911409944040034, "language_loss": 0.85572237, "learning_rate": 3.7694339479760647e-06, "loss": 0.87810892, "num_input_tokens_seen": 31570415, "step": 1495, "time_per_iteration": 2.5756149291992188 }, { "auxiliary_loss_clip": 0.01113288, "auxiliary_loss_mlp": 0.01000995, "balance_loss_clip": 1.03412819, "balance_loss_mlp": 0.99909937, "epoch": 0.17988336439608008, "flos": 68161875197760.0, "grad_norm": 1.2276577288918857, "language_loss": 0.57381207, "learning_rate": 3.769070712807476e-06, "loss": 0.59495491, "num_input_tokens_seen": 31632445, "step": 1496, "time_per_iteration": 3.2542407512664795 }, { "auxiliary_loss_clip": 0.01136716, "auxiliary_loss_mlp": 0.01036348, "balance_loss_clip": 1.05234718, "balance_loss_mlp": 1.02713895, "epoch": 0.18000360728671919, "flos": 21945050792640.0, "grad_norm": 2.4819266867406338, "language_loss": 0.78850919, "learning_rate": 3.768707209272266e-06, "loss": 0.81023979, "num_input_tokens_seen": 31652575, "step": 1497, "time_per_iteration": 2.740133762359619 }, { "auxiliary_loss_clip": 0.01185773, "auxiliary_loss_mlp": 0.01037374, "balance_loss_clip": 1.05607677, "balance_loss_mlp": 1.02758718, "epoch": 0.18012385017735827, "flos": 18986333995200.0, "grad_norm": 2.9095426257707278, "language_loss": 0.7652967, "learning_rate": 3.768343437425579e-06, "loss": 0.78752816, "num_input_tokens_seen": 31671145, "step": 1498, "time_per_iteration": 2.7026796340942383 }, { "auxiliary_loss_clip": 0.01121515, "auxiliary_loss_mlp": 0.01030157, "balance_loss_clip": 1.04892457, "balance_loss_mlp": 1.02090669, "epoch": 0.18024409306799735, "flos": 19748106312960.0, "grad_norm": 2.401343675365043, "language_loss": 0.85857022, "learning_rate": 3.7679793973225987e-06, "loss": 0.88008702, "num_input_tokens_seen": 31686955, "step": 1499, "time_per_iteration": 4.284273624420166 }, { "auxiliary_loss_clip": 0.01077173, "auxiliary_loss_mlp": 0.01002567, "balance_loss_clip": 1.02699316, "balance_loss_mlp": 1.00065994, "epoch": 0.18036433595863643, "flos": 67227194659200.0, "grad_norm": 0.857070890455722, "language_loss": 0.61652172, "learning_rate": 3.767615089018549e-06, "loss": 0.63731921, "num_input_tokens_seen": 31749300, "step": 1500, "time_per_iteration": 3.283478260040283 }, { "auxiliary_loss_clip": 0.011856, "auxiliary_loss_mlp": 0.01034966, "balance_loss_clip": 1.05560613, "balance_loss_mlp": 1.02486885, "epoch": 0.18048457884927555, "flos": 18181467713760.0, "grad_norm": 2.115630378818585, "language_loss": 0.86069971, "learning_rate": 3.7672505125686966e-06, "loss": 0.88290536, "num_input_tokens_seen": 31765665, "step": 1501, "time_per_iteration": 2.5730857849121094 }, { "auxiliary_loss_clip": 0.0116082, "auxiliary_loss_mlp": 0.01032567, "balance_loss_clip": 1.05318451, "balance_loss_mlp": 1.02354932, "epoch": 0.18060482173991463, "flos": 15813770030880.0, "grad_norm": 2.679687759219937, "language_loss": 0.84214097, "learning_rate": 3.7668856680283455e-06, "loss": 0.86407483, "num_input_tokens_seen": 31782690, "step": 1502, "time_per_iteration": 2.6481106281280518 }, { "auxiliary_loss_clip": 0.01193778, "auxiliary_loss_mlp": 0.01044839, "balance_loss_clip": 1.05711555, "balance_loss_mlp": 1.03520703, "epoch": 0.1807250646305537, "flos": 18587402758080.0, "grad_norm": 1.9612332911685646, "language_loss": 0.82483947, "learning_rate": 3.7665205554528437e-06, "loss": 0.84722567, "num_input_tokens_seen": 31802045, "step": 1503, "time_per_iteration": 3.5109434127807617 }, { "auxiliary_loss_clip": 0.01190427, "auxiliary_loss_mlp": 0.01036139, "balance_loss_clip": 1.05974269, "balance_loss_mlp": 1.02687657, "epoch": 0.18084530752119282, "flos": 23149135646880.0, "grad_norm": 1.8995883018859572, "language_loss": 0.74299014, "learning_rate": 3.7661551748975782e-06, "loss": 0.76525581, "num_input_tokens_seen": 31820220, "step": 1504, "time_per_iteration": 2.5964975357055664 }, { "auxiliary_loss_clip": 0.01103148, "auxiliary_loss_mlp": 0.01005485, "balance_loss_clip": 1.02562261, "balance_loss_mlp": 1.00351834, "epoch": 0.1809655504118319, "flos": 59803162845600.0, "grad_norm": 0.8184916184227027, "language_loss": 0.60555375, "learning_rate": 3.7657895264179772e-06, "loss": 0.62663996, "num_input_tokens_seen": 31876195, "step": 1505, "time_per_iteration": 4.055802345275879 }, { "auxiliary_loss_clip": 0.01186088, "auxiliary_loss_mlp": 0.0103967, "balance_loss_clip": 1.05486703, "balance_loss_mlp": 1.03110492, "epoch": 0.181085793302471, "flos": 44201946971040.0, "grad_norm": 1.9747224969412016, "language_loss": 0.74655414, "learning_rate": 3.765423610069509e-06, "loss": 0.76881164, "num_input_tokens_seen": 31901585, "step": 1506, "time_per_iteration": 2.7924816608428955 }, { "auxiliary_loss_clip": 0.01195733, "auxiliary_loss_mlp": 0.01033095, "balance_loss_clip": 1.0604372, "balance_loss_mlp": 1.0237546, "epoch": 0.18120603619311007, "flos": 34898389580640.0, "grad_norm": 1.7896059345966158, "language_loss": 0.72255605, "learning_rate": 3.765057425907683e-06, "loss": 0.74484432, "num_input_tokens_seen": 31923045, "step": 1507, "time_per_iteration": 2.710080146789551 }, { "auxiliary_loss_clip": 0.01209564, "auxiliary_loss_mlp": 0.01036416, "balance_loss_clip": 1.05834675, "balance_loss_mlp": 1.02664661, "epoch": 0.18132627908374918, "flos": 21506760955680.0, "grad_norm": 1.795765674317946, "language_loss": 0.78407586, "learning_rate": 3.764690973988048e-06, "loss": 0.8065356, "num_input_tokens_seen": 31943385, "step": 1508, "time_per_iteration": 2.5810492038726807 }, { "auxiliary_loss_clip": 0.01179328, "auxiliary_loss_mlp": 0.01037639, "balance_loss_clip": 1.05599868, "balance_loss_mlp": 1.02878809, "epoch": 0.18144652197438826, "flos": 29057697752640.0, "grad_norm": 2.839841364921246, "language_loss": 0.7391119, "learning_rate": 3.7643242543661967e-06, "loss": 0.76128161, "num_input_tokens_seen": 31966045, "step": 1509, "time_per_iteration": 2.7216711044311523 }, { "auxiliary_loss_clip": 0.01095711, "auxiliary_loss_mlp": 0.01007182, "balance_loss_clip": 1.02251661, "balance_loss_mlp": 1.00527465, "epoch": 0.18156676486502735, "flos": 68675075036640.0, "grad_norm": 0.8235212739154494, "language_loss": 0.60510445, "learning_rate": 3.7639572670977573e-06, "loss": 0.62613332, "num_input_tokens_seen": 32021540, "step": 1510, "time_per_iteration": 3.1032841205596924 }, { "auxiliary_loss_clip": 0.01181529, "auxiliary_loss_mlp": 0.01032886, "balance_loss_clip": 1.0564146, "balance_loss_mlp": 1.02371907, "epoch": 0.18168700775566646, "flos": 26471519615040.0, "grad_norm": 1.9383402032492043, "language_loss": 0.7659651, "learning_rate": 3.7635900122384042e-06, "loss": 0.78810918, "num_input_tokens_seen": 32044535, "step": 1511, "time_per_iteration": 2.672884941101074 }, { "auxiliary_loss_clip": 0.01193735, "auxiliary_loss_mlp": 0.01037733, "balance_loss_clip": 1.05608559, "balance_loss_mlp": 1.02826142, "epoch": 0.18180725064630554, "flos": 15005671223040.0, "grad_norm": 3.130002548208409, "language_loss": 0.87089086, "learning_rate": 3.7632224898438477e-06, "loss": 0.89320552, "num_input_tokens_seen": 32061010, "step": 1512, "time_per_iteration": 2.5707924365997314 }, { "auxiliary_loss_clip": 0.01179059, "auxiliary_loss_mlp": 0.0102667, "balance_loss_clip": 1.05523872, "balance_loss_mlp": 1.01782489, "epoch": 0.18192749353694462, "flos": 19682391052800.0, "grad_norm": 1.6271674210281206, "language_loss": 0.79055548, "learning_rate": 3.762854699969842e-06, "loss": 0.81261277, "num_input_tokens_seen": 32081520, "step": 1513, "time_per_iteration": 2.6333250999450684 }, { "auxiliary_loss_clip": 0.01206893, "auxiliary_loss_mlp": 0.01040885, "balance_loss_clip": 1.0608598, "balance_loss_mlp": 1.03177726, "epoch": 0.1820477364275837, "flos": 20702720764320.0, "grad_norm": 1.9023822339246121, "language_loss": 0.73298621, "learning_rate": 3.762486642672179e-06, "loss": 0.75546402, "num_input_tokens_seen": 32098460, "step": 1514, "time_per_iteration": 2.5562186241149902 }, { "auxiliary_loss_clip": 0.01195311, "auxiliary_loss_mlp": 0.01036783, "balance_loss_clip": 1.05889809, "balance_loss_mlp": 1.02748525, "epoch": 0.18216797931822282, "flos": 17128711375680.0, "grad_norm": 1.8281780012138982, "language_loss": 0.8654598, "learning_rate": 3.7621183180066946e-06, "loss": 0.88778067, "num_input_tokens_seen": 32116420, "step": 1515, "time_per_iteration": 2.5672826766967773 }, { "auxiliary_loss_clip": 0.01190929, "auxiliary_loss_mlp": 0.01032287, "balance_loss_clip": 1.05509591, "balance_loss_mlp": 1.02328038, "epoch": 0.1822882222088619, "flos": 29242566321120.0, "grad_norm": 2.0055217796399694, "language_loss": 0.73602009, "learning_rate": 3.7617497260292625e-06, "loss": 0.75825226, "num_input_tokens_seen": 32138475, "step": 1516, "time_per_iteration": 2.6759133338928223 }, { "auxiliary_loss_clip": 0.01187753, "auxiliary_loss_mlp": 0.01036293, "balance_loss_clip": 1.05844223, "balance_loss_mlp": 1.02642202, "epoch": 0.18240846509950098, "flos": 17702741768160.0, "grad_norm": 3.7428174419149456, "language_loss": 0.79405439, "learning_rate": 3.7613808667957967e-06, "loss": 0.81629479, "num_input_tokens_seen": 32151165, "step": 1517, "time_per_iteration": 2.547095775604248 }, { "auxiliary_loss_clip": 0.01199845, "auxiliary_loss_mlp": 0.01040225, "balance_loss_clip": 1.06034863, "balance_loss_mlp": 1.030581, "epoch": 0.1825287079901401, "flos": 14790028208160.0, "grad_norm": 1.8674469673360283, "language_loss": 0.91235626, "learning_rate": 3.7610117403622547e-06, "loss": 0.93475688, "num_input_tokens_seen": 32167725, "step": 1518, "time_per_iteration": 2.571753978729248 }, { "auxiliary_loss_clip": 0.01171597, "auxiliary_loss_mlp": 0.01035625, "balance_loss_clip": 1.05462193, "balance_loss_mlp": 1.02686274, "epoch": 0.18264895088077918, "flos": 21946236052320.0, "grad_norm": 1.5631967947883378, "language_loss": 0.90326238, "learning_rate": 3.7606423467846313e-06, "loss": 0.92533451, "num_input_tokens_seen": 32187330, "step": 1519, "time_per_iteration": 2.6565003395080566 }, { "auxiliary_loss_clip": 0.0118836, "auxiliary_loss_mlp": 0.01049557, "balance_loss_clip": 1.06251466, "balance_loss_mlp": 1.04033029, "epoch": 0.18276919377141826, "flos": 20886763242720.0, "grad_norm": 1.5201366256410793, "language_loss": 0.7946856, "learning_rate": 3.760272686118964e-06, "loss": 0.81706476, "num_input_tokens_seen": 32205550, "step": 1520, "time_per_iteration": 2.6201043128967285 }, { "auxiliary_loss_clip": 0.0119965, "auxiliary_loss_mlp": 0.0103555, "balance_loss_clip": 1.05990458, "balance_loss_mlp": 1.02663958, "epoch": 0.18288943666205737, "flos": 21469880626080.0, "grad_norm": 1.9926704551556698, "language_loss": 0.92438519, "learning_rate": 3.7599027584213297e-06, "loss": 0.94673711, "num_input_tokens_seen": 32224430, "step": 1521, "time_per_iteration": 2.636569023132324 }, { "auxiliary_loss_clip": 0.01215524, "auxiliary_loss_mlp": 0.01039021, "balance_loss_clip": 1.06170797, "balance_loss_mlp": 1.03036058, "epoch": 0.18300967955269645, "flos": 21539367167040.0, "grad_norm": 2.0126437060297535, "language_loss": 0.78472114, "learning_rate": 3.7595325637478465e-06, "loss": 0.80726659, "num_input_tokens_seen": 32242455, "step": 1522, "time_per_iteration": 2.5680058002471924 }, { "auxiliary_loss_clip": 0.01183551, "auxiliary_loss_mlp": 0.010423, "balance_loss_clip": 1.05872226, "balance_loss_mlp": 1.03314495, "epoch": 0.18312992244333554, "flos": 28876241295360.0, "grad_norm": 1.6238809354684791, "language_loss": 0.81434309, "learning_rate": 3.7591621021546723e-06, "loss": 0.83660161, "num_input_tokens_seen": 32264450, "step": 1523, "time_per_iteration": 2.6593751907348633 }, { "auxiliary_loss_clip": 0.01199849, "auxiliary_loss_mlp": 0.01034999, "balance_loss_clip": 1.05721593, "balance_loss_mlp": 1.02586126, "epoch": 0.18325016533397462, "flos": 20120106218400.0, "grad_norm": 1.7100308319260957, "language_loss": 0.81295156, "learning_rate": 3.7587913736980062e-06, "loss": 0.83530009, "num_input_tokens_seen": 32284090, "step": 1524, "time_per_iteration": 3.301814079284668 }, { "auxiliary_loss_clip": 0.01131902, "auxiliary_loss_mlp": 0.01040825, "balance_loss_clip": 1.04750407, "balance_loss_mlp": 1.03129363, "epoch": 0.18337040822461373, "flos": 23329191342720.0, "grad_norm": 1.7961050840556805, "language_loss": 0.84405553, "learning_rate": 3.7584203784340865e-06, "loss": 0.86578274, "num_input_tokens_seen": 32303260, "step": 1525, "time_per_iteration": 3.388646125793457 }, { "auxiliary_loss_clip": 0.01188021, "auxiliary_loss_mlp": 0.01036027, "balance_loss_clip": 1.05591178, "balance_loss_mlp": 1.02661514, "epoch": 0.1834906511152528, "flos": 25009559789280.0, "grad_norm": 1.9111501347463713, "language_loss": 0.85759449, "learning_rate": 3.7580491164191938e-06, "loss": 0.87983501, "num_input_tokens_seen": 32321570, "step": 1526, "time_per_iteration": 2.5984835624694824 }, { "auxiliary_loss_clip": 0.01110744, "auxiliary_loss_mlp": 0.01003224, "balance_loss_clip": 1.02201295, "balance_loss_mlp": 1.00135291, "epoch": 0.1836108940058919, "flos": 67251504052800.0, "grad_norm": 0.7664737137924853, "language_loss": 0.61312604, "learning_rate": 3.757677587709648e-06, "loss": 0.63426566, "num_input_tokens_seen": 32384835, "step": 1527, "time_per_iteration": 3.235013484954834 }, { "auxiliary_loss_clip": 0.01175028, "auxiliary_loss_mlp": 0.01036402, "balance_loss_clip": 1.05834877, "balance_loss_mlp": 1.02777755, "epoch": 0.183731136896531, "flos": 25738725895680.0, "grad_norm": 1.89403102167769, "language_loss": 0.75588918, "learning_rate": 3.7573057923618095e-06, "loss": 0.77800345, "num_input_tokens_seen": 32404930, "step": 1528, "time_per_iteration": 2.6290507316589355 }, { "auxiliary_loss_clip": 0.01160018, "auxiliary_loss_mlp": 0.01034834, "balance_loss_clip": 1.05154693, "balance_loss_mlp": 1.02489805, "epoch": 0.1838513797871701, "flos": 20449407246720.0, "grad_norm": 2.4894799604635702, "language_loss": 0.74300092, "learning_rate": 3.7569337304320793e-06, "loss": 0.76494944, "num_input_tokens_seen": 32424515, "step": 1529, "time_per_iteration": 3.5278375148773193 }, { "auxiliary_loss_clip": 0.01095573, "auxiliary_loss_mlp": 0.0100301, "balance_loss_clip": 1.020643, "balance_loss_mlp": 1.00119829, "epoch": 0.18397162267780917, "flos": 68565152387040.0, "grad_norm": 0.8399398990660073, "language_loss": 0.64509213, "learning_rate": 3.756561401976899e-06, "loss": 0.66607797, "num_input_tokens_seen": 32484220, "step": 1530, "time_per_iteration": 3.0575175285339355 }, { "auxiliary_loss_clip": 0.01227036, "auxiliary_loss_mlp": 0.01034615, "balance_loss_clip": 1.06476521, "balance_loss_mlp": 1.02467847, "epoch": 0.18409186556844825, "flos": 31941109461600.0, "grad_norm": 1.755298420385587, "language_loss": 0.82421744, "learning_rate": 3.7561888070527514e-06, "loss": 0.84683394, "num_input_tokens_seen": 32506260, "step": 1531, "time_per_iteration": 3.5365872383117676 }, { "auxiliary_loss_clip": 0.01164609, "auxiliary_loss_mlp": 0.00764686, "balance_loss_clip": 1.05491078, "balance_loss_mlp": 1.00033045, "epoch": 0.18421210845908736, "flos": 20120537221920.0, "grad_norm": 2.0312254681538167, "language_loss": 0.8020069, "learning_rate": 3.7558159457161577e-06, "loss": 0.82129985, "num_input_tokens_seen": 32524225, "step": 1532, "time_per_iteration": 2.597487688064575 }, { "auxiliary_loss_clip": 0.01195494, "auxiliary_loss_mlp": 0.00764068, "balance_loss_clip": 1.0596838, "balance_loss_mlp": 1.00040615, "epoch": 0.18433235134972645, "flos": 23110495386240.0, "grad_norm": 2.043623811015698, "language_loss": 0.78051186, "learning_rate": 3.755442818023681e-06, "loss": 0.80010748, "num_input_tokens_seen": 32543850, "step": 1533, "time_per_iteration": 2.6182971000671387 }, { "auxiliary_loss_clip": 0.0117572, "auxiliary_loss_mlp": 0.01036185, "balance_loss_clip": 1.05653703, "balance_loss_mlp": 1.0268271, "epoch": 0.18445259424036553, "flos": 18291354446400.0, "grad_norm": 2.083149260569801, "language_loss": 0.76240933, "learning_rate": 3.7550694240319246e-06, "loss": 0.78452837, "num_input_tokens_seen": 32561725, "step": 1534, "time_per_iteration": 2.61263370513916 }, { "auxiliary_loss_clip": 0.01212355, "auxiliary_loss_mlp": 0.01035864, "balance_loss_clip": 1.05864167, "balance_loss_mlp": 1.02672088, "epoch": 0.18457283713100464, "flos": 21324083321760.0, "grad_norm": 2.68156135114367, "language_loss": 0.76451683, "learning_rate": 3.7546957637975326e-06, "loss": 0.78699905, "num_input_tokens_seen": 32579135, "step": 1535, "time_per_iteration": 2.5713653564453125 }, { "auxiliary_loss_clip": 0.01141084, "auxiliary_loss_mlp": 0.01031755, "balance_loss_clip": 1.04753971, "balance_loss_mlp": 1.02299333, "epoch": 0.18469308002164372, "flos": 20375682504480.0, "grad_norm": 1.460481930260825, "language_loss": 0.74072206, "learning_rate": 3.7543218373771873e-06, "loss": 0.7624504, "num_input_tokens_seen": 32598460, "step": 1536, "time_per_iteration": 2.673383951187134 }, { "auxiliary_loss_clip": 0.0114063, "auxiliary_loss_mlp": 0.00764601, "balance_loss_clip": 1.05135655, "balance_loss_mlp": 1.00030041, "epoch": 0.1848133229122828, "flos": 26435896379040.0, "grad_norm": 1.4991652675683353, "language_loss": 0.78118372, "learning_rate": 3.753947644827615e-06, "loss": 0.80023599, "num_input_tokens_seen": 32621920, "step": 1537, "time_per_iteration": 2.769084930419922 }, { "auxiliary_loss_clip": 0.01096139, "auxiliary_loss_mlp": 0.01008174, "balance_loss_clip": 1.02006817, "balance_loss_mlp": 1.00638568, "epoch": 0.1849335658029219, "flos": 70547459526720.0, "grad_norm": 0.9435646905991305, "language_loss": 0.57203627, "learning_rate": 3.753573186205579e-06, "loss": 0.59307939, "num_input_tokens_seen": 32690040, "step": 1538, "time_per_iteration": 3.308476686477661 }, { "auxiliary_loss_clip": 0.01180392, "auxiliary_loss_mlp": 0.00764358, "balance_loss_clip": 1.05142736, "balance_loss_mlp": 1.00031364, "epoch": 0.185053808693561, "flos": 17384143993920.0, "grad_norm": 1.974168470634556, "language_loss": 0.78100395, "learning_rate": 3.753198461567885e-06, "loss": 0.80045152, "num_input_tokens_seen": 32707285, "step": 1539, "time_per_iteration": 2.5695886611938477 }, { "auxiliary_loss_clip": 0.01175238, "auxiliary_loss_mlp": 0.0103439, "balance_loss_clip": 1.06174624, "balance_loss_mlp": 1.0259676, "epoch": 0.18517405158420008, "flos": 28986163944960.0, "grad_norm": 1.7032237962182675, "language_loss": 0.91929889, "learning_rate": 3.7528234709713783e-06, "loss": 0.94139516, "num_input_tokens_seen": 32730030, "step": 1540, "time_per_iteration": 2.688646078109741 }, { "auxiliary_loss_clip": 0.01214794, "auxiliary_loss_mlp": 0.0104114, "balance_loss_clip": 1.06408358, "balance_loss_mlp": 1.03224158, "epoch": 0.18529429447483917, "flos": 26794966178880.0, "grad_norm": 2.021195069365742, "language_loss": 0.84117603, "learning_rate": 3.7524482144729447e-06, "loss": 0.86373544, "num_input_tokens_seen": 32749485, "step": 1541, "time_per_iteration": 2.6081433296203613 }, { "auxiliary_loss_clip": 0.01169107, "auxiliary_loss_mlp": 0.01041078, "balance_loss_clip": 1.05264938, "balance_loss_mlp": 1.03151178, "epoch": 0.18541453736547828, "flos": 13581597401760.0, "grad_norm": 2.4615304118000076, "language_loss": 0.83952683, "learning_rate": 3.7520726921295106e-06, "loss": 0.86162865, "num_input_tokens_seen": 32766205, "step": 1542, "time_per_iteration": 2.649620294570923 }, { "auxiliary_loss_clip": 0.01205494, "auxiliary_loss_mlp": 0.01030068, "balance_loss_clip": 1.05870223, "balance_loss_mlp": 1.02158093, "epoch": 0.18553478025611736, "flos": 24025427985120.0, "grad_norm": 1.7291651861202115, "language_loss": 0.72168422, "learning_rate": 3.751696903998042e-06, "loss": 0.74403977, "num_input_tokens_seen": 32784840, "step": 1543, "time_per_iteration": 2.5811514854431152 }, { "auxiliary_loss_clip": 0.01207834, "auxiliary_loss_mlp": 0.01035677, "balance_loss_clip": 1.06307316, "balance_loss_mlp": 1.02720094, "epoch": 0.18565502314675644, "flos": 25885169705280.0, "grad_norm": 1.7076312858279732, "language_loss": 0.70258701, "learning_rate": 3.7513208501355456e-06, "loss": 0.72502214, "num_input_tokens_seen": 32805945, "step": 1544, "time_per_iteration": 2.6198313236236572 }, { "auxiliary_loss_clip": 0.01190577, "auxiliary_loss_mlp": 0.01041286, "balance_loss_clip": 1.05761659, "balance_loss_mlp": 1.03238082, "epoch": 0.18577526603739553, "flos": 19610067072000.0, "grad_norm": 1.7638806049682372, "language_loss": 0.83788931, "learning_rate": 3.750944530599069e-06, "loss": 0.86020792, "num_input_tokens_seen": 32825515, "step": 1545, "time_per_iteration": 2.5862934589385986 }, { "auxiliary_loss_clip": 0.01217092, "auxiliary_loss_mlp": 0.01033083, "balance_loss_clip": 1.06345069, "balance_loss_mlp": 1.02357054, "epoch": 0.18589550892803464, "flos": 18474894087360.0, "grad_norm": 1.8500059552490586, "language_loss": 0.80769265, "learning_rate": 3.7505679454456992e-06, "loss": 0.83019441, "num_input_tokens_seen": 32842125, "step": 1546, "time_per_iteration": 2.55861759185791 }, { "auxiliary_loss_clip": 0.01126868, "auxiliary_loss_mlp": 0.01036343, "balance_loss_clip": 1.04965854, "balance_loss_mlp": 1.02672875, "epoch": 0.18601575181867372, "flos": 23549970482880.0, "grad_norm": 1.8072044095532578, "language_loss": 0.70245594, "learning_rate": 3.750191094732564e-06, "loss": 0.72408807, "num_input_tokens_seen": 32862990, "step": 1547, "time_per_iteration": 2.813626527786255 }, { "auxiliary_loss_clip": 0.01126475, "auxiliary_loss_mlp": 0.00764488, "balance_loss_clip": 1.04806066, "balance_loss_mlp": 1.00022924, "epoch": 0.1861359947093128, "flos": 26360196204000.0, "grad_norm": 1.7476924721842688, "language_loss": 0.75378966, "learning_rate": 3.7498139785168313e-06, "loss": 0.7726993, "num_input_tokens_seen": 32883595, "step": 1548, "time_per_iteration": 2.858126163482666 }, { "auxiliary_loss_clip": 0.01204952, "auxiliary_loss_mlp": 0.0103533, "balance_loss_clip": 1.06108057, "balance_loss_mlp": 1.02671707, "epoch": 0.1862562375999519, "flos": 23331202692480.0, "grad_norm": 1.735302413118509, "language_loss": 0.77183276, "learning_rate": 3.749436596855709e-06, "loss": 0.79423559, "num_input_tokens_seen": 32902895, "step": 1549, "time_per_iteration": 2.565725564956665 }, { "auxiliary_loss_clip": 0.01201279, "auxiliary_loss_mlp": 0.01035696, "balance_loss_clip": 1.05787325, "balance_loss_mlp": 1.02657616, "epoch": 0.186376480490591, "flos": 16648225499040.0, "grad_norm": 2.1020242285021054, "language_loss": 0.90398145, "learning_rate": 3.749058949806446e-06, "loss": 0.92635119, "num_input_tokens_seen": 32919620, "step": 1550, "time_per_iteration": 4.0127904415130615 }, { "auxiliary_loss_clip": 0.01206922, "auxiliary_loss_mlp": 0.01027926, "balance_loss_clip": 1.05888093, "balance_loss_mlp": 1.01982582, "epoch": 0.18649672338123008, "flos": 21468659449440.0, "grad_norm": 1.9480223093388547, "language_loss": 0.84360516, "learning_rate": 3.748681037426331e-06, "loss": 0.86595368, "num_input_tokens_seen": 32938830, "step": 1551, "time_per_iteration": 2.579227924346924 }, { "auxiliary_loss_clip": 0.0122612, "auxiliary_loss_mlp": 0.01036322, "balance_loss_clip": 1.06574488, "balance_loss_mlp": 1.02741718, "epoch": 0.1866169662718692, "flos": 12312731128320.0, "grad_norm": 2.055819645756873, "language_loss": 0.91935074, "learning_rate": 3.7483028597726936e-06, "loss": 0.94197518, "num_input_tokens_seen": 32955600, "step": 1552, "time_per_iteration": 2.5118720531463623 }, { "auxiliary_loss_clip": 0.01176703, "auxiliary_loss_mlp": 0.01035586, "balance_loss_clip": 1.0589602, "balance_loss_mlp": 1.02631199, "epoch": 0.18673720916250827, "flos": 23581283683680.0, "grad_norm": 1.7508797074669855, "language_loss": 0.62507188, "learning_rate": 3.7479244169029017e-06, "loss": 0.64719474, "num_input_tokens_seen": 32975390, "step": 1553, "time_per_iteration": 2.6723878383636475 }, { "auxiliary_loss_clip": 0.01206297, "auxiliary_loss_mlp": 0.01029935, "balance_loss_clip": 1.05531633, "balance_loss_mlp": 1.02114391, "epoch": 0.18685745205314735, "flos": 19718373458400.0, "grad_norm": 2.687357811745109, "language_loss": 0.7332741, "learning_rate": 3.7475457088743658e-06, "loss": 0.75563645, "num_input_tokens_seen": 32992640, "step": 1554, "time_per_iteration": 2.5463626384735107 }, { "auxiliary_loss_clip": 0.01183765, "auxiliary_loss_mlp": 0.01035575, "balance_loss_clip": 1.05851626, "balance_loss_mlp": 1.02582383, "epoch": 0.18697769494378644, "flos": 34204128371040.0, "grad_norm": 1.8263647034254638, "language_loss": 0.74687541, "learning_rate": 3.7471667357445348e-06, "loss": 0.76906884, "num_input_tokens_seen": 33012470, "step": 1555, "time_per_iteration": 3.5637855529785156 }, { "auxiliary_loss_clip": 0.01146057, "auxiliary_loss_mlp": 0.01033521, "balance_loss_clip": 1.05218911, "balance_loss_mlp": 1.02510464, "epoch": 0.18709793783442555, "flos": 34241331953280.0, "grad_norm": 3.736255402519099, "language_loss": 0.72101176, "learning_rate": 3.7467874975709e-06, "loss": 0.74280751, "num_input_tokens_seen": 33033275, "step": 1556, "time_per_iteration": 2.788029909133911 }, { "auxiliary_loss_clip": 0.01213017, "auxiliary_loss_mlp": 0.01031574, "balance_loss_clip": 1.06308842, "balance_loss_mlp": 1.0230509, "epoch": 0.18721818072506463, "flos": 40734555871680.0, "grad_norm": 2.9754660445124568, "language_loss": 0.78389019, "learning_rate": 3.7464079944109904e-06, "loss": 0.8063361, "num_input_tokens_seen": 33055135, "step": 1557, "time_per_iteration": 3.5018961429595947 }, { "auxiliary_loss_clip": 0.01178076, "auxiliary_loss_mlp": 0.01033316, "balance_loss_clip": 1.05420685, "balance_loss_mlp": 1.02457166, "epoch": 0.18733842361570371, "flos": 22157389447200.0, "grad_norm": 1.7741724079872578, "language_loss": 0.77738094, "learning_rate": 3.746028226322376e-06, "loss": 0.79949486, "num_input_tokens_seen": 33071015, "step": 1558, "time_per_iteration": 2.6300008296966553 }, { "auxiliary_loss_clip": 0.011879, "auxiliary_loss_mlp": 0.0102971, "balance_loss_clip": 1.05770171, "balance_loss_mlp": 1.02111483, "epoch": 0.18745866650634282, "flos": 18914943855360.0, "grad_norm": 1.6351787185738016, "language_loss": 0.75248462, "learning_rate": 3.745648193362669e-06, "loss": 0.77466077, "num_input_tokens_seen": 33090370, "step": 1559, "time_per_iteration": 2.6060361862182617 }, { "auxiliary_loss_clip": 0.01190981, "auxiliary_loss_mlp": 0.01027335, "balance_loss_clip": 1.05990636, "balance_loss_mlp": 1.01905012, "epoch": 0.1875789093969819, "flos": 19314629348640.0, "grad_norm": 1.883451609990899, "language_loss": 0.71839583, "learning_rate": 3.745267895589518e-06, "loss": 0.74057901, "num_input_tokens_seen": 33108910, "step": 1560, "time_per_iteration": 2.6114084720611572 }, { "auxiliary_loss_clip": 0.01194316, "auxiliary_loss_mlp": 0.01033431, "balance_loss_clip": 1.06151438, "balance_loss_mlp": 1.02494299, "epoch": 0.187699152287621, "flos": 17018968310880.0, "grad_norm": 1.9696875971436947, "language_loss": 0.82133418, "learning_rate": 3.7448873330606154e-06, "loss": 0.84361172, "num_input_tokens_seen": 33126680, "step": 1561, "time_per_iteration": 2.559666395187378 }, { "auxiliary_loss_clip": 0.01169048, "auxiliary_loss_mlp": 0.01031752, "balance_loss_clip": 1.05756629, "balance_loss_mlp": 1.02273989, "epoch": 0.18781939517826007, "flos": 22346388466080.0, "grad_norm": 2.042497014098876, "language_loss": 0.86926579, "learning_rate": 3.7445065058336914e-06, "loss": 0.89127374, "num_input_tokens_seen": 33145550, "step": 1562, "time_per_iteration": 2.6519699096679688 }, { "auxiliary_loss_clip": 0.01145733, "auxiliary_loss_mlp": 0.01031611, "balance_loss_clip": 1.04763532, "balance_loss_mlp": 1.02326691, "epoch": 0.18793963806889918, "flos": 14611481024640.0, "grad_norm": 2.0933912520800417, "language_loss": 0.864577, "learning_rate": 3.7441254139665176e-06, "loss": 0.88635039, "num_input_tokens_seen": 33161735, "step": 1563, "time_per_iteration": 2.632664442062378 }, { "auxiliary_loss_clip": 0.012255, "auxiliary_loss_mlp": 0.01029311, "balance_loss_clip": 1.06640935, "balance_loss_mlp": 1.02058506, "epoch": 0.18805988095953827, "flos": 17457078563040.0, "grad_norm": 1.8677465277892669, "language_loss": 0.82417226, "learning_rate": 3.743744057516905e-06, "loss": 0.84672034, "num_input_tokens_seen": 33179795, "step": 1564, "time_per_iteration": 2.52826189994812 }, { "auxiliary_loss_clip": 0.01160304, "auxiliary_loss_mlp": 0.0104241, "balance_loss_clip": 1.05290747, "balance_loss_mlp": 1.03367186, "epoch": 0.18818012385017735, "flos": 15043880480160.0, "grad_norm": 3.0123010725996378, "language_loss": 0.87109542, "learning_rate": 3.743362436542706e-06, "loss": 0.89312255, "num_input_tokens_seen": 33194485, "step": 1565, "time_per_iteration": 2.6131224632263184 }, { "auxiliary_loss_clip": 0.01219397, "auxiliary_loss_mlp": 0.01034032, "balance_loss_clip": 1.06131363, "balance_loss_mlp": 1.02606273, "epoch": 0.18830036674081646, "flos": 47551980610080.0, "grad_norm": 1.7368813967114918, "language_loss": 0.76696527, "learning_rate": 3.7429805511018115e-06, "loss": 0.78949964, "num_input_tokens_seen": 33216145, "step": 1566, "time_per_iteration": 2.7558865547180176 }, { "auxiliary_loss_clip": 0.0117265, "auxiliary_loss_mlp": 0.00764008, "balance_loss_clip": 1.05731332, "balance_loss_mlp": 1.00005245, "epoch": 0.18842060963145554, "flos": 30044631079680.0, "grad_norm": 3.884247808526907, "language_loss": 0.77719748, "learning_rate": 3.7425984012521524e-06, "loss": 0.7965641, "num_input_tokens_seen": 33236345, "step": 1567, "time_per_iteration": 2.6916370391845703 }, { "auxiliary_loss_clip": 0.01073431, "auxiliary_loss_mlp": 0.00755421, "balance_loss_clip": 1.01904058, "balance_loss_mlp": 0.99996483, "epoch": 0.18854085252209463, "flos": 70318383568800.0, "grad_norm": 0.7411105350979146, "language_loss": 0.60470361, "learning_rate": 3.7422159870517025e-06, "loss": 0.62299216, "num_input_tokens_seen": 33301600, "step": 1568, "time_per_iteration": 3.2178144454956055 }, { "auxiliary_loss_clip": 0.01188016, "auxiliary_loss_mlp": 0.01030199, "balance_loss_clip": 1.05599213, "balance_loss_mlp": 1.02228951, "epoch": 0.1886610954127337, "flos": 21289322092800.0, "grad_norm": 1.4785430882820996, "language_loss": 0.78946805, "learning_rate": 3.7418333085584717e-06, "loss": 0.81165028, "num_input_tokens_seen": 33322785, "step": 1569, "time_per_iteration": 2.6558899879455566 }, { "auxiliary_loss_clip": 0.01179928, "auxiliary_loss_mlp": 0.01026108, "balance_loss_clip": 1.05978954, "balance_loss_mlp": 1.01748919, "epoch": 0.18878133830337282, "flos": 17266822450560.0, "grad_norm": 2.722108686057843, "language_loss": 0.90787053, "learning_rate": 3.7414503658305128e-06, "loss": 0.92993093, "num_input_tokens_seen": 33340020, "step": 1570, "time_per_iteration": 2.6090943813323975 }, { "auxiliary_loss_clip": 0.01167639, "auxiliary_loss_mlp": 0.01025402, "balance_loss_clip": 1.05301893, "balance_loss_mlp": 1.01644361, "epoch": 0.1889015811940119, "flos": 25775211138720.0, "grad_norm": 2.7009787128954077, "language_loss": 0.77886009, "learning_rate": 3.7410671589259185e-06, "loss": 0.80079049, "num_input_tokens_seen": 33358620, "step": 1571, "time_per_iteration": 2.715012311935425 }, { "auxiliary_loss_clip": 0.01223405, "auxiliary_loss_mlp": 0.01032123, "balance_loss_clip": 1.06445074, "balance_loss_mlp": 1.02367687, "epoch": 0.18902182408465099, "flos": 21032201377440.0, "grad_norm": 2.5998797084607914, "language_loss": 0.79956758, "learning_rate": 3.7406836879028205e-06, "loss": 0.82212287, "num_input_tokens_seen": 33378845, "step": 1572, "time_per_iteration": 2.540544271469116 }, { "auxiliary_loss_clip": 0.01198946, "auxiliary_loss_mlp": 0.0103616, "balance_loss_clip": 1.05934036, "balance_loss_mlp": 1.02755344, "epoch": 0.1891420669752901, "flos": 22272124969440.0, "grad_norm": 1.9729426765456435, "language_loss": 0.76860642, "learning_rate": 3.7402999528193907e-06, "loss": 0.79095745, "num_input_tokens_seen": 33398345, "step": 1573, "time_per_iteration": 2.5858731269836426 }, { "auxiliary_loss_clip": 0.01162158, "auxiliary_loss_mlp": 0.00764069, "balance_loss_clip": 1.05531335, "balance_loss_mlp": 1.00011587, "epoch": 0.18926230986592918, "flos": 22017805776960.0, "grad_norm": 2.765135497672378, "language_loss": 0.85657179, "learning_rate": 3.739915953733842e-06, "loss": 0.87583411, "num_input_tokens_seen": 33416390, "step": 1574, "time_per_iteration": 2.629783868789673 }, { "auxiliary_loss_clip": 0.01217574, "auxiliary_loss_mlp": 0.0102893, "balance_loss_clip": 1.06012225, "balance_loss_mlp": 1.0205853, "epoch": 0.18938255275656826, "flos": 24462676230240.0, "grad_norm": 1.85919497832446, "language_loss": 0.81809211, "learning_rate": 3.7395316907044264e-06, "loss": 0.8405571, "num_input_tokens_seen": 33437175, "step": 1575, "time_per_iteration": 2.5643537044525146 }, { "auxiliary_loss_clip": 0.0120405, "auxiliary_loss_mlp": 0.01029081, "balance_loss_clip": 1.05869412, "balance_loss_mlp": 1.02028346, "epoch": 0.18950279564720737, "flos": 24427052994240.0, "grad_norm": 1.7839700062286845, "language_loss": 0.79791981, "learning_rate": 3.7391471637894364e-06, "loss": 0.82025111, "num_input_tokens_seen": 33459440, "step": 1576, "time_per_iteration": 4.15323805809021 }, { "auxiliary_loss_clip": 0.01176215, "auxiliary_loss_mlp": 0.0103456, "balance_loss_clip": 1.05515504, "balance_loss_mlp": 1.0260036, "epoch": 0.18962303853784646, "flos": 19756295379840.0, "grad_norm": 1.8004494578520547, "language_loss": 0.84775454, "learning_rate": 3.738762373047205e-06, "loss": 0.86986232, "num_input_tokens_seen": 33479360, "step": 1577, "time_per_iteration": 2.645991563796997 }, { "auxiliary_loss_clip": 0.01177583, "auxiliary_loss_mlp": 0.0103436, "balance_loss_clip": 1.05679584, "balance_loss_mlp": 1.02631319, "epoch": 0.18974328142848554, "flos": 21032057709600.0, "grad_norm": 1.6719672137240338, "language_loss": 0.83331037, "learning_rate": 3.738377318536103e-06, "loss": 0.85542977, "num_input_tokens_seen": 33499245, "step": 1578, "time_per_iteration": 2.6465797424316406 }, { "auxiliary_loss_clip": 0.01213714, "auxiliary_loss_mlp": 0.01030357, "balance_loss_clip": 1.06129837, "balance_loss_mlp": 1.02198863, "epoch": 0.18986352431912462, "flos": 12966125225760.0, "grad_norm": 2.229667885658858, "language_loss": 0.7115227, "learning_rate": 3.7379920003145447e-06, "loss": 0.73396343, "num_input_tokens_seen": 33513520, "step": 1579, "time_per_iteration": 2.4886372089385986 }, { "auxiliary_loss_clip": 0.01184947, "auxiliary_loss_mlp": 0.01036992, "balance_loss_clip": 1.06140327, "balance_loss_mlp": 1.02799749, "epoch": 0.18998376720976373, "flos": 23767912183200.0, "grad_norm": 1.6324199043327305, "language_loss": 0.83558357, "learning_rate": 3.7376064184409817e-06, "loss": 0.85780299, "num_input_tokens_seen": 33533100, "step": 1580, "time_per_iteration": 2.6358346939086914 }, { "auxiliary_loss_clip": 0.01189631, "auxiliary_loss_mlp": 0.01032733, "balance_loss_clip": 1.05925071, "balance_loss_mlp": 1.02372694, "epoch": 0.19010401010040281, "flos": 22966027009440.0, "grad_norm": 1.4672195208388208, "language_loss": 0.86933374, "learning_rate": 3.7372205729739063e-06, "loss": 0.89155734, "num_input_tokens_seen": 33554915, "step": 1581, "time_per_iteration": 3.511106252670288 }, { "auxiliary_loss_clip": 0.01209499, "auxiliary_loss_mlp": 0.01036417, "balance_loss_clip": 1.06089044, "balance_loss_mlp": 1.02758932, "epoch": 0.1902242529910419, "flos": 19135651161600.0, "grad_norm": 2.1617702535054835, "language_loss": 0.7155624, "learning_rate": 3.7368344639718514e-06, "loss": 0.73802155, "num_input_tokens_seen": 33572850, "step": 1582, "time_per_iteration": 2.5653626918792725 }, { "auxiliary_loss_clip": 0.01208653, "auxiliary_loss_mlp": 0.01031324, "balance_loss_clip": 1.06085873, "balance_loss_mlp": 1.02302098, "epoch": 0.190344495881681, "flos": 25483939782720.0, "grad_norm": 1.6440633041991406, "language_loss": 0.80596459, "learning_rate": 3.7364480914933895e-06, "loss": 0.82836437, "num_input_tokens_seen": 33593090, "step": 1583, "time_per_iteration": 3.5186731815338135 }, { "auxiliary_loss_clip": 0.01158099, "auxiliary_loss_mlp": 0.00764422, "balance_loss_clip": 1.05587792, "balance_loss_mlp": 1.00010049, "epoch": 0.1904647387723201, "flos": 26792847078240.0, "grad_norm": 1.8273218021083053, "language_loss": 0.80969834, "learning_rate": 3.7360614555971325e-06, "loss": 0.82892358, "num_input_tokens_seen": 33612745, "step": 1584, "time_per_iteration": 2.6918771266937256 }, { "auxiliary_loss_clip": 0.01207354, "auxiliary_loss_mlp": 0.00763961, "balance_loss_clip": 1.06255031, "balance_loss_mlp": 1.00002575, "epoch": 0.19058498166295917, "flos": 23987757482400.0, "grad_norm": 2.0057267911767465, "language_loss": 0.84950238, "learning_rate": 3.735674556341733e-06, "loss": 0.86921561, "num_input_tokens_seen": 33632360, "step": 1585, "time_per_iteration": 2.6118013858795166 }, { "auxiliary_loss_clip": 0.01191119, "auxiliary_loss_mlp": 0.01031159, "balance_loss_clip": 1.06359291, "balance_loss_mlp": 1.02281487, "epoch": 0.19070522455359826, "flos": 28293303496800.0, "grad_norm": 1.986034120989433, "language_loss": 0.82833493, "learning_rate": 3.7352873937858835e-06, "loss": 0.85055774, "num_input_tokens_seen": 33653895, "step": 1586, "time_per_iteration": 2.633615255355835 }, { "auxiliary_loss_clip": 0.01165713, "auxiliary_loss_mlp": 0.00763795, "balance_loss_clip": 1.05329633, "balance_loss_mlp": 1.00008059, "epoch": 0.19082546744423737, "flos": 25660224197760.0, "grad_norm": 2.3072822587014548, "language_loss": 0.71952212, "learning_rate": 3.734899967988316e-06, "loss": 0.73881721, "num_input_tokens_seen": 33672075, "step": 1587, "time_per_iteration": 2.6896934509277344 }, { "auxiliary_loss_clip": 0.01165571, "auxiliary_loss_mlp": 0.01028076, "balance_loss_clip": 1.0540154, "balance_loss_mlp": 1.01927233, "epoch": 0.19094571033487645, "flos": 19719486884160.0, "grad_norm": 4.028641658389406, "language_loss": 0.83709455, "learning_rate": 3.7345122790078026e-06, "loss": 0.85903102, "num_input_tokens_seen": 33689640, "step": 1588, "time_per_iteration": 2.5980019569396973 }, { "auxiliary_loss_clip": 0.01201554, "auxiliary_loss_mlp": 0.01034242, "balance_loss_clip": 1.05867863, "balance_loss_mlp": 1.02518845, "epoch": 0.19106595322551553, "flos": 21616324435680.0, "grad_norm": 3.341549804878232, "language_loss": 0.92669523, "learning_rate": 3.7341243269031556e-06, "loss": 0.94905317, "num_input_tokens_seen": 33708630, "step": 1589, "time_per_iteration": 2.5774757862091064 }, { "auxiliary_loss_clip": 0.01184022, "auxiliary_loss_mlp": 0.01029937, "balance_loss_clip": 1.05797958, "balance_loss_mlp": 1.02100778, "epoch": 0.19118619611615464, "flos": 29896894259520.0, "grad_norm": 1.7722019650728313, "language_loss": 0.77359694, "learning_rate": 3.7337361117332275e-06, "loss": 0.79573655, "num_input_tokens_seen": 33730370, "step": 1590, "time_per_iteration": 2.654031753540039 }, { "auxiliary_loss_clip": 0.01173318, "auxiliary_loss_mlp": 0.010404, "balance_loss_clip": 1.05351651, "balance_loss_mlp": 1.03200126, "epoch": 0.19130643900679373, "flos": 17273431171200.0, "grad_norm": 2.1063846431115287, "language_loss": 0.77293235, "learning_rate": 3.7333476335569087e-06, "loss": 0.79506958, "num_input_tokens_seen": 33748370, "step": 1591, "time_per_iteration": 2.627575397491455 }, { "auxiliary_loss_clip": 0.01187089, "auxiliary_loss_mlp": 0.01035687, "balance_loss_clip": 1.05791378, "balance_loss_mlp": 1.02690768, "epoch": 0.1914266818974328, "flos": 24826343400960.0, "grad_norm": 2.1482430089567206, "language_loss": 0.66823715, "learning_rate": 3.7329588924331325e-06, "loss": 0.69046491, "num_input_tokens_seen": 33769575, "step": 1592, "time_per_iteration": 2.6233150959014893 }, { "auxiliary_loss_clip": 0.01165825, "auxiliary_loss_mlp": 0.01036295, "balance_loss_clip": 1.05260336, "balance_loss_mlp": 1.02763498, "epoch": 0.1915469247880719, "flos": 18952470690240.0, "grad_norm": 1.8231585517395295, "language_loss": 0.82679534, "learning_rate": 3.732569888420871e-06, "loss": 0.84881651, "num_input_tokens_seen": 33789110, "step": 1593, "time_per_iteration": 2.6404924392700195 }, { "auxiliary_loss_clip": 0.01218912, "auxiliary_loss_mlp": 0.01032856, "balance_loss_clip": 1.0589695, "balance_loss_mlp": 1.0238142, "epoch": 0.191667167678711, "flos": 21032955633600.0, "grad_norm": 3.46211988945981, "language_loss": 0.82306457, "learning_rate": 3.732180621579134e-06, "loss": 0.84558231, "num_input_tokens_seen": 33808325, "step": 1594, "time_per_iteration": 2.5289199352264404 }, { "auxiliary_loss_clip": 0.01190013, "auxiliary_loss_mlp": 0.0104293, "balance_loss_clip": 1.06217837, "balance_loss_mlp": 1.03372693, "epoch": 0.1917874105693501, "flos": 34237668423360.0, "grad_norm": 2.370920626770873, "language_loss": 0.81123424, "learning_rate": 3.7317910919669745e-06, "loss": 0.83356369, "num_input_tokens_seen": 33829520, "step": 1595, "time_per_iteration": 2.7459299564361572 }, { "auxiliary_loss_clip": 0.01203639, "auxiliary_loss_mlp": 0.01029742, "balance_loss_clip": 1.06028295, "balance_loss_mlp": 1.02030694, "epoch": 0.19190765345998917, "flos": 23550616988160.0, "grad_norm": 2.076050065708934, "language_loss": 0.76260877, "learning_rate": 3.7314012996434826e-06, "loss": 0.78494257, "num_input_tokens_seen": 33848250, "step": 1596, "time_per_iteration": 2.5622806549072266 }, { "auxiliary_loss_clip": 0.01189901, "auxiliary_loss_mlp": 0.01030779, "balance_loss_clip": 1.05842471, "balance_loss_mlp": 1.02211869, "epoch": 0.19202789635062828, "flos": 19861333322880.0, "grad_norm": 1.9238476168512693, "language_loss": 0.80771923, "learning_rate": 3.7310112446677907e-06, "loss": 0.82992613, "num_input_tokens_seen": 33866160, "step": 1597, "time_per_iteration": 2.6686038970947266 }, { "auxiliary_loss_clip": 0.01222221, "auxiliary_loss_mlp": 0.01031919, "balance_loss_clip": 1.06328773, "balance_loss_mlp": 1.02297282, "epoch": 0.19214813924126736, "flos": 20922961150080.0, "grad_norm": 2.7172793607510335, "language_loss": 0.68802124, "learning_rate": 3.7306209270990695e-06, "loss": 0.71056265, "num_input_tokens_seen": 33884165, "step": 1598, "time_per_iteration": 2.566356658935547 }, { "auxiliary_loss_clip": 0.0119022, "auxiliary_loss_mlp": 0.01036306, "balance_loss_clip": 1.05891538, "balance_loss_mlp": 1.02773476, "epoch": 0.19226838213190645, "flos": 26359729283520.0, "grad_norm": 1.8613987911941912, "language_loss": 0.86533153, "learning_rate": 3.7302303469965292e-06, "loss": 0.88759679, "num_input_tokens_seen": 33903705, "step": 1599, "time_per_iteration": 2.6355082988739014 }, { "auxiliary_loss_clip": 0.01201382, "auxiliary_loss_mlp": 0.01029547, "balance_loss_clip": 1.060251, "balance_loss_mlp": 1.0207907, "epoch": 0.19238862502254553, "flos": 20850529418400.0, "grad_norm": 1.6782432076591238, "language_loss": 0.70862532, "learning_rate": 3.7298395044194206e-06, "loss": 0.73093468, "num_input_tokens_seen": 33922515, "step": 1600, "time_per_iteration": 2.5784804821014404 }, { "auxiliary_loss_clip": 0.0122166, "auxiliary_loss_mlp": 0.01032532, "balance_loss_clip": 1.0638454, "balance_loss_mlp": 1.02407408, "epoch": 0.19250886791318464, "flos": 21726067500480.0, "grad_norm": 1.7928914465742678, "language_loss": 0.94215477, "learning_rate": 3.7294483994270356e-06, "loss": 0.96469659, "num_input_tokens_seen": 33940840, "step": 1601, "time_per_iteration": 2.5269722938537598 }, { "auxiliary_loss_clip": 0.01152651, "auxiliary_loss_mlp": 0.01039104, "balance_loss_clip": 1.05399394, "balance_loss_mlp": 1.03061056, "epoch": 0.19262911080382372, "flos": 23367831603360.0, "grad_norm": 2.013198964801049, "language_loss": 0.77824128, "learning_rate": 3.7290570320787033e-06, "loss": 0.80015886, "num_input_tokens_seen": 33960420, "step": 1602, "time_per_iteration": 4.222150564193726 }, { "auxiliary_loss_clip": 0.01200823, "auxiliary_loss_mlp": 0.01031475, "balance_loss_clip": 1.05934012, "balance_loss_mlp": 1.02256393, "epoch": 0.1927493536944628, "flos": 21943506363360.0, "grad_norm": 2.292106146200165, "language_loss": 0.71358293, "learning_rate": 3.728665402433793e-06, "loss": 0.73590589, "num_input_tokens_seen": 33978990, "step": 1603, "time_per_iteration": 2.5631015300750732 }, { "auxiliary_loss_clip": 0.0118683, "auxiliary_loss_mlp": 0.01036235, "balance_loss_clip": 1.05854666, "balance_loss_mlp": 1.02763963, "epoch": 0.19286959658510192, "flos": 16545593992320.0, "grad_norm": 2.444445884758292, "language_loss": 0.85732341, "learning_rate": 3.7282735105517164e-06, "loss": 0.87955403, "num_input_tokens_seen": 33997115, "step": 1604, "time_per_iteration": 2.585517406463623 }, { "auxiliary_loss_clip": 0.01165036, "auxiliary_loss_mlp": 0.01034848, "balance_loss_clip": 1.05350697, "balance_loss_mlp": 1.0254066, "epoch": 0.192989839475741, "flos": 21616970940960.0, "grad_norm": 3.0448289317175083, "language_loss": 0.67270494, "learning_rate": 3.727881356491922e-06, "loss": 0.6947037, "num_input_tokens_seen": 34015525, "step": 1605, "time_per_iteration": 2.649437189102173 }, { "auxiliary_loss_clip": 0.01218423, "auxiliary_loss_mlp": 0.01036521, "balance_loss_clip": 1.06332231, "balance_loss_mlp": 1.02870095, "epoch": 0.19311008236638008, "flos": 19281520299840.0, "grad_norm": 2.009128647539327, "language_loss": 0.75909477, "learning_rate": 3.7274889403139002e-06, "loss": 0.78164423, "num_input_tokens_seen": 34033150, "step": 1606, "time_per_iteration": 2.5275051593780518 }, { "auxiliary_loss_clip": 0.01151789, "auxiliary_loss_mlp": 0.01028961, "balance_loss_clip": 1.05622196, "balance_loss_mlp": 1.02078366, "epoch": 0.1932303252570192, "flos": 28652373296640.0, "grad_norm": 2.3751813661541505, "language_loss": 0.78153938, "learning_rate": 3.727096262077179e-06, "loss": 0.80334687, "num_input_tokens_seen": 34052145, "step": 1607, "time_per_iteration": 3.630953788757324 }, { "auxiliary_loss_clip": 0.01206919, "auxiliary_loss_mlp": 0.01029455, "balance_loss_clip": 1.063012, "balance_loss_mlp": 1.02121782, "epoch": 0.19335056814765827, "flos": 18368994137280.0, "grad_norm": 1.8049408425342284, "language_loss": 0.85291505, "learning_rate": 3.7267033218413285e-06, "loss": 0.87527871, "num_input_tokens_seen": 34069940, "step": 1608, "time_per_iteration": 2.5049686431884766 }, { "auxiliary_loss_clip": 0.01145955, "auxiliary_loss_mlp": 0.01036818, "balance_loss_clip": 1.05125451, "balance_loss_mlp": 1.02823508, "epoch": 0.19347081103829736, "flos": 13260880526880.0, "grad_norm": 2.7495259540510473, "language_loss": 0.81344241, "learning_rate": 3.726310119665957e-06, "loss": 0.83527017, "num_input_tokens_seen": 34086275, "step": 1609, "time_per_iteration": 3.461958885192871 }, { "auxiliary_loss_clip": 0.01200363, "auxiliary_loss_mlp": 0.01036077, "balance_loss_clip": 1.05918467, "balance_loss_mlp": 1.02676666, "epoch": 0.19359105392893644, "flos": 20300125997280.0, "grad_norm": 1.7753030066152957, "language_loss": 0.8570888, "learning_rate": 3.725916655610713e-06, "loss": 0.87945318, "num_input_tokens_seen": 34105605, "step": 1610, "time_per_iteration": 2.5252175331115723 }, { "auxiliary_loss_clip": 0.01179816, "auxiliary_loss_mlp": 0.0103246, "balance_loss_clip": 1.05611539, "balance_loss_mlp": 1.02349591, "epoch": 0.19371129681957555, "flos": 20484599479200.0, "grad_norm": 2.315391262483644, "language_loss": 0.75476867, "learning_rate": 3.725522929735284e-06, "loss": 0.77689147, "num_input_tokens_seen": 34122540, "step": 1611, "time_per_iteration": 2.5725760459899902 }, { "auxiliary_loss_clip": 0.01194612, "auxiliary_loss_mlp": 0.01032464, "balance_loss_clip": 1.05667877, "balance_loss_mlp": 1.02364874, "epoch": 0.19383153971021463, "flos": 30445501832640.0, "grad_norm": 2.3037933048396373, "language_loss": 0.74285316, "learning_rate": 3.725128942099399e-06, "loss": 0.76512396, "num_input_tokens_seen": 34142940, "step": 1612, "time_per_iteration": 2.6657843589782715 }, { "auxiliary_loss_clip": 0.01179181, "auxiliary_loss_mlp": 0.01029457, "balance_loss_clip": 1.05645442, "balance_loss_mlp": 1.02095127, "epoch": 0.19395178260085372, "flos": 24569941024800.0, "grad_norm": 2.031447936009846, "language_loss": 0.79760963, "learning_rate": 3.7247346927628245e-06, "loss": 0.81969595, "num_input_tokens_seen": 34162875, "step": 1613, "time_per_iteration": 2.640533685684204 }, { "auxiliary_loss_clip": 0.01188114, "auxiliary_loss_mlp": 0.00764621, "balance_loss_clip": 1.06017637, "balance_loss_mlp": 0.99998891, "epoch": 0.19407202549149283, "flos": 28950612542880.0, "grad_norm": 1.8418713064744054, "language_loss": 0.78778857, "learning_rate": 3.7243401817853694e-06, "loss": 0.80731595, "num_input_tokens_seen": 34183565, "step": 1614, "time_per_iteration": 2.6671829223632812 }, { "auxiliary_loss_clip": 0.01194344, "auxiliary_loss_mlp": 0.01035346, "balance_loss_clip": 1.05786824, "balance_loss_mlp": 1.02631593, "epoch": 0.1941922683821319, "flos": 18004500876480.0, "grad_norm": 1.7660785016445073, "language_loss": 0.71864772, "learning_rate": 3.723945409226879e-06, "loss": 0.74094456, "num_input_tokens_seen": 34202055, "step": 1615, "time_per_iteration": 2.5417966842651367 }, { "auxiliary_loss_clip": 0.01205206, "auxiliary_loss_mlp": 0.01030408, "balance_loss_clip": 1.06041086, "balance_loss_mlp": 1.02249217, "epoch": 0.194312511272771, "flos": 9720339356640.0, "grad_norm": 2.1868296765014668, "language_loss": 0.79674506, "learning_rate": 3.723550375147241e-06, "loss": 0.81910121, "num_input_tokens_seen": 34216830, "step": 1616, "time_per_iteration": 2.5603108406066895 }, { "auxiliary_loss_clip": 0.01164012, "auxiliary_loss_mlp": 0.01030807, "balance_loss_clip": 1.05489469, "balance_loss_mlp": 1.02155638, "epoch": 0.19443275416341008, "flos": 27016212239520.0, "grad_norm": 2.394673016106551, "language_loss": 0.80281436, "learning_rate": 3.7231550796063816e-06, "loss": 0.82476258, "num_input_tokens_seen": 34236840, "step": 1617, "time_per_iteration": 2.664294481277466 }, { "auxiliary_loss_clip": 0.01199844, "auxiliary_loss_mlp": 0.01030376, "balance_loss_clip": 1.06482673, "balance_loss_mlp": 1.02054095, "epoch": 0.1945529970540492, "flos": 15846627660960.0, "grad_norm": 2.063817336064151, "language_loss": 0.65299606, "learning_rate": 3.722759522664266e-06, "loss": 0.67529821, "num_input_tokens_seen": 34254140, "step": 1618, "time_per_iteration": 2.5693135261535645 }, { "auxiliary_loss_clip": 0.01163093, "auxiliary_loss_mlp": 0.01031091, "balance_loss_clip": 1.05601323, "balance_loss_mlp": 1.02253771, "epoch": 0.19467323994468827, "flos": 19314988518240.0, "grad_norm": 1.7095837754196945, "language_loss": 0.81182063, "learning_rate": 3.7223637043809016e-06, "loss": 0.83376247, "num_input_tokens_seen": 34273120, "step": 1619, "time_per_iteration": 2.642812728881836 }, { "auxiliary_loss_clip": 0.0118089, "auxiliary_loss_mlp": 0.01036117, "balance_loss_clip": 1.06308007, "balance_loss_mlp": 1.02725363, "epoch": 0.19479348283532735, "flos": 24133231534080.0, "grad_norm": 1.8416881178399322, "language_loss": 0.86675805, "learning_rate": 3.7219676248163322e-06, "loss": 0.88892812, "num_input_tokens_seen": 34290285, "step": 1620, "time_per_iteration": 2.6450796127319336 }, { "auxiliary_loss_clip": 0.01212789, "auxiliary_loss_mlp": 0.01037639, "balance_loss_clip": 1.06668437, "balance_loss_mlp": 1.02894819, "epoch": 0.19491372572596646, "flos": 25775642142240.0, "grad_norm": 2.386031321743139, "language_loss": 0.93316865, "learning_rate": 3.721571284030643e-06, "loss": 0.95567292, "num_input_tokens_seen": 34310095, "step": 1621, "time_per_iteration": 2.6173455715179443 }, { "auxiliary_loss_clip": 0.01208058, "auxiliary_loss_mlp": 0.01027899, "balance_loss_clip": 1.06156945, "balance_loss_mlp": 1.01956081, "epoch": 0.19503396861660555, "flos": 19645223387520.0, "grad_norm": 2.0798692693407244, "language_loss": 0.79438788, "learning_rate": 3.7211746820839587e-06, "loss": 0.81674743, "num_input_tokens_seen": 34327190, "step": 1622, "time_per_iteration": 2.5690581798553467 }, { "auxiliary_loss_clip": 0.01115671, "auxiliary_loss_mlp": 0.01029652, "balance_loss_clip": 1.04993629, "balance_loss_mlp": 1.02060413, "epoch": 0.19515421150724463, "flos": 21033027467520.0, "grad_norm": 2.1577720020417406, "language_loss": 0.80834228, "learning_rate": 3.7207778190364437e-06, "loss": 0.82979548, "num_input_tokens_seen": 34345615, "step": 1623, "time_per_iteration": 2.711073398590088 }, { "auxiliary_loss_clip": 0.01124582, "auxiliary_loss_mlp": 0.01031193, "balance_loss_clip": 1.04939103, "balance_loss_mlp": 1.02294397, "epoch": 0.1952744543978837, "flos": 32961259588320.0, "grad_norm": 1.5261275718484901, "language_loss": 0.73849308, "learning_rate": 3.720380694948302e-06, "loss": 0.76005089, "num_input_tokens_seen": 34368500, "step": 1624, "time_per_iteration": 2.8145692348480225 }, { "auxiliary_loss_clip": 0.01075917, "auxiliary_loss_mlp": 0.01003953, "balance_loss_clip": 1.02317774, "balance_loss_mlp": 1.00216472, "epoch": 0.19539469728852282, "flos": 64044322527360.0, "grad_norm": 1.0436839865677245, "language_loss": 0.71302891, "learning_rate": 3.719983309879777e-06, "loss": 0.73382759, "num_input_tokens_seen": 34428280, "step": 1625, "time_per_iteration": 3.214569568634033 }, { "auxiliary_loss_clip": 0.01163977, "auxiliary_loss_mlp": 0.01034069, "balance_loss_clip": 1.05541158, "balance_loss_mlp": 1.02497292, "epoch": 0.1955149401791619, "flos": 13370910927360.0, "grad_norm": 1.7147655843589182, "language_loss": 0.77540326, "learning_rate": 3.719585663891151e-06, "loss": 0.79738379, "num_input_tokens_seen": 34445815, "step": 1626, "time_per_iteration": 2.6261205673217773 }, { "auxiliary_loss_clip": 0.01155671, "auxiliary_loss_mlp": 0.01037979, "balance_loss_clip": 1.05764341, "balance_loss_mlp": 1.02898479, "epoch": 0.195635183069801, "flos": 18728890027200.0, "grad_norm": 1.91110786527953, "language_loss": 0.793051, "learning_rate": 3.719187757042747e-06, "loss": 0.81498754, "num_input_tokens_seen": 34463635, "step": 1627, "time_per_iteration": 2.6519010066986084 }, { "auxiliary_loss_clip": 0.01094903, "auxiliary_loss_mlp": 0.01001017, "balance_loss_clip": 1.02560961, "balance_loss_mlp": 0.9993009, "epoch": 0.1957554259604401, "flos": 69313965070560.0, "grad_norm": 0.7271108139819289, "language_loss": 0.55041945, "learning_rate": 3.7187895893949275e-06, "loss": 0.57137865, "num_input_tokens_seen": 34530105, "step": 1628, "time_per_iteration": 4.273730278015137 }, { "auxiliary_loss_clip": 0.01147036, "auxiliary_loss_mlp": 0.01032237, "balance_loss_clip": 1.05265307, "balance_loss_mlp": 1.02259326, "epoch": 0.19587566885107918, "flos": 21069261291840.0, "grad_norm": 2.8465797237267636, "language_loss": 0.76120687, "learning_rate": 3.7183911610080937e-06, "loss": 0.78299963, "num_input_tokens_seen": 34546970, "step": 1629, "time_per_iteration": 3.4131269454956055 }, { "auxiliary_loss_clip": 0.01177444, "auxiliary_loss_mlp": 0.01034359, "balance_loss_clip": 1.05772328, "balance_loss_mlp": 1.02504289, "epoch": 0.19599591174171827, "flos": 22194664863360.0, "grad_norm": 2.545729508265524, "language_loss": 0.74847376, "learning_rate": 3.7179924719426872e-06, "loss": 0.77059174, "num_input_tokens_seen": 34564865, "step": 1630, "time_per_iteration": 2.6533257961273193 }, { "auxiliary_loss_clip": 0.01210076, "auxiliary_loss_mlp": 0.01036771, "balance_loss_clip": 1.06530893, "balance_loss_mlp": 1.02799749, "epoch": 0.19611615463235738, "flos": 23768379103680.0, "grad_norm": 3.259606934447909, "language_loss": 0.75696689, "learning_rate": 3.7175935222591885e-06, "loss": 0.7794354, "num_input_tokens_seen": 34584165, "step": 1631, "time_per_iteration": 2.5766682624816895 }, { "auxiliary_loss_clip": 0.01193526, "auxiliary_loss_mlp": 0.01034998, "balance_loss_clip": 1.0642693, "balance_loss_mlp": 1.02607477, "epoch": 0.19623639752299646, "flos": 28618222656000.0, "grad_norm": 1.9698641488323567, "language_loss": 0.74492049, "learning_rate": 3.717194312018118e-06, "loss": 0.76720572, "num_input_tokens_seen": 34603150, "step": 1632, "time_per_iteration": 3.5481059551239014 }, { "auxiliary_loss_clip": 0.01209572, "auxiliary_loss_mlp": 0.01039542, "balance_loss_clip": 1.06395686, "balance_loss_mlp": 1.03015447, "epoch": 0.19635664041363554, "flos": 21032704214880.0, "grad_norm": 2.36652071848309, "language_loss": 0.76145262, "learning_rate": 3.716794841280036e-06, "loss": 0.78394377, "num_input_tokens_seen": 34621855, "step": 1633, "time_per_iteration": 2.5657217502593994 }, { "auxiliary_loss_clip": 0.01213356, "auxiliary_loss_mlp": 0.01029167, "balance_loss_clip": 1.06364393, "balance_loss_mlp": 1.02085805, "epoch": 0.19647688330427462, "flos": 18879751622880.0, "grad_norm": 2.0822291567558646, "language_loss": 0.77452689, "learning_rate": 3.7163951101055407e-06, "loss": 0.79695213, "num_input_tokens_seen": 34639915, "step": 1634, "time_per_iteration": 2.563208818435669 }, { "auxiliary_loss_clip": 0.01184377, "auxiliary_loss_mlp": 0.01036639, "balance_loss_clip": 1.06110179, "balance_loss_mlp": 1.02777553, "epoch": 0.19659712619491373, "flos": 24242507678400.0, "grad_norm": 1.6803710278444641, "language_loss": 0.79023188, "learning_rate": 3.715995118555273e-06, "loss": 0.81244206, "num_input_tokens_seen": 34659890, "step": 1635, "time_per_iteration": 3.5436835289001465 }, { "auxiliary_loss_clip": 0.01156365, "auxiliary_loss_mlp": 0.0102864, "balance_loss_clip": 1.05449772, "balance_loss_mlp": 1.01952672, "epoch": 0.19671736908555282, "flos": 24717426426240.0, "grad_norm": 2.114272030620645, "language_loss": 0.85633528, "learning_rate": 3.71559486668991e-06, "loss": 0.87818539, "num_input_tokens_seen": 34678750, "step": 1636, "time_per_iteration": 2.672368049621582 }, { "auxiliary_loss_clip": 0.01210427, "auxiliary_loss_mlp": 0.00763374, "balance_loss_clip": 1.06356239, "balance_loss_mlp": 0.9999373, "epoch": 0.1968376119761919, "flos": 23842283430720.0, "grad_norm": 1.842287463755436, "language_loss": 0.77066553, "learning_rate": 3.715194354570169e-06, "loss": 0.7904036, "num_input_tokens_seen": 34698755, "step": 1637, "time_per_iteration": 2.618558883666992 }, { "auxiliary_loss_clip": 0.01205494, "auxiliary_loss_mlp": 0.010317, "balance_loss_clip": 1.06496084, "balance_loss_mlp": 1.02330756, "epoch": 0.196957854866831, "flos": 18113920688640.0, "grad_norm": 1.817828454662259, "language_loss": 0.83276379, "learning_rate": 3.714793582256809e-06, "loss": 0.85513574, "num_input_tokens_seen": 34715820, "step": 1638, "time_per_iteration": 2.5492160320281982 }, { "auxiliary_loss_clip": 0.01220806, "auxiliary_loss_mlp": 0.01035058, "balance_loss_clip": 1.06552815, "balance_loss_mlp": 1.02690959, "epoch": 0.1970780977574701, "flos": 21653132931360.0, "grad_norm": 2.2912249260111364, "language_loss": 0.85238034, "learning_rate": 3.7143925498106253e-06, "loss": 0.87493902, "num_input_tokens_seen": 34734360, "step": 1639, "time_per_iteration": 2.557776927947998 }, { "auxiliary_loss_clip": 0.01191633, "auxiliary_loss_mlp": 0.01034171, "balance_loss_clip": 1.05767453, "balance_loss_mlp": 1.0253551, "epoch": 0.19719834064810918, "flos": 20811817323840.0, "grad_norm": 1.88691283302152, "language_loss": 0.79222107, "learning_rate": 3.7139912572924558e-06, "loss": 0.81447911, "num_input_tokens_seen": 34753390, "step": 1640, "time_per_iteration": 2.601590633392334 }, { "auxiliary_loss_clip": 0.01199986, "auxiliary_loss_mlp": 0.01034874, "balance_loss_clip": 1.05754805, "balance_loss_mlp": 1.02595139, "epoch": 0.19731858353874826, "flos": 23434803957120.0, "grad_norm": 4.541608292349139, "language_loss": 0.80525386, "learning_rate": 3.7135897047631744e-06, "loss": 0.82760245, "num_input_tokens_seen": 34771275, "step": 1641, "time_per_iteration": 2.589700222015381 }, { "auxiliary_loss_clip": 0.01191271, "auxiliary_loss_mlp": 0.01032409, "balance_loss_clip": 1.06023502, "balance_loss_mlp": 1.02376616, "epoch": 0.19743882642938737, "flos": 23988188485920.0, "grad_norm": 2.212405972536695, "language_loss": 0.75888842, "learning_rate": 3.713187892283698e-06, "loss": 0.78112525, "num_input_tokens_seen": 34790885, "step": 1642, "time_per_iteration": 2.619173288345337 }, { "auxiliary_loss_clip": 0.0115905, "auxiliary_loss_mlp": 0.01029951, "balance_loss_clip": 1.05408382, "balance_loss_mlp": 1.02135062, "epoch": 0.19755906932002645, "flos": 15004342295520.0, "grad_norm": 2.1846098460009977, "language_loss": 0.86757362, "learning_rate": 3.71278581991498e-06, "loss": 0.88946366, "num_input_tokens_seen": 34806745, "step": 1643, "time_per_iteration": 2.649517774581909 }, { "auxiliary_loss_clip": 0.0117664, "auxiliary_loss_mlp": 0.00763535, "balance_loss_clip": 1.06013858, "balance_loss_mlp": 0.99992424, "epoch": 0.19767931221066554, "flos": 19494469542720.0, "grad_norm": 2.5090271257677657, "language_loss": 0.78903449, "learning_rate": 3.712383487718015e-06, "loss": 0.80843627, "num_input_tokens_seen": 34824985, "step": 1644, "time_per_iteration": 2.635206460952759 }, { "auxiliary_loss_clip": 0.01139271, "auxiliary_loss_mlp": 0.01034087, "balance_loss_clip": 1.05577314, "balance_loss_mlp": 1.02522409, "epoch": 0.19779955510130465, "flos": 25737899805600.0, "grad_norm": 1.8006880214837198, "language_loss": 0.87237012, "learning_rate": 3.7119808957538365e-06, "loss": 0.89410377, "num_input_tokens_seen": 34843980, "step": 1645, "time_per_iteration": 2.7295961380004883 }, { "auxiliary_loss_clip": 0.01180376, "auxiliary_loss_mlp": 0.01036994, "balance_loss_clip": 1.05471098, "balance_loss_mlp": 1.02794552, "epoch": 0.19791979799194373, "flos": 20777702600160.0, "grad_norm": 2.023068812682375, "language_loss": 0.7996074, "learning_rate": 3.711578044083517e-06, "loss": 0.8217811, "num_input_tokens_seen": 34860780, "step": 1646, "time_per_iteration": 2.5915451049804688 }, { "auxiliary_loss_clip": 0.01189767, "auxiliary_loss_mlp": 0.01033461, "balance_loss_clip": 1.05899167, "balance_loss_mlp": 1.02487779, "epoch": 0.1980400408825828, "flos": 25589013642720.0, "grad_norm": 2.2028559829046674, "language_loss": 0.74707198, "learning_rate": 3.7111749327681698e-06, "loss": 0.76930428, "num_input_tokens_seen": 34880815, "step": 1647, "time_per_iteration": 2.6408491134643555 }, { "auxiliary_loss_clip": 0.01207743, "auxiliary_loss_mlp": 0.01028193, "balance_loss_clip": 1.0635041, "balance_loss_mlp": 1.01950228, "epoch": 0.1981602837732219, "flos": 23513844409440.0, "grad_norm": 2.0955747260121127, "language_loss": 0.8616432, "learning_rate": 3.7107715618689455e-06, "loss": 0.88400257, "num_input_tokens_seen": 34899790, "step": 1648, "time_per_iteration": 2.5688881874084473 }, { "auxiliary_loss_clip": 0.0120206, "auxiliary_loss_mlp": 0.01037754, "balance_loss_clip": 1.06287885, "balance_loss_mlp": 1.02926052, "epoch": 0.198280526663861, "flos": 23185369471200.0, "grad_norm": 1.4684601864765385, "language_loss": 0.83608973, "learning_rate": 3.710367931447035e-06, "loss": 0.85848784, "num_input_tokens_seen": 34921570, "step": 1649, "time_per_iteration": 2.600818157196045 }, { "auxiliary_loss_clip": 0.01213805, "auxiliary_loss_mlp": 0.0104683, "balance_loss_clip": 1.06423843, "balance_loss_mlp": 1.03771615, "epoch": 0.1984007695545001, "flos": 21689474506560.0, "grad_norm": 1.9940282422820783, "language_loss": 0.87114358, "learning_rate": 3.70996404156367e-06, "loss": 0.89374989, "num_input_tokens_seen": 34941205, "step": 1650, "time_per_iteration": 2.5815250873565674 }, { "auxiliary_loss_clip": 0.01148969, "auxiliary_loss_mlp": 0.01028777, "balance_loss_clip": 1.05373573, "balance_loss_mlp": 1.01967514, "epoch": 0.19852101244513917, "flos": 36064013759040.0, "grad_norm": 1.8947246823790818, "language_loss": 0.73135448, "learning_rate": 3.7095598922801187e-06, "loss": 0.75313193, "num_input_tokens_seen": 34963280, "step": 1651, "time_per_iteration": 2.7750494480133057 }, { "auxiliary_loss_clip": 0.01220796, "auxiliary_loss_mlp": 0.01030668, "balance_loss_clip": 1.06510282, "balance_loss_mlp": 1.02269304, "epoch": 0.19864125533577828, "flos": 23105898015360.0, "grad_norm": 2.91118528801905, "language_loss": 0.76189345, "learning_rate": 3.7091554836576914e-06, "loss": 0.78440815, "num_input_tokens_seen": 34979955, "step": 1652, "time_per_iteration": 2.5308971405029297 }, { "auxiliary_loss_clip": 0.01204022, "auxiliary_loss_mlp": 0.00763152, "balance_loss_clip": 1.06313801, "balance_loss_mlp": 0.9999184, "epoch": 0.19876149822641737, "flos": 24608509451520.0, "grad_norm": 2.5494691616673726, "language_loss": 0.82870996, "learning_rate": 3.708750815757736e-06, "loss": 0.84838164, "num_input_tokens_seen": 35000725, "step": 1653, "time_per_iteration": 3.373138427734375 }, { "auxiliary_loss_clip": 0.01206305, "auxiliary_loss_mlp": 0.01035853, "balance_loss_clip": 1.06329632, "balance_loss_mlp": 1.02694225, "epoch": 0.19888174111705645, "flos": 32196649830720.0, "grad_norm": 2.2492931624975023, "language_loss": 0.73249674, "learning_rate": 3.7083458886416407e-06, "loss": 0.75491828, "num_input_tokens_seen": 35019920, "step": 1654, "time_per_iteration": 3.3935606479644775 }, { "auxiliary_loss_clip": 0.01145042, "auxiliary_loss_mlp": 0.01032124, "balance_loss_clip": 1.05562162, "balance_loss_mlp": 1.02330828, "epoch": 0.19900198400769553, "flos": 24608473534560.0, "grad_norm": 1.9173391271462799, "language_loss": 0.8806228, "learning_rate": 3.707940702370832e-06, "loss": 0.90239447, "num_input_tokens_seen": 35040765, "step": 1655, "time_per_iteration": 2.713728189468384 }, { "auxiliary_loss_clip": 0.01099504, "auxiliary_loss_mlp": 0.01006071, "balance_loss_clip": 1.02187526, "balance_loss_mlp": 1.00425911, "epoch": 0.19912222689833464, "flos": 67915816906080.0, "grad_norm": 0.7574106420375869, "language_loss": 0.5831067, "learning_rate": 3.707535257006777e-06, "loss": 0.60416245, "num_input_tokens_seen": 35106390, "step": 1656, "time_per_iteration": 3.2239809036254883 }, { "auxiliary_loss_clip": 0.01189391, "auxiliary_loss_mlp": 0.01031458, "balance_loss_clip": 1.05835915, "balance_loss_mlp": 1.0222491, "epoch": 0.19924246978897373, "flos": 15742343974080.0, "grad_norm": 2.6276628140568628, "language_loss": 0.88787675, "learning_rate": 3.707129552610981e-06, "loss": 0.9100852, "num_input_tokens_seen": 35125040, "step": 1657, "time_per_iteration": 2.6063807010650635 }, { "auxiliary_loss_clip": 0.01180154, "auxiliary_loss_mlp": 0.01033107, "balance_loss_clip": 1.05803394, "balance_loss_mlp": 1.02404141, "epoch": 0.1993627126796128, "flos": 17566570209120.0, "grad_norm": 1.832498636849933, "language_loss": 0.73967081, "learning_rate": 3.70672358924499e-06, "loss": 0.76180339, "num_input_tokens_seen": 35144280, "step": 1658, "time_per_iteration": 3.4261934757232666 }, { "auxiliary_loss_clip": 0.01175742, "auxiliary_loss_mlp": 0.01032754, "balance_loss_clip": 1.06224573, "balance_loss_mlp": 1.02425492, "epoch": 0.19948295557025192, "flos": 40843832016000.0, "grad_norm": 2.613407350644294, "language_loss": 0.78843188, "learning_rate": 3.706317366970386e-06, "loss": 0.81051689, "num_input_tokens_seen": 35165280, "step": 1659, "time_per_iteration": 2.7950263023376465 }, { "auxiliary_loss_clip": 0.01224149, "auxiliary_loss_mlp": 0.00764508, "balance_loss_clip": 1.06405044, "balance_loss_mlp": 0.99992436, "epoch": 0.199603198460891, "flos": 25082422524480.0, "grad_norm": 2.053953210416391, "language_loss": 0.83682764, "learning_rate": 3.705910885848795e-06, "loss": 0.85671419, "num_input_tokens_seen": 35183655, "step": 1660, "time_per_iteration": 2.5746662616729736 }, { "auxiliary_loss_clip": 0.01202688, "auxiliary_loss_mlp": 0.01035945, "balance_loss_clip": 1.06161368, "balance_loss_mlp": 1.02799082, "epoch": 0.19972344135153008, "flos": 20084123812800.0, "grad_norm": 1.9726720212921693, "language_loss": 0.84404874, "learning_rate": 3.705504145941879e-06, "loss": 0.86643505, "num_input_tokens_seen": 35201825, "step": 1661, "time_per_iteration": 3.4236865043640137 }, { "auxiliary_loss_clip": 0.01217037, "auxiliary_loss_mlp": 0.01028248, "balance_loss_clip": 1.06175303, "balance_loss_mlp": 1.01950967, "epoch": 0.1998436842421692, "flos": 23727476074560.0, "grad_norm": 1.9083302962591178, "language_loss": 0.78813607, "learning_rate": 3.7050971473113403e-06, "loss": 0.8105889, "num_input_tokens_seen": 35221600, "step": 1662, "time_per_iteration": 2.5567996501922607 }, { "auxiliary_loss_clip": 0.01199775, "auxiliary_loss_mlp": 0.00763777, "balance_loss_clip": 1.05913281, "balance_loss_mlp": 0.99995971, "epoch": 0.19996392713280828, "flos": 36102366684000.0, "grad_norm": 1.8439175207895657, "language_loss": 0.80089527, "learning_rate": 3.7046898900189196e-06, "loss": 0.82053077, "num_input_tokens_seen": 35245935, "step": 1663, "time_per_iteration": 2.7208449840545654 }, { "auxiliary_loss_clip": 0.0117942, "auxiliary_loss_mlp": 0.0103625, "balance_loss_clip": 1.06105232, "balance_loss_mlp": 1.0266118, "epoch": 0.20008417002344736, "flos": 23657666280960.0, "grad_norm": 1.657557577309439, "language_loss": 0.83152765, "learning_rate": 3.704282374126398e-06, "loss": 0.85368431, "num_input_tokens_seen": 35265615, "step": 1664, "time_per_iteration": 2.6640894412994385 }, { "auxiliary_loss_clip": 0.01170492, "auxiliary_loss_mlp": 0.01032509, "balance_loss_clip": 1.05534029, "balance_loss_mlp": 1.02374756, "epoch": 0.20020441291408644, "flos": 21872080306560.0, "grad_norm": 1.6848495190401513, "language_loss": 0.87229925, "learning_rate": 3.7038745996955954e-06, "loss": 0.89432919, "num_input_tokens_seen": 35284960, "step": 1665, "time_per_iteration": 2.6417365074157715 }, { "auxiliary_loss_clip": 0.0118477, "auxiliary_loss_mlp": 0.01041687, "balance_loss_clip": 1.05939913, "balance_loss_mlp": 1.03216267, "epoch": 0.20032465580472555, "flos": 23179694591520.0, "grad_norm": 2.554012348550441, "language_loss": 0.7231217, "learning_rate": 3.703466566788371e-06, "loss": 0.7453863, "num_input_tokens_seen": 35304090, "step": 1666, "time_per_iteration": 2.641608238220215 }, { "auxiliary_loss_clip": 0.01182226, "auxiliary_loss_mlp": 0.01030069, "balance_loss_clip": 1.059582, "balance_loss_mlp": 1.0209856, "epoch": 0.20044489869536464, "flos": 23873524797600.0, "grad_norm": 3.391725293524286, "language_loss": 0.74479079, "learning_rate": 3.703058275466622e-06, "loss": 0.76691371, "num_input_tokens_seen": 35323325, "step": 1667, "time_per_iteration": 2.6713383197784424 }, { "auxiliary_loss_clip": 0.01190235, "auxiliary_loss_mlp": 0.01032483, "balance_loss_clip": 1.05913377, "balance_loss_mlp": 1.02404332, "epoch": 0.20056514158600372, "flos": 21945230377440.0, "grad_norm": 1.557421940035051, "language_loss": 0.7748962, "learning_rate": 3.7026497257922877e-06, "loss": 0.79712337, "num_input_tokens_seen": 35343635, "step": 1668, "time_per_iteration": 2.586513042449951 }, { "auxiliary_loss_clip": 0.01153596, "auxiliary_loss_mlp": 0.01034568, "balance_loss_clip": 1.05409122, "balance_loss_mlp": 1.02568078, "epoch": 0.20068538447664283, "flos": 23879163760320.0, "grad_norm": 1.6395240478658963, "language_loss": 0.84792531, "learning_rate": 3.7022409178273436e-06, "loss": 0.86980689, "num_input_tokens_seen": 35364615, "step": 1669, "time_per_iteration": 2.6432266235351562 }, { "auxiliary_loss_clip": 0.01203031, "auxiliary_loss_mlp": 0.01028859, "balance_loss_clip": 1.06112325, "balance_loss_mlp": 1.02004969, "epoch": 0.2008056273672819, "flos": 18442826630400.0, "grad_norm": 1.7504159238124648, "language_loss": 0.7840293, "learning_rate": 3.7018318516338054e-06, "loss": 0.80634826, "num_input_tokens_seen": 35383775, "step": 1670, "time_per_iteration": 2.515733003616333 }, { "auxiliary_loss_clip": 0.01209306, "auxiliary_loss_mlp": 0.01036952, "balance_loss_clip": 1.06256294, "balance_loss_mlp": 1.02829146, "epoch": 0.200925870257921, "flos": 23659533962880.0, "grad_norm": 2.4594529493347674, "language_loss": 0.82124388, "learning_rate": 3.7014225272737284e-06, "loss": 0.84370637, "num_input_tokens_seen": 35403000, "step": 1671, "time_per_iteration": 2.5612027645111084 }, { "auxiliary_loss_clip": 0.0119282, "auxiliary_loss_mlp": 0.01034189, "balance_loss_clip": 1.05691981, "balance_loss_mlp": 1.02425337, "epoch": 0.20104611314856008, "flos": 16217119054080.0, "grad_norm": 4.7435981243913945, "language_loss": 0.73947769, "learning_rate": 3.701012944809207e-06, "loss": 0.76174784, "num_input_tokens_seen": 35420115, "step": 1672, "time_per_iteration": 2.556344509124756 }, { "auxiliary_loss_clip": 0.01191591, "auxiliary_loss_mlp": 0.0076387, "balance_loss_clip": 1.06143737, "balance_loss_mlp": 0.99994171, "epoch": 0.2011663560391992, "flos": 21397377060480.0, "grad_norm": 1.9187462915889675, "language_loss": 0.79043132, "learning_rate": 3.700603104302374e-06, "loss": 0.80998588, "num_input_tokens_seen": 35439925, "step": 1673, "time_per_iteration": 2.635767698287964 }, { "auxiliary_loss_clip": 0.01059899, "auxiliary_loss_mlp": 0.01002593, "balance_loss_clip": 1.01647973, "balance_loss_mlp": 1.00090063, "epoch": 0.20128659892983827, "flos": 62229470618880.0, "grad_norm": 0.8903742938815814, "language_loss": 0.55936515, "learning_rate": 3.7001930058154027e-06, "loss": 0.57999003, "num_input_tokens_seen": 35504885, "step": 1674, "time_per_iteration": 3.2265915870666504 }, { "auxiliary_loss_clip": 0.01173507, "auxiliary_loss_mlp": 0.01030326, "balance_loss_clip": 1.05899787, "balance_loss_mlp": 1.02148604, "epoch": 0.20140684182047736, "flos": 28438741631520.0, "grad_norm": 2.5115017468892487, "language_loss": 0.79536164, "learning_rate": 3.6997826494105037e-06, "loss": 0.81740004, "num_input_tokens_seen": 35525330, "step": 1675, "time_per_iteration": 2.689924478530884 }, { "auxiliary_loss_clip": 0.01191656, "auxiliary_loss_mlp": 0.01037966, "balance_loss_clip": 1.0601753, "balance_loss_mlp": 1.02888215, "epoch": 0.20152708471111647, "flos": 28074068785920.0, "grad_norm": 2.2083510511638424, "language_loss": 0.69704247, "learning_rate": 3.6993720351499286e-06, "loss": 0.71933866, "num_input_tokens_seen": 35546455, "step": 1676, "time_per_iteration": 2.637558698654175 }, { "auxiliary_loss_clip": 0.01180367, "auxiliary_loss_mlp": 0.01034289, "balance_loss_clip": 1.05935836, "balance_loss_mlp": 1.02618325, "epoch": 0.20164732760175555, "flos": 23549754981120.0, "grad_norm": 1.653871864446005, "language_loss": 0.77367246, "learning_rate": 3.6989611630959666e-06, "loss": 0.79581904, "num_input_tokens_seen": 35565010, "step": 1677, "time_per_iteration": 2.631244659423828 }, { "auxiliary_loss_clip": 0.01096951, "auxiliary_loss_mlp": 0.01003114, "balance_loss_clip": 1.01889133, "balance_loss_mlp": 1.00132585, "epoch": 0.20176757049239463, "flos": 71100628553760.0, "grad_norm": 0.6886101730448227, "language_loss": 0.58385551, "learning_rate": 3.6985500333109474e-06, "loss": 0.60485625, "num_input_tokens_seen": 35633340, "step": 1678, "time_per_iteration": 3.227537155151367 }, { "auxiliary_loss_clip": 0.01166199, "auxiliary_loss_mlp": 0.0104055, "balance_loss_clip": 1.05634558, "balance_loss_mlp": 1.0311203, "epoch": 0.20188781338303372, "flos": 21430162856640.0, "grad_norm": 2.1800947943715734, "language_loss": 0.76814783, "learning_rate": 3.6981386458572385e-06, "loss": 0.79021531, "num_input_tokens_seen": 35651315, "step": 1679, "time_per_iteration": 3.3676540851593018 }, { "auxiliary_loss_clip": 0.01168605, "auxiliary_loss_mlp": 0.01036269, "balance_loss_clip": 1.0557512, "balance_loss_mlp": 1.02731669, "epoch": 0.20200805627367283, "flos": 11546217771840.0, "grad_norm": 2.3121208529416717, "language_loss": 0.76403272, "learning_rate": 3.6977270007972468e-06, "loss": 0.78608149, "num_input_tokens_seen": 35668850, "step": 1680, "time_per_iteration": 3.6674752235412598 }, { "auxiliary_loss_clip": 0.01195659, "auxiliary_loss_mlp": 0.01035521, "balance_loss_clip": 1.06182289, "balance_loss_mlp": 1.02668738, "epoch": 0.2021282991643119, "flos": 28545395837760.0, "grad_norm": 2.1603277107574175, "language_loss": 0.72571087, "learning_rate": 3.6973150981934196e-06, "loss": 0.74802268, "num_input_tokens_seen": 35690080, "step": 1681, "time_per_iteration": 2.650455951690674 }, { "auxiliary_loss_clip": 0.01224904, "auxiliary_loss_mlp": 0.01036781, "balance_loss_clip": 1.06478834, "balance_loss_mlp": 1.02756071, "epoch": 0.202248542054951, "flos": 17923449074400.0, "grad_norm": 2.619985296653117, "language_loss": 0.8330164, "learning_rate": 3.6969029381082415e-06, "loss": 0.85563326, "num_input_tokens_seen": 35706075, "step": 1682, "time_per_iteration": 2.537087917327881 }, { "auxiliary_loss_clip": 0.01188536, "auxiliary_loss_mlp": 0.01031478, "balance_loss_clip": 1.06154466, "balance_loss_mlp": 1.02356279, "epoch": 0.2023687849455901, "flos": 19864637683200.0, "grad_norm": 1.9620307201192873, "language_loss": 0.79509515, "learning_rate": 3.696490520604237e-06, "loss": 0.81729531, "num_input_tokens_seen": 35724765, "step": 1683, "time_per_iteration": 2.5709002017974854 }, { "auxiliary_loss_clip": 0.01200409, "auxiliary_loss_mlp": 0.0103901, "balance_loss_clip": 1.06060731, "balance_loss_mlp": 1.03098094, "epoch": 0.20248902783622919, "flos": 22564725252960.0, "grad_norm": 1.6362137665576997, "language_loss": 0.80448121, "learning_rate": 3.696077845743968e-06, "loss": 0.82687533, "num_input_tokens_seen": 35744355, "step": 1684, "time_per_iteration": 3.526873826980591 }, { "auxiliary_loss_clip": 0.0121704, "auxiliary_loss_mlp": 0.01035529, "balance_loss_clip": 1.0596478, "balance_loss_mlp": 1.02654672, "epoch": 0.20260927072686827, "flos": 22709732384160.0, "grad_norm": 2.576683378070055, "language_loss": 0.73501784, "learning_rate": 3.69566491359004e-06, "loss": 0.75754356, "num_input_tokens_seen": 35761000, "step": 1685, "time_per_iteration": 2.579930067062378 }, { "auxiliary_loss_clip": 0.01186988, "auxiliary_loss_mlp": 0.01035983, "balance_loss_clip": 1.05908585, "balance_loss_mlp": 1.02739429, "epoch": 0.20272951361750738, "flos": 51024004997280.0, "grad_norm": 1.7679714777367783, "language_loss": 0.69527352, "learning_rate": 3.695251724205092e-06, "loss": 0.71750325, "num_input_tokens_seen": 35785360, "step": 1686, "time_per_iteration": 3.735586166381836 }, { "auxiliary_loss_clip": 0.01218043, "auxiliary_loss_mlp": 0.01026209, "balance_loss_clip": 1.06295753, "balance_loss_mlp": 1.01778722, "epoch": 0.20284975650814646, "flos": 26578137904320.0, "grad_norm": 1.5843099590691068, "language_loss": 0.86253119, "learning_rate": 3.6948382776518054e-06, "loss": 0.8849737, "num_input_tokens_seen": 35806065, "step": 1687, "time_per_iteration": 2.5769402980804443 }, { "auxiliary_loss_clip": 0.01182625, "auxiliary_loss_mlp": 0.01029461, "balance_loss_clip": 1.05620623, "balance_loss_mlp": 1.02074039, "epoch": 0.20296999939878554, "flos": 16034225918400.0, "grad_norm": 2.2103506772966335, "language_loss": 0.795564, "learning_rate": 3.6944245739929e-06, "loss": 0.81768483, "num_input_tokens_seen": 35822225, "step": 1688, "time_per_iteration": 2.5933589935302734 }, { "auxiliary_loss_clip": 0.01209289, "auxiliary_loss_mlp": 0.01034738, "balance_loss_clip": 1.06519842, "balance_loss_mlp": 1.02567768, "epoch": 0.20309024228942463, "flos": 19203377771520.0, "grad_norm": 2.2875447798198847, "language_loss": 0.72033083, "learning_rate": 3.6940106132911332e-06, "loss": 0.74277103, "num_input_tokens_seen": 35839410, "step": 1689, "time_per_iteration": 2.561265468597412 }, { "auxiliary_loss_clip": 0.0121012, "auxiliary_loss_mlp": 0.0103042, "balance_loss_clip": 1.06538951, "balance_loss_mlp": 1.02102017, "epoch": 0.20321048518006374, "flos": 22821091712160.0, "grad_norm": 3.0833692248241404, "language_loss": 0.88665164, "learning_rate": 3.6935963956093037e-06, "loss": 0.90905702, "num_input_tokens_seen": 35859495, "step": 1690, "time_per_iteration": 2.613555431365967 }, { "auxiliary_loss_clip": 0.0119516, "auxiliary_loss_mlp": 0.01027296, "balance_loss_clip": 1.06004274, "balance_loss_mlp": 1.01906455, "epoch": 0.20333072807070282, "flos": 19096400312640.0, "grad_norm": 1.6863840161755372, "language_loss": 0.68777525, "learning_rate": 3.6931819210102474e-06, "loss": 0.70999986, "num_input_tokens_seen": 35878890, "step": 1691, "time_per_iteration": 2.58447265625 }, { "auxiliary_loss_clip": 0.01219335, "auxiliary_loss_mlp": 0.01027263, "balance_loss_clip": 1.06229985, "balance_loss_mlp": 1.01923418, "epoch": 0.2034509709613419, "flos": 18180964876320.0, "grad_norm": 1.7092048361595056, "language_loss": 0.8410821, "learning_rate": 3.6927671895568402e-06, "loss": 0.86354804, "num_input_tokens_seen": 35897950, "step": 1692, "time_per_iteration": 2.559523582458496 }, { "auxiliary_loss_clip": 0.0122194, "auxiliary_loss_mlp": 0.01032405, "balance_loss_clip": 1.06522155, "balance_loss_mlp": 1.02282655, "epoch": 0.20357121385198101, "flos": 22923902803680.0, "grad_norm": 3.1797899660415108, "language_loss": 0.86789823, "learning_rate": 3.692352201311996e-06, "loss": 0.89044166, "num_input_tokens_seen": 35916800, "step": 1693, "time_per_iteration": 2.5340824127197266 }, { "auxiliary_loss_clip": 0.01171346, "auxiliary_loss_mlp": 0.01031456, "balance_loss_clip": 1.05733883, "balance_loss_mlp": 1.0225625, "epoch": 0.2036914567426201, "flos": 20922135060000.0, "grad_norm": 2.1747670694038903, "language_loss": 0.76916575, "learning_rate": 3.6919369563386687e-06, "loss": 0.79119384, "num_input_tokens_seen": 35936600, "step": 1694, "time_per_iteration": 2.645721673965454 }, { "auxiliary_loss_clip": 0.01185624, "auxiliary_loss_mlp": 0.01033454, "balance_loss_clip": 1.05984402, "balance_loss_mlp": 1.0246799, "epoch": 0.20381169963325918, "flos": 15519158397600.0, "grad_norm": 2.3212688036562366, "language_loss": 0.79391742, "learning_rate": 3.69152145469985e-06, "loss": 0.81610823, "num_input_tokens_seen": 35953645, "step": 1695, "time_per_iteration": 2.561762571334839 }, { "auxiliary_loss_clip": 0.01161547, "auxiliary_loss_mlp": 0.01031352, "balance_loss_clip": 1.0533731, "balance_loss_mlp": 1.02259612, "epoch": 0.20393194252389826, "flos": 28833147331680.0, "grad_norm": 1.7083343008972982, "language_loss": 0.81886411, "learning_rate": 3.691105696458572e-06, "loss": 0.84079313, "num_input_tokens_seen": 35970940, "step": 1696, "time_per_iteration": 2.7226619720458984 }, { "auxiliary_loss_clip": 0.01217099, "auxiliary_loss_mlp": 0.01036168, "balance_loss_clip": 1.06407332, "balance_loss_mlp": 1.02791905, "epoch": 0.20405218541453737, "flos": 22488558157440.0, "grad_norm": 2.9511175480669953, "language_loss": 0.68005443, "learning_rate": 3.690689681677904e-06, "loss": 0.70258713, "num_input_tokens_seen": 35989410, "step": 1697, "time_per_iteration": 2.517850637435913 }, { "auxiliary_loss_clip": 0.0119092, "auxiliary_loss_mlp": 0.01029472, "balance_loss_clip": 1.05870783, "balance_loss_mlp": 1.02120447, "epoch": 0.20417242830517646, "flos": 25374412219680.0, "grad_norm": 1.735711778392426, "language_loss": 0.88167202, "learning_rate": 3.690273410420956e-06, "loss": 0.90387595, "num_input_tokens_seen": 36009175, "step": 1698, "time_per_iteration": 2.640824556350708 }, { "auxiliary_loss_clip": 0.01198243, "auxiliary_loss_mlp": 0.01031847, "balance_loss_clip": 1.05776942, "balance_loss_mlp": 1.02288246, "epoch": 0.20429267119581554, "flos": 14793081149760.0, "grad_norm": 19.650132368972887, "language_loss": 0.76663649, "learning_rate": 3.689856882750875e-06, "loss": 0.78893739, "num_input_tokens_seen": 36024375, "step": 1699, "time_per_iteration": 2.525184154510498 }, { "auxiliary_loss_clip": 0.01199916, "auxiliary_loss_mlp": 0.01030638, "balance_loss_clip": 1.06156385, "balance_loss_mlp": 1.02241874, "epoch": 0.20441291408645465, "flos": 17781854054400.0, "grad_norm": 1.6340773242833508, "language_loss": 0.78520262, "learning_rate": 3.6894400987308486e-06, "loss": 0.80750811, "num_input_tokens_seen": 36041895, "step": 1700, "time_per_iteration": 2.559912919998169 }, { "auxiliary_loss_clip": 0.01207117, "auxiliary_loss_mlp": 0.01033047, "balance_loss_clip": 1.06178176, "balance_loss_mlp": 1.02438092, "epoch": 0.20453315697709373, "flos": 16435671342720.0, "grad_norm": 1.9200222320977047, "language_loss": 0.84686494, "learning_rate": 3.6890230584241024e-06, "loss": 0.86926651, "num_input_tokens_seen": 36058825, "step": 1701, "time_per_iteration": 2.5324835777282715 }, { "auxiliary_loss_clip": 0.01109182, "auxiliary_loss_mlp": 0.01007405, "balance_loss_clip": 1.01799226, "balance_loss_mlp": 1.00576007, "epoch": 0.20465339986773282, "flos": 66713096896320.0, "grad_norm": 1.0659078722203001, "language_loss": 0.66494566, "learning_rate": 3.6886057618939016e-06, "loss": 0.68611151, "num_input_tokens_seen": 36121645, "step": 1702, "time_per_iteration": 3.194105625152588 }, { "auxiliary_loss_clip": 0.01163967, "auxiliary_loss_mlp": 0.01038772, "balance_loss_clip": 1.05427337, "balance_loss_mlp": 1.03006351, "epoch": 0.2047736427583719, "flos": 41974120294080.0, "grad_norm": 2.756412928040745, "language_loss": 0.6926651, "learning_rate": 3.6881882092035492e-06, "loss": 0.71469247, "num_input_tokens_seen": 36143030, "step": 1703, "time_per_iteration": 2.7662744522094727 }, { "auxiliary_loss_clip": 0.01081523, "auxiliary_loss_mlp": 0.00755264, "balance_loss_clip": 1.01757812, "balance_loss_mlp": 0.99966604, "epoch": 0.204893885649011, "flos": 69940888368480.0, "grad_norm": 0.9278789462650422, "language_loss": 0.61236966, "learning_rate": 3.6877704004163873e-06, "loss": 0.63073742, "num_input_tokens_seen": 36203435, "step": 1704, "time_per_iteration": 3.3252358436584473 }, { "auxiliary_loss_clip": 0.01219514, "auxiliary_loss_mlp": 0.01031795, "balance_loss_clip": 1.06356466, "balance_loss_mlp": 1.02315879, "epoch": 0.2050141285396501, "flos": 22200016490400.0, "grad_norm": 1.6520245152528399, "language_loss": 0.77810133, "learning_rate": 3.6873523355957984e-06, "loss": 0.80061448, "num_input_tokens_seen": 36222435, "step": 1705, "time_per_iteration": 4.053423643112183 }, { "auxiliary_loss_clip": 0.01108709, "auxiliary_loss_mlp": 0.01000859, "balance_loss_clip": 1.01761532, "balance_loss_mlp": 0.99924964, "epoch": 0.20513437143028918, "flos": 46283731313280.0, "grad_norm": 0.9824096964026501, "language_loss": 0.64142424, "learning_rate": 3.686934014805201e-06, "loss": 0.66251993, "num_input_tokens_seen": 36273065, "step": 1706, "time_per_iteration": 2.9567618370056152 }, { "auxiliary_loss_clip": 0.01201589, "auxiliary_loss_mlp": 0.01032924, "balance_loss_clip": 1.0621767, "balance_loss_mlp": 1.02443004, "epoch": 0.20525461432092829, "flos": 21904327348320.0, "grad_norm": 1.8179855203786175, "language_loss": 0.81239271, "learning_rate": 3.6865154381080552e-06, "loss": 0.83473784, "num_input_tokens_seen": 36293750, "step": 1707, "time_per_iteration": 2.626085042953491 }, { "auxiliary_loss_clip": 0.01127724, "auxiliary_loss_mlp": 0.01034324, "balance_loss_clip": 1.0509218, "balance_loss_mlp": 1.0257585, "epoch": 0.20537485721156737, "flos": 21214268423040.0, "grad_norm": 2.157219011999669, "language_loss": 0.82618761, "learning_rate": 3.6860966055678585e-06, "loss": 0.84780812, "num_input_tokens_seen": 36310105, "step": 1708, "time_per_iteration": 2.708197593688965 }, { "auxiliary_loss_clip": 0.01203293, "auxiliary_loss_mlp": 0.01032382, "balance_loss_clip": 1.06141615, "balance_loss_mlp": 1.0227735, "epoch": 0.20549510010220645, "flos": 20191209022560.0, "grad_norm": 1.7249996582055376, "language_loss": 0.86199272, "learning_rate": 3.685677517248147e-06, "loss": 0.88434947, "num_input_tokens_seen": 36328995, "step": 1709, "time_per_iteration": 2.594188690185547 }, { "auxiliary_loss_clip": 0.0118637, "auxiliary_loss_mlp": 0.00762957, "balance_loss_clip": 1.06228006, "balance_loss_mlp": 0.99995685, "epoch": 0.20561534299284553, "flos": 17016705542400.0, "grad_norm": 1.9193546520957947, "language_loss": 0.80288577, "learning_rate": 3.6852581732124967e-06, "loss": 0.82237905, "num_input_tokens_seen": 36346340, "step": 1710, "time_per_iteration": 3.4811620712280273 }, { "auxiliary_loss_clip": 0.01204381, "auxiliary_loss_mlp": 0.01036517, "balance_loss_clip": 1.06091678, "balance_loss_mlp": 1.02786803, "epoch": 0.20573558588348465, "flos": 22890470502240.0, "grad_norm": 1.8397855668796976, "language_loss": 0.76306731, "learning_rate": 3.6848385735245213e-06, "loss": 0.78547633, "num_input_tokens_seen": 36365430, "step": 1711, "time_per_iteration": 2.5776684284210205 }, { "auxiliary_loss_clip": 0.011893, "auxiliary_loss_mlp": 0.01032773, "balance_loss_clip": 1.0550611, "balance_loss_mlp": 1.02469015, "epoch": 0.20585582877412373, "flos": 24643127012640.0, "grad_norm": 1.8306726896800705, "language_loss": 0.86294675, "learning_rate": 3.6844187182478734e-06, "loss": 0.88516748, "num_input_tokens_seen": 36386285, "step": 1712, "time_per_iteration": 2.600839138031006 }, { "auxiliary_loss_clip": 0.01175531, "auxiliary_loss_mlp": 0.01030672, "balance_loss_clip": 1.05395269, "balance_loss_mlp": 1.02254176, "epoch": 0.2059760716647628, "flos": 24206956276320.0, "grad_norm": 3.56368251072286, "language_loss": 0.75177795, "learning_rate": 3.683998607446246e-06, "loss": 0.77383995, "num_input_tokens_seen": 36404935, "step": 1713, "time_per_iteration": 3.511029005050659 }, { "auxiliary_loss_clip": 0.01203927, "auxiliary_loss_mlp": 0.01028096, "balance_loss_clip": 1.06218171, "balance_loss_mlp": 1.0192802, "epoch": 0.20609631455540192, "flos": 20229526030560.0, "grad_norm": 2.0883130471906637, "language_loss": 0.75166404, "learning_rate": 3.6835782411833686e-06, "loss": 0.77398419, "num_input_tokens_seen": 36424455, "step": 1714, "time_per_iteration": 2.5708541870117188 }, { "auxiliary_loss_clip": 0.01158104, "auxiliary_loss_mlp": 0.01025509, "balance_loss_clip": 1.05316019, "balance_loss_mlp": 1.01718783, "epoch": 0.206216557446041, "flos": 19864960935840.0, "grad_norm": 1.6775959314801594, "language_loss": 0.74134791, "learning_rate": 3.68315761952301e-06, "loss": 0.76318407, "num_input_tokens_seen": 36441685, "step": 1715, "time_per_iteration": 2.604310989379883 }, { "auxiliary_loss_clip": 0.01221175, "auxiliary_loss_mlp": 0.01036367, "balance_loss_clip": 1.06387782, "balance_loss_mlp": 1.02821875, "epoch": 0.2063368003366801, "flos": 24096315287520.0, "grad_norm": 1.9406788133094022, "language_loss": 0.82781196, "learning_rate": 3.6827367425289797e-06, "loss": 0.85038733, "num_input_tokens_seen": 36461460, "step": 1716, "time_per_iteration": 2.5660903453826904 }, { "auxiliary_loss_clip": 0.01186833, "auxiliary_loss_mlp": 0.01029189, "balance_loss_clip": 1.05935538, "balance_loss_mlp": 1.0197413, "epoch": 0.2064570432273192, "flos": 20340167019360.0, "grad_norm": 2.2456960672765502, "language_loss": 0.72747684, "learning_rate": 3.6823156102651225e-06, "loss": 0.74963713, "num_input_tokens_seen": 36479615, "step": 1717, "time_per_iteration": 2.6019468307495117 }, { "auxiliary_loss_clip": 0.01132619, "auxiliary_loss_mlp": 0.01034976, "balance_loss_clip": 1.05397344, "balance_loss_mlp": 1.02642822, "epoch": 0.20657728611795828, "flos": 20520366383040.0, "grad_norm": 1.7910538367025164, "language_loss": 0.70934081, "learning_rate": 3.6818942227953257e-06, "loss": 0.7310167, "num_input_tokens_seen": 36500160, "step": 1718, "time_per_iteration": 2.6691551208496094 }, { "auxiliary_loss_clip": 0.01174374, "auxiliary_loss_mlp": 0.01027954, "balance_loss_clip": 1.05899024, "balance_loss_mlp": 1.0194838, "epoch": 0.20669752900859736, "flos": 21799289405280.0, "grad_norm": 1.8589742378728564, "language_loss": 0.69022119, "learning_rate": 3.681472580183512e-06, "loss": 0.71224451, "num_input_tokens_seen": 36518810, "step": 1719, "time_per_iteration": 2.6332108974456787 }, { "auxiliary_loss_clip": 0.01198925, "auxiliary_loss_mlp": 0.01033285, "balance_loss_clip": 1.06217504, "balance_loss_mlp": 1.02508378, "epoch": 0.20681777189923645, "flos": 15122023008480.0, "grad_norm": 10.547639886758079, "language_loss": 0.85962486, "learning_rate": 3.6810506824936455e-06, "loss": 0.88194692, "num_input_tokens_seen": 36536890, "step": 1720, "time_per_iteration": 2.534945249557495 }, { "auxiliary_loss_clip": 0.01081222, "auxiliary_loss_mlp": 0.01005132, "balance_loss_clip": 1.01317859, "balance_loss_mlp": 1.00351131, "epoch": 0.20693801478987556, "flos": 56481030296160.0, "grad_norm": 1.0344268839838298, "language_loss": 0.625875, "learning_rate": 3.680628529789726e-06, "loss": 0.64673853, "num_input_tokens_seen": 36589300, "step": 1721, "time_per_iteration": 3.0225026607513428 }, { "auxiliary_loss_clip": 0.01224298, "auxiliary_loss_mlp": 0.01034015, "balance_loss_clip": 1.06512237, "balance_loss_mlp": 1.02508044, "epoch": 0.20705825768051464, "flos": 21614205335040.0, "grad_norm": 1.7683851295710353, "language_loss": 0.86252779, "learning_rate": 3.680206122135796e-06, "loss": 0.88511091, "num_input_tokens_seen": 36609905, "step": 1722, "time_per_iteration": 2.5542423725128174 }, { "auxiliary_loss_clip": 0.01165531, "auxiliary_loss_mlp": 0.01035213, "balance_loss_clip": 1.05975771, "balance_loss_mlp": 1.02726793, "epoch": 0.20717850057115372, "flos": 25848899964000.0, "grad_norm": 2.048607328547552, "language_loss": 0.78033918, "learning_rate": 3.6797834595959323e-06, "loss": 0.80234665, "num_input_tokens_seen": 36629805, "step": 1723, "time_per_iteration": 2.7160286903381348 }, { "auxiliary_loss_clip": 0.01146962, "auxiliary_loss_mlp": 0.01033293, "balance_loss_clip": 1.0502429, "balance_loss_mlp": 1.02451956, "epoch": 0.20729874346179283, "flos": 29130811906560.0, "grad_norm": 2.492164194952611, "language_loss": 0.7785036, "learning_rate": 3.679360542234254e-06, "loss": 0.8003062, "num_input_tokens_seen": 36649150, "step": 1724, "time_per_iteration": 2.702908992767334 }, { "auxiliary_loss_clip": 0.01177867, "auxiliary_loss_mlp": 0.00763257, "balance_loss_clip": 1.05355775, "balance_loss_mlp": 0.99990582, "epoch": 0.20741898635243192, "flos": 29023367527200.0, "grad_norm": 1.7430268906749609, "language_loss": 0.72107804, "learning_rate": 3.678937370114916e-06, "loss": 0.74048924, "num_input_tokens_seen": 36668955, "step": 1725, "time_per_iteration": 2.6900393962860107 }, { "auxiliary_loss_clip": 0.01178344, "auxiliary_loss_mlp": 0.01034661, "balance_loss_clip": 1.05649948, "balance_loss_mlp": 1.02708578, "epoch": 0.207539229243071, "flos": 15559450838400.0, "grad_norm": 2.0468104903685065, "language_loss": 0.78883028, "learning_rate": 3.678513943302114e-06, "loss": 0.81096035, "num_input_tokens_seen": 36685730, "step": 1726, "time_per_iteration": 2.568601131439209 }, { "auxiliary_loss_clip": 0.01214941, "auxiliary_loss_mlp": 0.01030631, "balance_loss_clip": 1.06161106, "balance_loss_mlp": 1.02218485, "epoch": 0.20765947213371008, "flos": 20521084722240.0, "grad_norm": 1.7609365596665145, "language_loss": 0.85066915, "learning_rate": 3.678090261860082e-06, "loss": 0.87312484, "num_input_tokens_seen": 36705460, "step": 1727, "time_per_iteration": 2.556105852127075 }, { "auxiliary_loss_clip": 0.01172277, "auxiliary_loss_mlp": 0.01027721, "balance_loss_clip": 1.05260456, "balance_loss_mlp": 1.01928711, "epoch": 0.2077797150243492, "flos": 19354418952000.0, "grad_norm": 1.9004880389967838, "language_loss": 0.77469325, "learning_rate": 3.6776663258530906e-06, "loss": 0.79669327, "num_input_tokens_seen": 36724110, "step": 1728, "time_per_iteration": 2.6118743419647217 }, { "auxiliary_loss_clip": 0.01202512, "auxiliary_loss_mlp": 0.01032291, "balance_loss_clip": 1.05790591, "balance_loss_mlp": 1.02395236, "epoch": 0.20789995791498828, "flos": 21829956100800.0, "grad_norm": 3.1425443713120207, "language_loss": 0.70740175, "learning_rate": 3.6772421353454516e-06, "loss": 0.7297498, "num_input_tokens_seen": 36742705, "step": 1729, "time_per_iteration": 2.5918307304382324 }, { "auxiliary_loss_clip": 0.01201665, "auxiliary_loss_mlp": 0.01031453, "balance_loss_clip": 1.06210899, "balance_loss_mlp": 1.02266741, "epoch": 0.20802020080562736, "flos": 23148848311200.0, "grad_norm": 1.9213424356833395, "language_loss": 0.88510263, "learning_rate": 3.6768176904015153e-06, "loss": 0.90743387, "num_input_tokens_seen": 36762510, "step": 1730, "time_per_iteration": 3.3268191814422607 }, { "auxiliary_loss_clip": 0.012021, "auxiliary_loss_mlp": 0.01028621, "balance_loss_clip": 1.05742729, "balance_loss_mlp": 1.01969206, "epoch": 0.20814044369626647, "flos": 23072681215680.0, "grad_norm": 3.291652700316526, "language_loss": 0.59814674, "learning_rate": 3.6763929910856674e-06, "loss": 0.62045395, "num_input_tokens_seen": 36780960, "step": 1731, "time_per_iteration": 3.32710862159729 }, { "auxiliary_loss_clip": 0.01201419, "auxiliary_loss_mlp": 0.01034241, "balance_loss_clip": 1.06097364, "balance_loss_mlp": 1.02494276, "epoch": 0.20826068658690555, "flos": 19608019805280.0, "grad_norm": 2.217867055247413, "language_loss": 0.77664524, "learning_rate": 3.6759680374623365e-06, "loss": 0.79900187, "num_input_tokens_seen": 36798875, "step": 1732, "time_per_iteration": 2.5546698570251465 }, { "auxiliary_loss_clip": 0.01213122, "auxiliary_loss_mlp": 0.01031715, "balance_loss_clip": 1.06156039, "balance_loss_mlp": 1.02315545, "epoch": 0.20838092947754464, "flos": 25374053050080.0, "grad_norm": 2.2692397517289042, "language_loss": 0.75984669, "learning_rate": 3.675542829595986e-06, "loss": 0.78229511, "num_input_tokens_seen": 36818540, "step": 1733, "time_per_iteration": 2.5420432090759277 }, { "auxiliary_loss_clip": 0.01186059, "auxiliary_loss_mlp": 0.01029876, "balance_loss_clip": 1.05766273, "balance_loss_mlp": 1.02157855, "epoch": 0.20850117236818372, "flos": 24061733643360.0, "grad_norm": 1.3624906146297766, "language_loss": 0.7912755, "learning_rate": 3.6751173675511213e-06, "loss": 0.81343484, "num_input_tokens_seen": 36840585, "step": 1734, "time_per_iteration": 2.635770082473755 }, { "auxiliary_loss_clip": 0.01181598, "auxiliary_loss_mlp": 0.0102943, "balance_loss_clip": 1.05151308, "balance_loss_mlp": 1.02172327, "epoch": 0.20862141525882283, "flos": 20077802427840.0, "grad_norm": 2.0436177191701406, "language_loss": 0.87540525, "learning_rate": 3.674691651392283e-06, "loss": 0.89751554, "num_input_tokens_seen": 36858255, "step": 1735, "time_per_iteration": 2.588062286376953 }, { "auxiliary_loss_clip": 0.01192155, "auxiliary_loss_mlp": 0.01043516, "balance_loss_clip": 1.0602566, "balance_loss_mlp": 1.03420568, "epoch": 0.2087416581494619, "flos": 39015187994880.0, "grad_norm": 2.3191372682964855, "language_loss": 0.7580471, "learning_rate": 3.674265681184053e-06, "loss": 0.78040385, "num_input_tokens_seen": 36881515, "step": 1736, "time_per_iteration": 3.5979018211364746 }, { "auxiliary_loss_clip": 0.01186066, "auxiliary_loss_mlp": 0.0103183, "balance_loss_clip": 1.05648875, "balance_loss_mlp": 1.02296674, "epoch": 0.208861901040101, "flos": 26101997979840.0, "grad_norm": 1.6205199021715044, "language_loss": 0.86536622, "learning_rate": 3.6738394569910504e-06, "loss": 0.88754511, "num_input_tokens_seen": 36902055, "step": 1737, "time_per_iteration": 2.6679630279541016 }, { "auxiliary_loss_clip": 0.01199467, "auxiliary_loss_mlp": 0.01026039, "balance_loss_clip": 1.05908501, "balance_loss_mlp": 1.01735449, "epoch": 0.2089821439307401, "flos": 28398736526400.0, "grad_norm": 2.064114701760869, "language_loss": 0.82795715, "learning_rate": 3.6734129788779333e-06, "loss": 0.85021228, "num_input_tokens_seen": 36921230, "step": 1738, "time_per_iteration": 2.5997960567474365 }, { "auxiliary_loss_clip": 0.01169805, "auxiliary_loss_mlp": 0.0103708, "balance_loss_clip": 1.05742693, "balance_loss_mlp": 1.02928376, "epoch": 0.2091023868213792, "flos": 21069620461440.0, "grad_norm": 1.8725918747136516, "language_loss": 0.89961904, "learning_rate": 3.6729862469093976e-06, "loss": 0.92168784, "num_input_tokens_seen": 36940325, "step": 1739, "time_per_iteration": 3.542715549468994 }, { "auxiliary_loss_clip": 0.01172407, "auxiliary_loss_mlp": 0.01030394, "balance_loss_clip": 1.05524564, "balance_loss_mlp": 1.02219272, "epoch": 0.20922262971201827, "flos": 22455485025600.0, "grad_norm": 2.064120323438612, "language_loss": 0.82659942, "learning_rate": 3.6725592611501782e-06, "loss": 0.84862745, "num_input_tokens_seen": 36959000, "step": 1740, "time_per_iteration": 2.612793207168579 }, { "auxiliary_loss_clip": 0.01200035, "auxiliary_loss_mlp": 0.01031301, "balance_loss_clip": 1.05825591, "balance_loss_mlp": 1.02202654, "epoch": 0.20934287260265738, "flos": 27852248053920.0, "grad_norm": 1.8857594157634403, "language_loss": 0.76352876, "learning_rate": 3.6721320216650496e-06, "loss": 0.78584206, "num_input_tokens_seen": 36979615, "step": 1741, "time_per_iteration": 2.629528522491455 }, { "auxiliary_loss_clip": 0.01183068, "auxiliary_loss_mlp": 0.01026809, "balance_loss_clip": 1.0580256, "balance_loss_mlp": 1.01832175, "epoch": 0.20946311549329646, "flos": 16435312173120.0, "grad_norm": 1.6524086032649477, "language_loss": 0.83136725, "learning_rate": 3.6717045285188215e-06, "loss": 0.85346609, "num_input_tokens_seen": 36997310, "step": 1742, "time_per_iteration": 2.5692644119262695 }, { "auxiliary_loss_clip": 0.01141785, "auxiliary_loss_mlp": 0.01026086, "balance_loss_clip": 1.04925334, "balance_loss_mlp": 1.01703238, "epoch": 0.20958335838393555, "flos": 22492724524800.0, "grad_norm": 2.2573316304508326, "language_loss": 0.86771196, "learning_rate": 3.671276781776346e-06, "loss": 0.88939065, "num_input_tokens_seen": 37015965, "step": 1743, "time_per_iteration": 2.659499406814575 }, { "auxiliary_loss_clip": 0.01178451, "auxiliary_loss_mlp": 0.01030557, "balance_loss_clip": 1.05315685, "balance_loss_mlp": 1.02201009, "epoch": 0.20970360127457463, "flos": 25224771800640.0, "grad_norm": 2.0682755402731527, "language_loss": 0.67363715, "learning_rate": 3.6708487815025128e-06, "loss": 0.69572717, "num_input_tokens_seen": 37036545, "step": 1744, "time_per_iteration": 2.64369535446167 }, { "auxiliary_loss_clip": 0.01170796, "auxiliary_loss_mlp": 0.01026413, "balance_loss_clip": 1.05630326, "balance_loss_mlp": 1.01763344, "epoch": 0.20982384416521374, "flos": 18479168205600.0, "grad_norm": 2.3301745494945134, "language_loss": 0.74866092, "learning_rate": 3.6704205277622463e-06, "loss": 0.77063298, "num_input_tokens_seen": 37054985, "step": 1745, "time_per_iteration": 2.6175649166107178 }, { "auxiliary_loss_clip": 0.01189617, "auxiliary_loss_mlp": 0.01028921, "balance_loss_clip": 1.05597508, "balance_loss_mlp": 1.0203259, "epoch": 0.20994408705585282, "flos": 25373550212640.0, "grad_norm": 2.135027471477059, "language_loss": 0.80593669, "learning_rate": 3.6699920206205146e-06, "loss": 0.82812208, "num_input_tokens_seen": 37075725, "step": 1746, "time_per_iteration": 2.6228697299957275 }, { "auxiliary_loss_clip": 0.01202714, "auxiliary_loss_mlp": 0.01030322, "balance_loss_clip": 1.05820453, "balance_loss_mlp": 1.02201867, "epoch": 0.2100643299464919, "flos": 21320958546240.0, "grad_norm": 1.704255748425031, "language_loss": 0.82356334, "learning_rate": 3.669563260142321e-06, "loss": 0.84589374, "num_input_tokens_seen": 37094615, "step": 1747, "time_per_iteration": 2.586181640625 }, { "auxiliary_loss_clip": 0.01183068, "auxiliary_loss_mlp": 0.01026883, "balance_loss_clip": 1.05880833, "balance_loss_mlp": 1.01751304, "epoch": 0.21018457283713102, "flos": 19354383035040.0, "grad_norm": 2.2087600332656407, "language_loss": 0.84266174, "learning_rate": 3.6691342463927083e-06, "loss": 0.86476123, "num_input_tokens_seen": 37113610, "step": 1748, "time_per_iteration": 2.576345443725586 }, { "auxiliary_loss_clip": 0.01175787, "auxiliary_loss_mlp": 0.01046163, "balance_loss_clip": 1.0571425, "balance_loss_mlp": 1.03691852, "epoch": 0.2103048157277701, "flos": 28330040158560.0, "grad_norm": 1.612096386612615, "language_loss": 0.81946838, "learning_rate": 3.668704979436758e-06, "loss": 0.84168786, "num_input_tokens_seen": 37133705, "step": 1749, "time_per_iteration": 2.6734323501586914 }, { "auxiliary_loss_clip": 0.01180134, "auxiliary_loss_mlp": 0.01037906, "balance_loss_clip": 1.05411124, "balance_loss_mlp": 1.02871513, "epoch": 0.21042505861840918, "flos": 17457294064800.0, "grad_norm": 1.8730616147489394, "language_loss": 0.78612328, "learning_rate": 3.668275459339588e-06, "loss": 0.80830371, "num_input_tokens_seen": 37152185, "step": 1750, "time_per_iteration": 2.552128553390503 }, { "auxiliary_loss_clip": 0.01216708, "auxiliary_loss_mlp": 0.01031695, "balance_loss_clip": 1.06230116, "balance_loss_mlp": 1.02315927, "epoch": 0.21054530150904827, "flos": 14209820098560.0, "grad_norm": 1.8818781900266657, "language_loss": 0.79944676, "learning_rate": 3.667845686166358e-06, "loss": 0.82193077, "num_input_tokens_seen": 37169110, "step": 1751, "time_per_iteration": 2.4612677097320557 }, { "auxiliary_loss_clip": 0.01153956, "auxiliary_loss_mlp": 0.01028455, "balance_loss_clip": 1.05299711, "balance_loss_mlp": 1.01957989, "epoch": 0.21066554439968738, "flos": 18618213121440.0, "grad_norm": 1.9547112013183967, "language_loss": 0.85792243, "learning_rate": 3.6674156599822634e-06, "loss": 0.87974656, "num_input_tokens_seen": 37184905, "step": 1752, "time_per_iteration": 2.6019115447998047 }, { "auxiliary_loss_clip": 0.01161062, "auxiliary_loss_mlp": 0.01034348, "balance_loss_clip": 1.05208158, "balance_loss_mlp": 1.0252111, "epoch": 0.21078578729032646, "flos": 23658887457600.0, "grad_norm": 1.9191215895227347, "language_loss": 0.81619525, "learning_rate": 3.666985380852539e-06, "loss": 0.83814943, "num_input_tokens_seen": 37203910, "step": 1753, "time_per_iteration": 2.637267827987671 }, { "auxiliary_loss_clip": 0.01185568, "auxiliary_loss_mlp": 0.01036937, "balance_loss_clip": 1.05825555, "balance_loss_mlp": 1.02826428, "epoch": 0.21090603018096554, "flos": 29346383087520.0, "grad_norm": 2.2635507000097816, "language_loss": 0.73801208, "learning_rate": 3.6665548488424576e-06, "loss": 0.7602371, "num_input_tokens_seen": 37222670, "step": 1754, "time_per_iteration": 2.6563775539398193 }, { "auxiliary_loss_clip": 0.01219595, "auxiliary_loss_mlp": 0.01035672, "balance_loss_clip": 1.06354654, "balance_loss_mlp": 1.02606964, "epoch": 0.21102627307160465, "flos": 23261249231040.0, "grad_norm": 1.5913288659683762, "language_loss": 0.87771714, "learning_rate": 3.6661240640173307e-06, "loss": 0.90026975, "num_input_tokens_seen": 37244140, "step": 1755, "time_per_iteration": 2.558943033218384 }, { "auxiliary_loss_clip": 0.01074135, "auxiliary_loss_mlp": 0.01001773, "balance_loss_clip": 1.01754141, "balance_loss_mlp": 0.9999612, "epoch": 0.21114651596224374, "flos": 54633464425440.0, "grad_norm": 0.8472611793258711, "language_loss": 0.57933033, "learning_rate": 3.6656930264425085e-06, "loss": 0.60008943, "num_input_tokens_seen": 37308185, "step": 1756, "time_per_iteration": 3.950410842895508 }, { "auxiliary_loss_clip": 0.01218541, "auxiliary_loss_mlp": 0.01030729, "balance_loss_clip": 1.06343246, "balance_loss_mlp": 1.02169907, "epoch": 0.21126675885288282, "flos": 21543318032640.0, "grad_norm": 1.834527443481418, "language_loss": 0.75442719, "learning_rate": 3.665261736183378e-06, "loss": 0.7769199, "num_input_tokens_seen": 37328220, "step": 1757, "time_per_iteration": 3.310379981994629 }, { "auxiliary_loss_clip": 0.01170787, "auxiliary_loss_mlp": 0.0102913, "balance_loss_clip": 1.05610526, "balance_loss_mlp": 1.02009439, "epoch": 0.2113870017435219, "flos": 10961879211840.0, "grad_norm": 3.914008003535937, "language_loss": 0.8914113, "learning_rate": 3.664830193305366e-06, "loss": 0.91341048, "num_input_tokens_seen": 37345995, "step": 1758, "time_per_iteration": 2.586718797683716 }, { "auxiliary_loss_clip": 0.01167644, "auxiliary_loss_mlp": 0.01037503, "balance_loss_clip": 1.05365634, "balance_loss_mlp": 1.02843761, "epoch": 0.211507244634161, "flos": 16653828544800.0, "grad_norm": 3.218842642618694, "language_loss": 0.76912546, "learning_rate": 3.6643983978739373e-06, "loss": 0.79117703, "num_input_tokens_seen": 37362610, "step": 1759, "time_per_iteration": 2.608234405517578 }, { "auxiliary_loss_clip": 0.01179059, "auxiliary_loss_mlp": 0.01036533, "balance_loss_clip": 1.05814028, "balance_loss_mlp": 1.02793789, "epoch": 0.2116274875248001, "flos": 20954094766080.0, "grad_norm": 1.834907279998123, "language_loss": 0.81974435, "learning_rate": 3.663966349954596e-06, "loss": 0.84190023, "num_input_tokens_seen": 37382790, "step": 1760, "time_per_iteration": 2.6225943565368652 }, { "auxiliary_loss_clip": 0.0109448, "auxiliary_loss_mlp": 0.01001253, "balance_loss_clip": 1.01423669, "balance_loss_mlp": 0.99946469, "epoch": 0.21174773041543918, "flos": 68196959679360.0, "grad_norm": 0.7830306859206431, "language_loss": 0.59705234, "learning_rate": 3.6635340496128816e-06, "loss": 0.61800969, "num_input_tokens_seen": 37439720, "step": 1761, "time_per_iteration": 3.084758996963501 }, { "auxiliary_loss_clip": 0.01151281, "auxiliary_loss_mlp": 0.01031931, "balance_loss_clip": 1.05264556, "balance_loss_mlp": 1.02369976, "epoch": 0.2118679733060783, "flos": 20668318704960.0, "grad_norm": 1.854175705265015, "language_loss": 0.92702436, "learning_rate": 3.6631014969143747e-06, "loss": 0.94885647, "num_input_tokens_seen": 37459410, "step": 1762, "time_per_iteration": 3.520620822906494 }, { "auxiliary_loss_clip": 0.01204308, "auxiliary_loss_mlp": 0.01035903, "balance_loss_clip": 1.06218338, "balance_loss_mlp": 1.02720714, "epoch": 0.21198821619671737, "flos": 23223435060480.0, "grad_norm": 1.8257328621310618, "language_loss": 0.88669097, "learning_rate": 3.662668691924693e-06, "loss": 0.90909314, "num_input_tokens_seen": 37480460, "step": 1763, "time_per_iteration": 2.5992960929870605 }, { "auxiliary_loss_clip": 0.0116978, "auxiliary_loss_mlp": 0.01031994, "balance_loss_clip": 1.05341935, "balance_loss_mlp": 1.02304101, "epoch": 0.21210845908735645, "flos": 24498550884960.0, "grad_norm": 2.242713838513768, "language_loss": 0.71360487, "learning_rate": 3.6622356347094927e-06, "loss": 0.73562258, "num_input_tokens_seen": 37502025, "step": 1764, "time_per_iteration": 2.6751668453216553 }, { "auxiliary_loss_clip": 0.01174302, "auxiliary_loss_mlp": 0.01036675, "balance_loss_clip": 1.05592692, "balance_loss_mlp": 1.02681684, "epoch": 0.21222870197799554, "flos": 27089793313920.0, "grad_norm": 1.8854060151142085, "language_loss": 0.78776717, "learning_rate": 3.6618023253344684e-06, "loss": 0.80987692, "num_input_tokens_seen": 37520885, "step": 1765, "time_per_iteration": 3.5389461517333984 }, { "auxiliary_loss_clip": 0.01204414, "auxiliary_loss_mlp": 0.01041246, "balance_loss_clip": 1.05847383, "balance_loss_mlp": 1.03160214, "epoch": 0.21234894486863465, "flos": 16873853428800.0, "grad_norm": 1.4707721599348704, "language_loss": 0.83406109, "learning_rate": 3.6613687638653527e-06, "loss": 0.85651767, "num_input_tokens_seen": 37539055, "step": 1766, "time_per_iteration": 2.572326898574829 }, { "auxiliary_loss_clip": 0.0118542, "auxiliary_loss_mlp": 0.01029612, "balance_loss_clip": 1.06036782, "balance_loss_mlp": 1.0208447, "epoch": 0.21246918775927373, "flos": 23474952730080.0, "grad_norm": 1.684412616987458, "language_loss": 0.77868658, "learning_rate": 3.660934950367916e-06, "loss": 0.80083692, "num_input_tokens_seen": 37558300, "step": 1767, "time_per_iteration": 2.603872299194336 }, { "auxiliary_loss_clip": 0.01205951, "auxiliary_loss_mlp": 0.01034046, "balance_loss_clip": 1.05968881, "balance_loss_mlp": 1.02571332, "epoch": 0.21258943064991281, "flos": 22382299037760.0, "grad_norm": 2.7987110225426295, "language_loss": 0.83313453, "learning_rate": 3.660500884907968e-06, "loss": 0.85553455, "num_input_tokens_seen": 37579040, "step": 1768, "time_per_iteration": 2.588765859603882 }, { "auxiliary_loss_clip": 0.01062298, "auxiliary_loss_mlp": 0.01010095, "balance_loss_clip": 1.01699317, "balance_loss_mlp": 1.00844979, "epoch": 0.21270967354055192, "flos": 59440178097120.0, "grad_norm": 0.8218638733599346, "language_loss": 0.60132474, "learning_rate": 3.660066567551356e-06, "loss": 0.62204868, "num_input_tokens_seen": 37639185, "step": 1769, "time_per_iteration": 3.13335919380188 }, { "auxiliary_loss_clip": 0.01202178, "auxiliary_loss_mlp": 0.00764117, "balance_loss_clip": 1.05851805, "balance_loss_mlp": 0.9999727, "epoch": 0.212829916431191, "flos": 21544036371840.0, "grad_norm": 2.2244436360154545, "language_loss": 0.83989066, "learning_rate": 3.6596319983639657e-06, "loss": 0.85955358, "num_input_tokens_seen": 37657765, "step": 1770, "time_per_iteration": 2.5856354236602783 }, { "auxiliary_loss_clip": 0.01173893, "auxiliary_loss_mlp": 0.00764595, "balance_loss_clip": 1.0581944, "balance_loss_mlp": 0.99996459, "epoch": 0.2129501593218301, "flos": 28987744291200.0, "grad_norm": 1.695894153193752, "language_loss": 0.8603605, "learning_rate": 3.6591971774117214e-06, "loss": 0.87974536, "num_input_tokens_seen": 37680740, "step": 1771, "time_per_iteration": 2.681140184402466 }, { "auxiliary_loss_clip": 0.01209915, "auxiliary_loss_mlp": 0.01034464, "balance_loss_clip": 1.06412554, "balance_loss_mlp": 1.02552295, "epoch": 0.2130704022124692, "flos": 18806996638560.0, "grad_norm": 2.23587659762024, "language_loss": 0.80023623, "learning_rate": 3.6587621047605833e-06, "loss": 0.82268, "num_input_tokens_seen": 37697910, "step": 1772, "time_per_iteration": 2.5704169273376465 }, { "auxiliary_loss_clip": 0.0120305, "auxiliary_loss_mlp": 0.01034292, "balance_loss_clip": 1.06132317, "balance_loss_mlp": 1.02511311, "epoch": 0.21319064510310828, "flos": 13918153656000.0, "grad_norm": 1.8494610977651418, "language_loss": 0.86657286, "learning_rate": 3.6583267804765542e-06, "loss": 0.88894629, "num_input_tokens_seen": 37712245, "step": 1773, "time_per_iteration": 2.5337398052215576 }, { "auxiliary_loss_clip": 0.01201989, "auxiliary_loss_mlp": 0.01032521, "balance_loss_clip": 1.06135178, "balance_loss_mlp": 1.02316272, "epoch": 0.21331088799374737, "flos": 20959697811840.0, "grad_norm": 3.7652376647265355, "language_loss": 0.85577178, "learning_rate": 3.6578912046256702e-06, "loss": 0.87811691, "num_input_tokens_seen": 37730765, "step": 1774, "time_per_iteration": 2.563141345977783 }, { "auxiliary_loss_clip": 0.01167853, "auxiliary_loss_mlp": 0.01034621, "balance_loss_clip": 1.05541515, "balance_loss_mlp": 1.02537632, "epoch": 0.21343113088438645, "flos": 18624642257280.0, "grad_norm": 2.3622751932029886, "language_loss": 0.76253963, "learning_rate": 3.6574553772740083e-06, "loss": 0.78456438, "num_input_tokens_seen": 37748695, "step": 1775, "time_per_iteration": 2.6218063831329346 }, { "auxiliary_loss_clip": 0.01102192, "auxiliary_loss_mlp": 0.0101972, "balance_loss_clip": 1.03108215, "balance_loss_mlp": 1.01802719, "epoch": 0.21355137377502556, "flos": 67413140736480.0, "grad_norm": 0.8605716933586092, "language_loss": 0.6190033, "learning_rate": 3.657019298487684e-06, "loss": 0.64022243, "num_input_tokens_seen": 37813705, "step": 1776, "time_per_iteration": 3.1733198165893555 }, { "auxiliary_loss_clip": 0.01210011, "auxiliary_loss_mlp": 0.00764474, "balance_loss_clip": 1.06202292, "balance_loss_mlp": 0.99997067, "epoch": 0.21367161666566464, "flos": 34532100471840.0, "grad_norm": 1.6042968958853199, "language_loss": 0.83136624, "learning_rate": 3.6565829683328495e-06, "loss": 0.85111105, "num_input_tokens_seen": 37836330, "step": 1777, "time_per_iteration": 2.6903929710388184 }, { "auxiliary_loss_clip": 0.0119734, "auxiliary_loss_mlp": 0.01033076, "balance_loss_clip": 1.05928624, "balance_loss_mlp": 1.02426052, "epoch": 0.21379185955630373, "flos": 18989351019840.0, "grad_norm": 1.7769514660993198, "language_loss": 0.85606945, "learning_rate": 3.6561463868756965e-06, "loss": 0.87837362, "num_input_tokens_seen": 37855030, "step": 1778, "time_per_iteration": 2.5596728324890137 }, { "auxiliary_loss_clip": 0.01204301, "auxiliary_loss_mlp": 0.0103101, "balance_loss_clip": 1.0624299, "balance_loss_mlp": 1.02163398, "epoch": 0.21391210244694284, "flos": 28218501245760.0, "grad_norm": 1.7737872439996105, "language_loss": 0.77857041, "learning_rate": 3.655709554182452e-06, "loss": 0.80092353, "num_input_tokens_seen": 37875370, "step": 1779, "time_per_iteration": 2.6388742923736572 }, { "auxiliary_loss_clip": 0.01206632, "auxiliary_loss_mlp": 0.01035957, "balance_loss_clip": 1.05996799, "balance_loss_mlp": 1.02661681, "epoch": 0.21403234533758192, "flos": 17455067213280.0, "grad_norm": 1.6346026502632973, "language_loss": 0.84389198, "learning_rate": 3.6552724703193855e-06, "loss": 0.86631787, "num_input_tokens_seen": 37892560, "step": 1780, "time_per_iteration": 2.5375454425811768 }, { "auxiliary_loss_clip": 0.01060024, "auxiliary_loss_mlp": 0.01002632, "balance_loss_clip": 1.01749527, "balance_loss_mlp": 1.00099897, "epoch": 0.214152588228221, "flos": 51637615879680.0, "grad_norm": 0.7885703383255063, "language_loss": 0.56007659, "learning_rate": 3.654835135352801e-06, "loss": 0.58070314, "num_input_tokens_seen": 37947370, "step": 1781, "time_per_iteration": 3.1176111698150635 }, { "auxiliary_loss_clip": 0.01156794, "auxiliary_loss_mlp": 0.01038563, "balance_loss_clip": 1.05154228, "balance_loss_mlp": 1.02956319, "epoch": 0.21427283111886009, "flos": 19496157639840.0, "grad_norm": 2.08439292790172, "language_loss": 0.87698495, "learning_rate": 3.654397549349043e-06, "loss": 0.89893854, "num_input_tokens_seen": 37964745, "step": 1782, "time_per_iteration": 2.68154239654541 }, { "auxiliary_loss_clip": 0.01186406, "auxiliary_loss_mlp": 0.01040626, "balance_loss_clip": 1.06125355, "balance_loss_mlp": 1.03154802, "epoch": 0.2143930740094992, "flos": 20084806235040.0, "grad_norm": 1.9701597370635748, "language_loss": 0.75284135, "learning_rate": 3.653959712374491e-06, "loss": 0.77511168, "num_input_tokens_seen": 37982850, "step": 1783, "time_per_iteration": 4.102065086364746 }, { "auxiliary_loss_clip": 0.01170085, "auxiliary_loss_mlp": 0.01029078, "balance_loss_clip": 1.05935168, "balance_loss_mlp": 1.01952934, "epoch": 0.21451331690013828, "flos": 21798606983040.0, "grad_norm": 2.0170302516331886, "language_loss": 0.82495832, "learning_rate": 3.6535216244955663e-06, "loss": 0.84694999, "num_input_tokens_seen": 38002745, "step": 1784, "time_per_iteration": 2.629697799682617 }, { "auxiliary_loss_clip": 0.01189544, "auxiliary_loss_mlp": 0.01030922, "balance_loss_clip": 1.06052613, "balance_loss_mlp": 1.02149248, "epoch": 0.21463355979077736, "flos": 32853887042880.0, "grad_norm": 1.577363143035923, "language_loss": 0.7065044, "learning_rate": 3.653083285778726e-06, "loss": 0.72870904, "num_input_tokens_seen": 38024115, "step": 1785, "time_per_iteration": 2.691041946411133 }, { "auxiliary_loss_clip": 0.01207663, "auxiliary_loss_mlp": 0.01032729, "balance_loss_clip": 1.06014228, "balance_loss_mlp": 1.02366316, "epoch": 0.21475380268141647, "flos": 21543820870080.0, "grad_norm": 2.1533380049664848, "language_loss": 0.81236315, "learning_rate": 3.6526446962904653e-06, "loss": 0.83476704, "num_input_tokens_seen": 38042830, "step": 1786, "time_per_iteration": 2.5880391597747803 }, { "auxiliary_loss_clip": 0.01200693, "auxiliary_loss_mlp": 0.0103022, "balance_loss_clip": 1.06272483, "balance_loss_mlp": 1.02104127, "epoch": 0.21487404557205556, "flos": 32159087078880.0, "grad_norm": 1.3670890453517681, "language_loss": 0.7410506, "learning_rate": 3.652205856097318e-06, "loss": 0.76335979, "num_input_tokens_seen": 38066015, "step": 1787, "time_per_iteration": 2.644348621368408 }, { "auxiliary_loss_clip": 0.01189316, "auxiliary_loss_mlp": 0.00764256, "balance_loss_clip": 1.06004107, "balance_loss_mlp": 1.00003076, "epoch": 0.21499428846269464, "flos": 12673093938720.0, "grad_norm": 1.9785676573847324, "language_loss": 0.79308033, "learning_rate": 3.651766765265856e-06, "loss": 0.81261599, "num_input_tokens_seen": 38083025, "step": 1788, "time_per_iteration": 3.4926769733428955 }, { "auxiliary_loss_clip": 0.01183795, "auxiliary_loss_mlp": 0.01036503, "balance_loss_clip": 1.05738902, "balance_loss_mlp": 1.02800941, "epoch": 0.21511453135333372, "flos": 23471576535840.0, "grad_norm": 2.779999745046395, "language_loss": 0.80983096, "learning_rate": 3.65132742386269e-06, "loss": 0.83203393, "num_input_tokens_seen": 38098245, "step": 1789, "time_per_iteration": 2.604800224304199 }, { "auxiliary_loss_clip": 0.01220517, "auxiliary_loss_mlp": 0.01039141, "balance_loss_clip": 1.06277156, "balance_loss_mlp": 1.02950263, "epoch": 0.21523477424397283, "flos": 26943565006080.0, "grad_norm": 1.8388530182422547, "language_loss": 0.84872544, "learning_rate": 3.6508878319544656e-06, "loss": 0.87132204, "num_input_tokens_seen": 38118460, "step": 1790, "time_per_iteration": 2.574644088745117 }, { "auxiliary_loss_clip": 0.01180536, "auxiliary_loss_mlp": 0.01038754, "balance_loss_clip": 1.06003928, "balance_loss_mlp": 1.0299325, "epoch": 0.21535501713461191, "flos": 18916165032000.0, "grad_norm": 2.525713851720579, "language_loss": 0.81143129, "learning_rate": 3.65044798960787e-06, "loss": 0.83362424, "num_input_tokens_seen": 38136800, "step": 1791, "time_per_iteration": 3.4824161529541016 }, { "auxiliary_loss_clip": 0.01167234, "auxiliary_loss_mlp": 0.01036229, "balance_loss_clip": 1.05518496, "balance_loss_mlp": 1.02721667, "epoch": 0.215475260025251, "flos": 17895116981280.0, "grad_norm": 1.9391839081354392, "language_loss": 0.78294694, "learning_rate": 3.650007896889627e-06, "loss": 0.80498153, "num_input_tokens_seen": 38155380, "step": 1792, "time_per_iteration": 2.6069695949554443 }, { "auxiliary_loss_clip": 0.01218042, "auxiliary_loss_mlp": 0.01036118, "balance_loss_clip": 1.06338763, "balance_loss_mlp": 1.02598524, "epoch": 0.2155955029158901, "flos": 16654295465280.0, "grad_norm": 1.7475638327912992, "language_loss": 0.80784452, "learning_rate": 3.6495675538664974e-06, "loss": 0.8303861, "num_input_tokens_seen": 38174395, "step": 1793, "time_per_iteration": 2.5362017154693604 }, { "auxiliary_loss_clip": 0.01188377, "auxiliary_loss_mlp": 0.01034816, "balance_loss_clip": 1.05669439, "balance_loss_mlp": 1.02546406, "epoch": 0.2157157458065292, "flos": 23621216954880.0, "grad_norm": 2.5970360471297185, "language_loss": 0.82421684, "learning_rate": 3.649126960605282e-06, "loss": 0.84644878, "num_input_tokens_seen": 38195380, "step": 1794, "time_per_iteration": 2.6233859062194824 }, { "auxiliary_loss_clip": 0.01183312, "auxiliary_loss_mlp": 0.01031874, "balance_loss_clip": 1.0589757, "balance_loss_mlp": 1.02275503, "epoch": 0.21583598869716827, "flos": 22127081921280.0, "grad_norm": 4.354212912240218, "language_loss": 0.83644271, "learning_rate": 3.6486861171728174e-06, "loss": 0.85859454, "num_input_tokens_seen": 38213775, "step": 1795, "time_per_iteration": 2.593940496444702 }, { "auxiliary_loss_clip": 0.01173388, "auxiliary_loss_mlp": 0.01035884, "balance_loss_clip": 1.05301023, "balance_loss_mlp": 1.02626419, "epoch": 0.21595623158780738, "flos": 23441233092960.0, "grad_norm": 1.6265923128603832, "language_loss": 0.78825128, "learning_rate": 3.6482450236359803e-06, "loss": 0.81034398, "num_input_tokens_seen": 38235630, "step": 1796, "time_per_iteration": 2.660846471786499 }, { "auxiliary_loss_clip": 0.01201845, "auxiliary_loss_mlp": 0.01035501, "balance_loss_clip": 1.06140447, "balance_loss_mlp": 1.02632189, "epoch": 0.21607647447844647, "flos": 26906505091680.0, "grad_norm": 2.2438524106036453, "language_loss": 0.77294213, "learning_rate": 3.647803680061683e-06, "loss": 0.79531556, "num_input_tokens_seen": 38256045, "step": 1797, "time_per_iteration": 2.6081948280334473 }, { "auxiliary_loss_clip": 0.01191006, "auxiliary_loss_mlp": 0.01030282, "balance_loss_clip": 1.06039357, "balance_loss_mlp": 1.02082264, "epoch": 0.21619671736908555, "flos": 14495380657920.0, "grad_norm": 2.8154975425796525, "language_loss": 0.75031829, "learning_rate": 3.6473620865168776e-06, "loss": 0.77253115, "num_input_tokens_seen": 38272915, "step": 1798, "time_per_iteration": 2.5751256942749023 }, { "auxiliary_loss_clip": 0.01191693, "auxiliary_loss_mlp": 0.01035324, "balance_loss_clip": 1.06187463, "balance_loss_mlp": 1.02610898, "epoch": 0.21631696025972463, "flos": 17931099386880.0, "grad_norm": 1.8560085000223019, "language_loss": 0.81660032, "learning_rate": 3.646920243068554e-06, "loss": 0.83887053, "num_input_tokens_seen": 38290810, "step": 1799, "time_per_iteration": 2.5826730728149414 }, { "auxiliary_loss_clip": 0.01176413, "auxiliary_loss_mlp": 0.01029177, "balance_loss_clip": 1.05783427, "balance_loss_mlp": 1.0193665, "epoch": 0.21643720315036374, "flos": 24462388894560.0, "grad_norm": 1.5320189985667176, "language_loss": 0.74775666, "learning_rate": 3.6464781497837384e-06, "loss": 0.76981258, "num_input_tokens_seen": 38312785, "step": 1800, "time_per_iteration": 2.618666887283325 }, { "auxiliary_loss_clip": 0.01194362, "auxiliary_loss_mlp": 0.01037315, "balance_loss_clip": 1.05710125, "balance_loss_mlp": 1.02774882, "epoch": 0.21655744604100283, "flos": 28474436701440.0, "grad_norm": 1.5741183188654555, "language_loss": 0.72791374, "learning_rate": 3.6460358067294965e-06, "loss": 0.75023055, "num_input_tokens_seen": 38334015, "step": 1801, "time_per_iteration": 2.6538007259368896 }, { "auxiliary_loss_clip": 0.01222053, "auxiliary_loss_mlp": 0.01030212, "balance_loss_clip": 1.06255579, "balance_loss_mlp": 1.02003157, "epoch": 0.2166776889316419, "flos": 20152963848480.0, "grad_norm": 2.0808097370883845, "language_loss": 0.77661467, "learning_rate": 3.645593213972932e-06, "loss": 0.79913729, "num_input_tokens_seen": 38352920, "step": 1802, "time_per_iteration": 2.5572361946105957 }, { "auxiliary_loss_clip": 0.0119629, "auxiliary_loss_mlp": 0.01034397, "balance_loss_clip": 1.05855489, "balance_loss_mlp": 1.02498484, "epoch": 0.21679793182228102, "flos": 15193484982240.0, "grad_norm": 1.9876144822984612, "language_loss": 0.797279, "learning_rate": 3.6451503715811852e-06, "loss": 0.81958586, "num_input_tokens_seen": 38371230, "step": 1803, "time_per_iteration": 2.5343177318573 }, { "auxiliary_loss_clip": 0.01185638, "auxiliary_loss_mlp": 0.01031405, "balance_loss_clip": 1.05983114, "balance_loss_mlp": 1.02295327, "epoch": 0.2169181747129201, "flos": 17384467246560.0, "grad_norm": 1.9943017490678223, "language_loss": 0.79776776, "learning_rate": 3.6447072796214345e-06, "loss": 0.81993818, "num_input_tokens_seen": 38389795, "step": 1804, "time_per_iteration": 2.5869481563568115 }, { "auxiliary_loss_clip": 0.01058965, "auxiliary_loss_mlp": 0.0100042, "balance_loss_clip": 1.01777387, "balance_loss_mlp": 0.99859583, "epoch": 0.21703841760355919, "flos": 58760966093760.0, "grad_norm": 0.9186654214192223, "language_loss": 0.63229603, "learning_rate": 3.644263938160898e-06, "loss": 0.65288991, "num_input_tokens_seen": 38445760, "step": 1805, "time_per_iteration": 3.1079304218292236 }, { "auxiliary_loss_clip": 0.01172426, "auxiliary_loss_mlp": 0.01036833, "balance_loss_clip": 1.05676818, "balance_loss_mlp": 1.02755845, "epoch": 0.21715866049419827, "flos": 22418461028160.0, "grad_norm": 1.8866487243010022, "language_loss": 0.7197032, "learning_rate": 3.6438203472668293e-06, "loss": 0.74179584, "num_input_tokens_seen": 38465405, "step": 1806, "time_per_iteration": 2.645799398422241 }, { "auxiliary_loss_clip": 0.01192219, "auxiliary_loss_mlp": 0.01036108, "balance_loss_clip": 1.06058812, "balance_loss_mlp": 1.02721477, "epoch": 0.21727890338483738, "flos": 17237736101280.0, "grad_norm": 1.938783695961625, "language_loss": 0.81625021, "learning_rate": 3.6433765070065206e-06, "loss": 0.83853346, "num_input_tokens_seen": 38483195, "step": 1807, "time_per_iteration": 2.588085651397705 }, { "auxiliary_loss_clip": 0.01219121, "auxiliary_loss_mlp": 0.0103544, "balance_loss_clip": 1.06297517, "balance_loss_mlp": 1.02615952, "epoch": 0.21739914627547646, "flos": 13434794422560.0, "grad_norm": 2.312249803629242, "language_loss": 0.87384284, "learning_rate": 3.6429324174473025e-06, "loss": 0.89638841, "num_input_tokens_seen": 38496735, "step": 1808, "time_per_iteration": 3.2602553367614746 }, { "auxiliary_loss_clip": 0.01205872, "auxiliary_loss_mlp": 0.01031833, "balance_loss_clip": 1.06146681, "balance_loss_mlp": 1.02317834, "epoch": 0.21751938916611555, "flos": 20959518227040.0, "grad_norm": 4.4570500317604305, "language_loss": 0.84949279, "learning_rate": 3.6424880786565425e-06, "loss": 0.8718698, "num_input_tokens_seen": 38512880, "step": 1809, "time_per_iteration": 2.546529769897461 }, { "auxiliary_loss_clip": 0.01155046, "auxiliary_loss_mlp": 0.01040337, "balance_loss_clip": 1.05856299, "balance_loss_mlp": 1.03070498, "epoch": 0.21763963205675466, "flos": 27599940211200.0, "grad_norm": 1.9531784710859919, "language_loss": 0.80447757, "learning_rate": 3.6420434907016482e-06, "loss": 0.82643139, "num_input_tokens_seen": 38532570, "step": 1810, "time_per_iteration": 3.413646697998047 }, { "auxiliary_loss_clip": 0.01205137, "auxiliary_loss_mlp": 0.01031846, "balance_loss_clip": 1.06509519, "balance_loss_mlp": 1.02269721, "epoch": 0.21775987494739374, "flos": 21430414275360.0, "grad_norm": 1.6755404184355058, "language_loss": 0.8130365, "learning_rate": 3.6415986536500606e-06, "loss": 0.83540642, "num_input_tokens_seen": 38550900, "step": 1811, "time_per_iteration": 2.588350296020508 }, { "auxiliary_loss_clip": 0.0115416, "auxiliary_loss_mlp": 0.01037839, "balance_loss_clip": 1.06033409, "balance_loss_mlp": 1.02869606, "epoch": 0.21788011783803282, "flos": 18332975814720.0, "grad_norm": 1.6569306549466367, "language_loss": 0.80673659, "learning_rate": 3.641153567569263e-06, "loss": 0.82865661, "num_input_tokens_seen": 38569215, "step": 1812, "time_per_iteration": 2.6157784461975098 }, { "auxiliary_loss_clip": 0.0119983, "auxiliary_loss_mlp": 0.01039563, "balance_loss_clip": 1.06048846, "balance_loss_mlp": 1.03049672, "epoch": 0.2180003607286719, "flos": 30262752364800.0, "grad_norm": 2.142398228759643, "language_loss": 0.95137191, "learning_rate": 3.640708232526774e-06, "loss": 0.97376585, "num_input_tokens_seen": 38587870, "step": 1813, "time_per_iteration": 2.6546835899353027 }, { "auxiliary_loss_clip": 0.01141051, "auxiliary_loss_mlp": 0.010386, "balance_loss_clip": 1.04977441, "balance_loss_mlp": 1.02886105, "epoch": 0.21812060361931102, "flos": 25480276252800.0, "grad_norm": 1.943784926951332, "language_loss": 0.78373623, "learning_rate": 3.6402626485901504e-06, "loss": 0.80553275, "num_input_tokens_seen": 38606965, "step": 1814, "time_per_iteration": 3.5965850353240967 }, { "auxiliary_loss_clip": 0.01200651, "auxiliary_loss_mlp": 0.01044435, "balance_loss_clip": 1.0637387, "balance_loss_mlp": 1.03613782, "epoch": 0.2182408465099501, "flos": 21908170463040.0, "grad_norm": 1.919413721927181, "language_loss": 0.78317273, "learning_rate": 3.639816815826988e-06, "loss": 0.80562353, "num_input_tokens_seen": 38626290, "step": 1815, "time_per_iteration": 2.5567145347595215 }, { "auxiliary_loss_clip": 0.01186125, "auxiliary_loss_mlp": 0.01039518, "balance_loss_clip": 1.05885124, "balance_loss_mlp": 1.03017759, "epoch": 0.21836108940058918, "flos": 23657343028320.0, "grad_norm": 2.2173801340701083, "language_loss": 0.78011537, "learning_rate": 3.6393707343049176e-06, "loss": 0.80237186, "num_input_tokens_seen": 38646620, "step": 1816, "time_per_iteration": 3.461289882659912 }, { "auxiliary_loss_clip": 0.01206551, "auxiliary_loss_mlp": 0.01033648, "balance_loss_clip": 1.06025171, "balance_loss_mlp": 1.0240581, "epoch": 0.2184813322912283, "flos": 24681013017120.0, "grad_norm": 2.3095864499940846, "language_loss": 0.73348463, "learning_rate": 3.6389244040916104e-06, "loss": 0.75588667, "num_input_tokens_seen": 38665695, "step": 1817, "time_per_iteration": 2.582048177719116 }, { "auxiliary_loss_clip": 0.01177432, "auxiliary_loss_mlp": 0.00764323, "balance_loss_clip": 1.05742514, "balance_loss_mlp": 0.99999225, "epoch": 0.21860157518186737, "flos": 26574653959200.0, "grad_norm": 2.147856372969862, "language_loss": 0.79666841, "learning_rate": 3.6384778252547747e-06, "loss": 0.81608593, "num_input_tokens_seen": 38681575, "step": 1818, "time_per_iteration": 2.6597394943237305 }, { "auxiliary_loss_clip": 0.01185423, "auxiliary_loss_mlp": 0.00764384, "balance_loss_clip": 1.06182849, "balance_loss_mlp": 1.00005233, "epoch": 0.21872181807250646, "flos": 20886296322240.0, "grad_norm": 3.51848485622073, "language_loss": 0.77916765, "learning_rate": 3.638030997862155e-06, "loss": 0.79866576, "num_input_tokens_seen": 38700510, "step": 1819, "time_per_iteration": 2.579953193664551 }, { "auxiliary_loss_clip": 0.01073557, "auxiliary_loss_mlp": 0.01006954, "balance_loss_clip": 1.01859021, "balance_loss_mlp": 1.00526166, "epoch": 0.21884206096314554, "flos": 61209464160000.0, "grad_norm": 0.9202011742406998, "language_loss": 0.59476483, "learning_rate": 3.6375839219815356e-06, "loss": 0.61556995, "num_input_tokens_seen": 38758310, "step": 1820, "time_per_iteration": 3.095414161682129 }, { "auxiliary_loss_clip": 0.01217641, "auxiliary_loss_mlp": 0.01038323, "balance_loss_clip": 1.06258273, "balance_loss_mlp": 1.02982903, "epoch": 0.21896230385378465, "flos": 23473839304320.0, "grad_norm": 2.4648627928752163, "language_loss": 0.82412517, "learning_rate": 3.6371365976807375e-06, "loss": 0.84668475, "num_input_tokens_seen": 38778705, "step": 1821, "time_per_iteration": 2.554250478744507 }, { "auxiliary_loss_clip": 0.01152639, "auxiliary_loss_mlp": 0.01036347, "balance_loss_clip": 1.05748379, "balance_loss_mlp": 1.02774, "epoch": 0.21908254674442373, "flos": 25081919687040.0, "grad_norm": 1.7424045793522236, "language_loss": 0.83597881, "learning_rate": 3.6366890250276185e-06, "loss": 0.85786867, "num_input_tokens_seen": 38799660, "step": 1822, "time_per_iteration": 2.715111255645752 }, { "auxiliary_loss_clip": 0.01219636, "auxiliary_loss_mlp": 0.01042621, "balance_loss_clip": 1.06521237, "balance_loss_mlp": 1.03368592, "epoch": 0.21920278963506282, "flos": 23513772575520.0, "grad_norm": 2.2595197109207428, "language_loss": 0.9011662, "learning_rate": 3.6362412040900764e-06, "loss": 0.92378873, "num_input_tokens_seen": 38819450, "step": 1823, "time_per_iteration": 2.5721659660339355 }, { "auxiliary_loss_clip": 0.01206734, "auxiliary_loss_mlp": 0.01039846, "balance_loss_clip": 1.06164789, "balance_loss_mlp": 1.03059506, "epoch": 0.21932303252570193, "flos": 29242243068480.0, "grad_norm": 1.8532583900527837, "language_loss": 0.80689299, "learning_rate": 3.635793134936044e-06, "loss": 0.82935882, "num_input_tokens_seen": 38840460, "step": 1824, "time_per_iteration": 2.642301559448242 }, { "auxiliary_loss_clip": 0.0120041, "auxiliary_loss_mlp": 0.01029229, "balance_loss_clip": 1.06070733, "balance_loss_mlp": 1.02038944, "epoch": 0.219443275416341, "flos": 20806860783360.0, "grad_norm": 1.7787758112067669, "language_loss": 0.73349285, "learning_rate": 3.635344817633494e-06, "loss": 0.75578922, "num_input_tokens_seen": 38859775, "step": 1825, "time_per_iteration": 2.5740714073181152 }, { "auxiliary_loss_clip": 0.01194548, "auxiliary_loss_mlp": 0.01029698, "balance_loss_clip": 1.05736494, "balance_loss_mlp": 1.02106714, "epoch": 0.2195635183069801, "flos": 14501558375040.0, "grad_norm": 2.4450894621349186, "language_loss": 0.75815755, "learning_rate": 3.634896252250436e-06, "loss": 0.78039998, "num_input_tokens_seen": 38876540, "step": 1826, "time_per_iteration": 2.565638542175293 }, { "auxiliary_loss_clip": 0.01222553, "auxiliary_loss_mlp": 0.01035067, "balance_loss_clip": 1.06536198, "balance_loss_mlp": 1.02557194, "epoch": 0.2196837611976192, "flos": 24243477436320.0, "grad_norm": 1.8034106167996105, "language_loss": 0.82068217, "learning_rate": 3.6344474388549157e-06, "loss": 0.84325838, "num_input_tokens_seen": 38896195, "step": 1827, "time_per_iteration": 2.5417213439941406 }, { "auxiliary_loss_clip": 0.01204863, "auxiliary_loss_mlp": 0.01035525, "balance_loss_clip": 1.0635891, "balance_loss_mlp": 1.02729952, "epoch": 0.2198040040882583, "flos": 18074526171840.0, "grad_norm": 1.9804582865624456, "language_loss": 0.80179822, "learning_rate": 3.6339983775150183e-06, "loss": 0.82420206, "num_input_tokens_seen": 38912755, "step": 1828, "time_per_iteration": 2.538438558578491 }, { "auxiliary_loss_clip": 0.01204476, "auxiliary_loss_mlp": 0.0103287, "balance_loss_clip": 1.06455493, "balance_loss_mlp": 1.02410853, "epoch": 0.21992424697889737, "flos": 17784188656800.0, "grad_norm": 2.942165642615822, "language_loss": 0.84051472, "learning_rate": 3.6335490682988664e-06, "loss": 0.86288822, "num_input_tokens_seen": 38928365, "step": 1829, "time_per_iteration": 2.524487018585205 }, { "auxiliary_loss_clip": 0.011297, "auxiliary_loss_mlp": 0.01031558, "balance_loss_clip": 1.04834092, "balance_loss_mlp": 1.02270043, "epoch": 0.22004448986953645, "flos": 17638499103360.0, "grad_norm": 1.8500979502139672, "language_loss": 0.82634664, "learning_rate": 3.63309951127462e-06, "loss": 0.84795922, "num_input_tokens_seen": 38945275, "step": 1830, "time_per_iteration": 2.6709983348846436 }, { "auxiliary_loss_clip": 0.01176062, "auxiliary_loss_mlp": 0.01032322, "balance_loss_clip": 1.05970311, "balance_loss_mlp": 1.02339363, "epoch": 0.22016473276017556, "flos": 22275537080640.0, "grad_norm": 2.530057011542651, "language_loss": 0.74909389, "learning_rate": 3.6326497065104757e-06, "loss": 0.77117777, "num_input_tokens_seen": 38965740, "step": 1831, "time_per_iteration": 2.6251068115234375 }, { "auxiliary_loss_clip": 0.01210005, "auxiliary_loss_mlp": 0.0103429, "balance_loss_clip": 1.06395257, "balance_loss_mlp": 1.02506876, "epoch": 0.22028497565081465, "flos": 25556263763520.0, "grad_norm": 2.7443181001006858, "language_loss": 0.77931356, "learning_rate": 3.6321996540746697e-06, "loss": 0.8017565, "num_input_tokens_seen": 38984815, "step": 1832, "time_per_iteration": 2.602400779724121 }, { "auxiliary_loss_clip": 0.01167081, "auxiliary_loss_mlp": 0.01036707, "balance_loss_clip": 1.05448484, "balance_loss_mlp": 1.02810645, "epoch": 0.22040521854145373, "flos": 36247337898240.0, "grad_norm": 1.876395734367707, "language_loss": 0.80562818, "learning_rate": 3.6317493540354733e-06, "loss": 0.82766604, "num_input_tokens_seen": 39008230, "step": 1833, "time_per_iteration": 2.7082300186157227 }, { "auxiliary_loss_clip": 0.01196242, "auxiliary_loss_mlp": 0.0102494, "balance_loss_clip": 1.05885553, "balance_loss_mlp": 1.0158267, "epoch": 0.22052546143209284, "flos": 11838422968800.0, "grad_norm": 2.207252972557314, "language_loss": 0.77125919, "learning_rate": 3.6312988064611976e-06, "loss": 0.79347098, "num_input_tokens_seen": 39026540, "step": 1834, "time_per_iteration": 3.270556688308716 }, { "auxiliary_loss_clip": 0.01174494, "auxiliary_loss_mlp": 0.01029415, "balance_loss_clip": 1.05448866, "balance_loss_mlp": 1.0204922, "epoch": 0.22064570432273192, "flos": 24209254961760.0, "grad_norm": 1.9705441995356066, "language_loss": 0.81407756, "learning_rate": 3.6308480114201896e-06, "loss": 0.83611655, "num_input_tokens_seen": 39048460, "step": 1835, "time_per_iteration": 2.6977829933166504 }, { "auxiliary_loss_clip": 0.01218308, "auxiliary_loss_mlp": 0.01036775, "balance_loss_clip": 1.06447566, "balance_loss_mlp": 1.02745283, "epoch": 0.220765947213371, "flos": 17931350805600.0, "grad_norm": 1.6879401603171837, "language_loss": 0.76365119, "learning_rate": 3.630396968980835e-06, "loss": 0.78620201, "num_input_tokens_seen": 39066335, "step": 1836, "time_per_iteration": 3.2735514640808105 }, { "auxiliary_loss_clip": 0.01190402, "auxiliary_loss_mlp": 0.01030797, "balance_loss_clip": 1.05996644, "balance_loss_mlp": 1.02150452, "epoch": 0.2208861901040101, "flos": 26757044257440.0, "grad_norm": 2.4322906553760006, "language_loss": 0.83529091, "learning_rate": 3.6299456792115575e-06, "loss": 0.85750282, "num_input_tokens_seen": 39087590, "step": 1837, "time_per_iteration": 2.635345935821533 }, { "auxiliary_loss_clip": 0.01110367, "auxiliary_loss_mlp": 0.01035024, "balance_loss_clip": 1.04694605, "balance_loss_mlp": 1.02592254, "epoch": 0.2210064329946492, "flos": 17817980127840.0, "grad_norm": 2.0618724167074327, "language_loss": 0.8106817, "learning_rate": 3.629494142180815e-06, "loss": 0.83213556, "num_input_tokens_seen": 39106335, "step": 1838, "time_per_iteration": 2.6965017318725586 }, { "auxiliary_loss_clip": 0.01218486, "auxiliary_loss_mlp": 0.01031916, "balance_loss_clip": 1.06451452, "balance_loss_mlp": 1.02316618, "epoch": 0.22112667588528828, "flos": 17967405045120.0, "grad_norm": 2.282093613829589, "language_loss": 0.8507241, "learning_rate": 3.6290423579571075e-06, "loss": 0.87322819, "num_input_tokens_seen": 39122875, "step": 1839, "time_per_iteration": 2.5076241493225098 }, { "auxiliary_loss_clip": 0.01196776, "auxiliary_loss_mlp": 0.01034745, "balance_loss_clip": 1.05902457, "balance_loss_mlp": 1.02557802, "epoch": 0.22124691877592736, "flos": 18369209639040.0, "grad_norm": 1.7866185829700134, "language_loss": 0.80256438, "learning_rate": 3.6285903266089694e-06, "loss": 0.82487953, "num_input_tokens_seen": 39142150, "step": 1840, "time_per_iteration": 3.4649410247802734 }, { "auxiliary_loss_clip": 0.0119203, "auxiliary_loss_mlp": 0.01030687, "balance_loss_clip": 1.06041098, "balance_loss_mlp": 1.02178252, "epoch": 0.22136716166656648, "flos": 20813289919200.0, "grad_norm": 1.8067488975846377, "language_loss": 0.76863873, "learning_rate": 3.628138048204974e-06, "loss": 0.79086584, "num_input_tokens_seen": 39162835, "step": 1841, "time_per_iteration": 2.5990421772003174 }, { "auxiliary_loss_clip": 0.01149876, "auxiliary_loss_mlp": 0.01041219, "balance_loss_clip": 1.05426621, "balance_loss_mlp": 1.03145027, "epoch": 0.22148740455720556, "flos": 17675702685600.0, "grad_norm": 1.7861285412539385, "language_loss": 0.76036811, "learning_rate": 3.6276855228137304e-06, "loss": 0.78227913, "num_input_tokens_seen": 39181040, "step": 1842, "time_per_iteration": 3.5342586040496826 }, { "auxiliary_loss_clip": 0.01218208, "auxiliary_loss_mlp": 0.00764628, "balance_loss_clip": 1.06371093, "balance_loss_mlp": 1.00010729, "epoch": 0.22160764744784464, "flos": 21726714005760.0, "grad_norm": 3.1389645076353943, "language_loss": 0.82173705, "learning_rate": 3.6272327505038874e-06, "loss": 0.84156537, "num_input_tokens_seen": 39197505, "step": 1843, "time_per_iteration": 2.521923303604126 }, { "auxiliary_loss_clip": 0.01161133, "auxiliary_loss_mlp": 0.01028363, "balance_loss_clip": 1.05508256, "balance_loss_mlp": 1.01987565, "epoch": 0.22172789033848372, "flos": 23764715573760.0, "grad_norm": 2.3701313342143027, "language_loss": 0.78270423, "learning_rate": 3.626779731344131e-06, "loss": 0.80459917, "num_input_tokens_seen": 39217295, "step": 1844, "time_per_iteration": 2.668968439102173 }, { "auxiliary_loss_clip": 0.01212651, "auxiliary_loss_mlp": 0.0103239, "balance_loss_clip": 1.06085873, "balance_loss_mlp": 1.0239675, "epoch": 0.22184813322912283, "flos": 16982303483040.0, "grad_norm": 2.4638702208589414, "language_loss": 0.85308027, "learning_rate": 3.6263264654031814e-06, "loss": 0.87553066, "num_input_tokens_seen": 39234195, "step": 1845, "time_per_iteration": 2.5189783573150635 }, { "auxiliary_loss_clip": 0.01067486, "auxiliary_loss_mlp": 0.01014625, "balance_loss_clip": 1.01848865, "balance_loss_mlp": 1.0126338, "epoch": 0.22196837611976192, "flos": 61823751076320.0, "grad_norm": 0.7030014667840329, "language_loss": 0.59225118, "learning_rate": 3.6258729527498008e-06, "loss": 0.61307228, "num_input_tokens_seen": 39295040, "step": 1846, "time_per_iteration": 3.1560959815979004 }, { "auxiliary_loss_clip": 0.0119356, "auxiliary_loss_mlp": 0.0103617, "balance_loss_clip": 1.06121767, "balance_loss_mlp": 1.02701497, "epoch": 0.222088619010401, "flos": 25558023694560.0, "grad_norm": 4.669284498345421, "language_loss": 0.64232033, "learning_rate": 3.6254191934527854e-06, "loss": 0.6646176, "num_input_tokens_seen": 39314395, "step": 1847, "time_per_iteration": 2.644908905029297 }, { "auxiliary_loss_clip": 0.01169089, "auxiliary_loss_mlp": 0.01036817, "balance_loss_clip": 1.05960107, "balance_loss_mlp": 1.02671456, "epoch": 0.2222088619010401, "flos": 19318616131200.0, "grad_norm": 1.8245408050593876, "language_loss": 0.65093464, "learning_rate": 3.6249651875809715e-06, "loss": 0.67299366, "num_input_tokens_seen": 39334275, "step": 1848, "time_per_iteration": 2.609147071838379 }, { "auxiliary_loss_clip": 0.01182107, "auxiliary_loss_mlp": 0.01038959, "balance_loss_clip": 1.05935133, "balance_loss_mlp": 1.03038788, "epoch": 0.2223291047916792, "flos": 19099345503360.0, "grad_norm": 1.867111960611578, "language_loss": 0.89140773, "learning_rate": 3.62451093520323e-06, "loss": 0.91361833, "num_input_tokens_seen": 39352180, "step": 1849, "time_per_iteration": 2.588837146759033 }, { "auxiliary_loss_clip": 0.01146019, "auxiliary_loss_mlp": 0.01032296, "balance_loss_clip": 1.04937649, "balance_loss_mlp": 1.0235821, "epoch": 0.22244934768231828, "flos": 20850421667520.0, "grad_norm": 2.580482411550487, "language_loss": 0.9058612, "learning_rate": 3.6240564363884714e-06, "loss": 0.92764431, "num_input_tokens_seen": 39372125, "step": 1850, "time_per_iteration": 2.650733709335327 }, { "auxiliary_loss_clip": 0.01207026, "auxiliary_loss_mlp": 0.01035877, "balance_loss_clip": 1.06145954, "balance_loss_mlp": 1.02672791, "epoch": 0.2225695905729574, "flos": 15632924161920.0, "grad_norm": 1.6861880592329157, "language_loss": 0.69921738, "learning_rate": 3.623601691205643e-06, "loss": 0.72164643, "num_input_tokens_seen": 39391200, "step": 1851, "time_per_iteration": 2.5630266666412354 }, { "auxiliary_loss_clip": 0.01197988, "auxiliary_loss_mlp": 0.01033996, "balance_loss_clip": 1.05869031, "balance_loss_mlp": 1.02554452, "epoch": 0.22268983346359647, "flos": 25373586129600.0, "grad_norm": 2.048585841747391, "language_loss": 0.81324607, "learning_rate": 3.623146699723729e-06, "loss": 0.83556592, "num_input_tokens_seen": 39410660, "step": 1852, "time_per_iteration": 2.603217601776123 }, { "auxiliary_loss_clip": 0.01187159, "auxiliary_loss_mlp": 0.01034435, "balance_loss_clip": 1.06290364, "balance_loss_mlp": 1.024827, "epoch": 0.22281007635423555, "flos": 13261455198240.0, "grad_norm": 1.9620562174078744, "language_loss": 0.77420425, "learning_rate": 3.6226914620117507e-06, "loss": 0.79642022, "num_input_tokens_seen": 39429280, "step": 1853, "time_per_iteration": 2.608569622039795 }, { "auxiliary_loss_clip": 0.01169807, "auxiliary_loss_mlp": 0.01027371, "balance_loss_clip": 1.05088997, "balance_loss_mlp": 1.01884103, "epoch": 0.22293031924487464, "flos": 15340539380160.0, "grad_norm": 1.98100361008954, "language_loss": 0.81387818, "learning_rate": 3.622235978138768e-06, "loss": 0.83585, "num_input_tokens_seen": 39446905, "step": 1854, "time_per_iteration": 2.6084067821502686 }, { "auxiliary_loss_clip": 0.01200943, "auxiliary_loss_mlp": 0.01037312, "balance_loss_clip": 1.06316555, "balance_loss_mlp": 1.02872276, "epoch": 0.22305056213551375, "flos": 22564653419040.0, "grad_norm": 2.3564549288950185, "language_loss": 0.81291533, "learning_rate": 3.621780248173877e-06, "loss": 0.83529782, "num_input_tokens_seen": 39465105, "step": 1855, "time_per_iteration": 2.5885872840881348 }, { "auxiliary_loss_clip": 0.01093906, "auxiliary_loss_mlp": 0.01001845, "balance_loss_clip": 1.01638651, "balance_loss_mlp": 0.99994963, "epoch": 0.22317080502615283, "flos": 64880430175680.0, "grad_norm": 0.8275140728988422, "language_loss": 0.61060846, "learning_rate": 3.6213242721862125e-06, "loss": 0.63156593, "num_input_tokens_seen": 39523560, "step": 1856, "time_per_iteration": 3.1600069999694824 }, { "auxiliary_loss_clip": 0.01173352, "auxiliary_loss_mlp": 0.01034123, "balance_loss_clip": 1.05484056, "balance_loss_mlp": 1.02522385, "epoch": 0.2232910479167919, "flos": 25776001311840.0, "grad_norm": 1.7863775926231573, "language_loss": 0.75124121, "learning_rate": 3.620868050244945e-06, "loss": 0.77331591, "num_input_tokens_seen": 39544040, "step": 1857, "time_per_iteration": 2.629551887512207 }, { "auxiliary_loss_clip": 0.01180053, "auxiliary_loss_mlp": 0.01031155, "balance_loss_clip": 1.05675459, "balance_loss_mlp": 1.02130198, "epoch": 0.22341129080743102, "flos": 23251803070560.0, "grad_norm": 8.89560033678704, "language_loss": 0.7751621, "learning_rate": 3.6204115824192817e-06, "loss": 0.79727411, "num_input_tokens_seen": 39561515, "step": 1858, "time_per_iteration": 2.603938579559326 }, { "auxiliary_loss_clip": 0.01176819, "auxiliary_loss_mlp": 0.01032696, "balance_loss_clip": 1.05524683, "balance_loss_mlp": 1.02351069, "epoch": 0.2235315336980701, "flos": 21214555758720.0, "grad_norm": 3.3607652177644707, "language_loss": 0.76504135, "learning_rate": 3.619954868778471e-06, "loss": 0.78713655, "num_input_tokens_seen": 39578210, "step": 1859, "time_per_iteration": 2.590332508087158 }, { "auxiliary_loss_clip": 0.01180142, "auxiliary_loss_mlp": 0.01034512, "balance_loss_clip": 1.05432355, "balance_loss_mlp": 1.02569008, "epoch": 0.2236517765887092, "flos": 19901949016320.0, "grad_norm": 1.9046967587108612, "language_loss": 0.82986814, "learning_rate": 3.6194979093917944e-06, "loss": 0.85201472, "num_input_tokens_seen": 39597625, "step": 1860, "time_per_iteration": 3.3544154167175293 }, { "auxiliary_loss_clip": 0.01179018, "auxiliary_loss_mlp": 0.01036092, "balance_loss_clip": 1.05548859, "balance_loss_mlp": 1.0275867, "epoch": 0.22377201947934827, "flos": 23214850907040.0, "grad_norm": 1.8192565335043742, "language_loss": 0.87175715, "learning_rate": 3.6190407043285724e-06, "loss": 0.89390826, "num_input_tokens_seen": 39615360, "step": 1861, "time_per_iteration": 2.60207200050354 }, { "auxiliary_loss_clip": 0.0121657, "auxiliary_loss_mlp": 0.01037592, "balance_loss_clip": 1.06298542, "balance_loss_mlp": 1.02856183, "epoch": 0.22389226236998738, "flos": 26794247839680.0, "grad_norm": 1.8703885064616819, "language_loss": 0.7555697, "learning_rate": 3.618583253658163e-06, "loss": 0.77811134, "num_input_tokens_seen": 39635460, "step": 1862, "time_per_iteration": 2.588287591934204 }, { "auxiliary_loss_clip": 0.01158224, "auxiliary_loss_mlp": 0.00764086, "balance_loss_clip": 1.05573869, "balance_loss_mlp": 1.00017333, "epoch": 0.22401250526062647, "flos": 24170363282400.0, "grad_norm": 1.820071137987658, "language_loss": 0.8660785, "learning_rate": 3.618125557449961e-06, "loss": 0.88530159, "num_input_tokens_seen": 39653515, "step": 1863, "time_per_iteration": 3.462242841720581 }, { "auxiliary_loss_clip": 0.01194342, "auxiliary_loss_mlp": 0.01032512, "balance_loss_clip": 1.05832684, "balance_loss_mlp": 1.02348781, "epoch": 0.22413274815126555, "flos": 16759764411840.0, "grad_norm": 1.866055902665445, "language_loss": 0.82729888, "learning_rate": 3.6176676157733983e-06, "loss": 0.84956741, "num_input_tokens_seen": 39668525, "step": 1864, "time_per_iteration": 2.5433568954467773 }, { "auxiliary_loss_clip": 0.01164571, "auxiliary_loss_mlp": 0.01036929, "balance_loss_clip": 1.05504382, "balance_loss_mlp": 1.02780306, "epoch": 0.22425299104190466, "flos": 21360209395200.0, "grad_norm": 1.9952300265774092, "language_loss": 0.75999284, "learning_rate": 3.6172094286979443e-06, "loss": 0.78200787, "num_input_tokens_seen": 39685895, "step": 1865, "time_per_iteration": 2.6187474727630615 }, { "auxiliary_loss_clip": 0.01182328, "auxiliary_loss_mlp": 0.01031699, "balance_loss_clip": 1.05507839, "balance_loss_mlp": 1.02302039, "epoch": 0.22437323393254374, "flos": 32165552131680.0, "grad_norm": 1.4321927871749884, "language_loss": 0.81404936, "learning_rate": 3.6167509962931064e-06, "loss": 0.83618969, "num_input_tokens_seen": 39711595, "step": 1866, "time_per_iteration": 3.602635145187378 }, { "auxiliary_loss_clip": 0.01159632, "auxiliary_loss_mlp": 0.01035965, "balance_loss_clip": 1.05625987, "balance_loss_mlp": 1.0272752, "epoch": 0.22449347682318282, "flos": 18002812779360.0, "grad_norm": 2.4593098440036285, "language_loss": 0.77011019, "learning_rate": 3.6162923186284276e-06, "loss": 0.79206616, "num_input_tokens_seen": 39727555, "step": 1867, "time_per_iteration": 2.649362802505493 }, { "auxiliary_loss_clip": 0.01184184, "auxiliary_loss_mlp": 0.01030747, "balance_loss_clip": 1.05750346, "balance_loss_mlp": 1.02215195, "epoch": 0.2246137197138219, "flos": 18697289490720.0, "grad_norm": 1.9262920509963262, "language_loss": 0.85926974, "learning_rate": 3.6158333957734888e-06, "loss": 0.88141906, "num_input_tokens_seen": 39746145, "step": 1868, "time_per_iteration": 3.4862799644470215 }, { "auxiliary_loss_clip": 0.01173631, "auxiliary_loss_mlp": 0.01033827, "balance_loss_clip": 1.05603278, "balance_loss_mlp": 1.02560735, "epoch": 0.22473396260446102, "flos": 15590656288320.0, "grad_norm": 2.903099536815666, "language_loss": 0.83025151, "learning_rate": 3.6153742277979088e-06, "loss": 0.8523261, "num_input_tokens_seen": 39763575, "step": 1869, "time_per_iteration": 2.632175922393799 }, { "auxiliary_loss_clip": 0.0118354, "auxiliary_loss_mlp": 0.01033308, "balance_loss_clip": 1.05679536, "balance_loss_mlp": 1.02494574, "epoch": 0.2248542054951001, "flos": 14465504135520.0, "grad_norm": 1.9526442948193041, "language_loss": 0.78333062, "learning_rate": 3.6149148147713434e-06, "loss": 0.80549914, "num_input_tokens_seen": 39781810, "step": 1870, "time_per_iteration": 2.581166982650757 }, { "auxiliary_loss_clip": 0.01206823, "auxiliary_loss_mlp": 0.0103279, "balance_loss_clip": 1.06505597, "balance_loss_mlp": 1.02433169, "epoch": 0.22497444838573918, "flos": 19243885714080.0, "grad_norm": 2.0564920271851825, "language_loss": 0.86298388, "learning_rate": 3.614455156763484e-06, "loss": 0.88538003, "num_input_tokens_seen": 39800115, "step": 1871, "time_per_iteration": 2.57969069480896 }, { "auxiliary_loss_clip": 0.01147893, "auxiliary_loss_mlp": 0.01038794, "balance_loss_clip": 1.05035496, "balance_loss_mlp": 1.03050268, "epoch": 0.2250946912763783, "flos": 16910302754880.0, "grad_norm": 2.2324388754923357, "language_loss": 0.71368766, "learning_rate": 3.613995253844061e-06, "loss": 0.73555458, "num_input_tokens_seen": 39817795, "step": 1872, "time_per_iteration": 2.6310696601867676 }, { "auxiliary_loss_clip": 0.01190901, "auxiliary_loss_mlp": 0.01034537, "balance_loss_clip": 1.05601645, "balance_loss_mlp": 1.02605569, "epoch": 0.22521493416701738, "flos": 24681372186720.0, "grad_norm": 1.8678419197071912, "language_loss": 0.8051157, "learning_rate": 3.6135351060828414e-06, "loss": 0.82737005, "num_input_tokens_seen": 39838270, "step": 1873, "time_per_iteration": 2.622751235961914 }, { "auxiliary_loss_clip": 0.01221496, "auxiliary_loss_mlp": 0.01034314, "balance_loss_clip": 1.06664526, "balance_loss_mlp": 1.0251168, "epoch": 0.22533517705765646, "flos": 17821966910400.0, "grad_norm": 2.0939954457918457, "language_loss": 0.69267261, "learning_rate": 3.6130747135496285e-06, "loss": 0.71523064, "num_input_tokens_seen": 39857270, "step": 1874, "time_per_iteration": 2.5528454780578613 }, { "auxiliary_loss_clip": 0.01210961, "auxiliary_loss_mlp": 0.01033859, "balance_loss_clip": 1.06036448, "balance_loss_mlp": 1.02507925, "epoch": 0.22545541994829554, "flos": 33691395452640.0, "grad_norm": 2.1750252902854057, "language_loss": 0.66134727, "learning_rate": 3.6126140763142646e-06, "loss": 0.68379545, "num_input_tokens_seen": 39882300, "step": 1875, "time_per_iteration": 2.6546456813812256 }, { "auxiliary_loss_clip": 0.01213996, "auxiliary_loss_mlp": 0.01033647, "balance_loss_clip": 1.06077087, "balance_loss_mlp": 1.02471185, "epoch": 0.22557566283893465, "flos": 19171597650240.0, "grad_norm": 2.5076043349686628, "language_loss": 0.85953081, "learning_rate": 3.6121531944466275e-06, "loss": 0.88200724, "num_input_tokens_seen": 39899625, "step": 1876, "time_per_iteration": 2.5334460735321045 }, { "auxiliary_loss_clip": 0.01196197, "auxiliary_loss_mlp": 0.01036818, "balance_loss_clip": 1.05965006, "balance_loss_mlp": 1.02781129, "epoch": 0.22569590572957374, "flos": 20773284814080.0, "grad_norm": 2.1263025079071274, "language_loss": 0.78287542, "learning_rate": 3.611692068016633e-06, "loss": 0.80520558, "num_input_tokens_seen": 39915955, "step": 1877, "time_per_iteration": 2.5540428161621094 }, { "auxiliary_loss_clip": 0.01164319, "auxiliary_loss_mlp": 0.0103504, "balance_loss_clip": 1.05342388, "balance_loss_mlp": 1.02490723, "epoch": 0.22581614862021282, "flos": 18442718879520.0, "grad_norm": 2.346774232036403, "language_loss": 0.74965739, "learning_rate": 3.611230697094233e-06, "loss": 0.77165097, "num_input_tokens_seen": 39932655, "step": 1878, "time_per_iteration": 2.6272878646850586 }, { "auxiliary_loss_clip": 0.01185681, "auxiliary_loss_mlp": 0.01027307, "balance_loss_clip": 1.05741739, "balance_loss_mlp": 1.01912379, "epoch": 0.22593639151085193, "flos": 20048392825920.0, "grad_norm": 1.7560826771868405, "language_loss": 0.87050164, "learning_rate": 3.6107690817494173e-06, "loss": 0.89263153, "num_input_tokens_seen": 39952875, "step": 1879, "time_per_iteration": 2.5930280685424805 }, { "auxiliary_loss_clip": 0.01144895, "auxiliary_loss_mlp": 0.01033577, "balance_loss_clip": 1.05133295, "balance_loss_mlp": 1.02497637, "epoch": 0.226056634401491, "flos": 13115119139520.0, "grad_norm": 2.441313708185933, "language_loss": 0.70710945, "learning_rate": 3.6103072220522117e-06, "loss": 0.72889423, "num_input_tokens_seen": 39968405, "step": 1880, "time_per_iteration": 2.6444103717803955 }, { "auxiliary_loss_clip": 0.01172824, "auxiliary_loss_mlp": 0.01032238, "balance_loss_clip": 1.05702174, "balance_loss_mlp": 1.02405441, "epoch": 0.2261768772921301, "flos": 18988381261920.0, "grad_norm": 1.8001107304104846, "language_loss": 0.91384828, "learning_rate": 3.609845118072682e-06, "loss": 0.9358989, "num_input_tokens_seen": 39987075, "step": 1881, "time_per_iteration": 2.6189639568328857 }, { "auxiliary_loss_clip": 0.01203211, "auxiliary_loss_mlp": 0.00764274, "balance_loss_clip": 1.05916548, "balance_loss_mlp": 1.00013304, "epoch": 0.2262971201827692, "flos": 19974057495360.0, "grad_norm": 1.7077811497095494, "language_loss": 0.79946065, "learning_rate": 3.6093827698809276e-06, "loss": 0.81913555, "num_input_tokens_seen": 40006175, "step": 1882, "time_per_iteration": 2.5963706970214844 }, { "auxiliary_loss_clip": 0.01195465, "auxiliary_loss_mlp": 0.01029338, "balance_loss_clip": 1.05569172, "balance_loss_mlp": 1.02075529, "epoch": 0.2264173630734083, "flos": 16654546884000.0, "grad_norm": 2.486699847802498, "language_loss": 0.84551758, "learning_rate": 3.6089201775470864e-06, "loss": 0.86776567, "num_input_tokens_seen": 40021630, "step": 1883, "time_per_iteration": 2.548128604888916 }, { "auxiliary_loss_clip": 0.01156251, "auxiliary_loss_mlp": 0.01030748, "balance_loss_clip": 1.05233717, "balance_loss_mlp": 1.02201605, "epoch": 0.22653760596404737, "flos": 24389813495040.0, "grad_norm": 1.3824516385616206, "language_loss": 0.77585286, "learning_rate": 3.6084573411413334e-06, "loss": 0.79772282, "num_input_tokens_seen": 40041025, "step": 1884, "time_per_iteration": 2.660585641860962 }, { "auxiliary_loss_clip": 0.01168724, "auxiliary_loss_mlp": 0.0103328, "balance_loss_clip": 1.05725312, "balance_loss_mlp": 1.02448249, "epoch": 0.22665784885468646, "flos": 18332544811200.0, "grad_norm": 3.3945630883409192, "language_loss": 0.80787641, "learning_rate": 3.607994260733881e-06, "loss": 0.82989645, "num_input_tokens_seen": 40060265, "step": 1885, "time_per_iteration": 2.629868745803833 }, { "auxiliary_loss_clip": 0.01185095, "auxiliary_loss_mlp": 0.01033514, "balance_loss_clip": 1.05513763, "balance_loss_mlp": 1.02564073, "epoch": 0.22677809174532557, "flos": 24058106030400.0, "grad_norm": 1.6777587585176026, "language_loss": 0.74798667, "learning_rate": 3.6075309363949776e-06, "loss": 0.77017272, "num_input_tokens_seen": 40079435, "step": 1886, "time_per_iteration": 3.361940622329712 }, { "auxiliary_loss_clip": 0.01212091, "auxiliary_loss_mlp": 0.01033297, "balance_loss_clip": 1.06009531, "balance_loss_mlp": 1.0250473, "epoch": 0.22689833463596465, "flos": 20374245826080.0, "grad_norm": 1.7993601759882158, "language_loss": 0.81397349, "learning_rate": 3.6070673681949094e-06, "loss": 0.83642739, "num_input_tokens_seen": 40097800, "step": 1887, "time_per_iteration": 2.5322859287261963 }, { "auxiliary_loss_clip": 0.01185613, "auxiliary_loss_mlp": 0.00763541, "balance_loss_clip": 1.05890465, "balance_loss_mlp": 1.00010896, "epoch": 0.22701857752660373, "flos": 30120403088640.0, "grad_norm": 2.1124184358498557, "language_loss": 0.8075785, "learning_rate": 3.606603556203999e-06, "loss": 0.82707006, "num_input_tokens_seen": 40122745, "step": 1888, "time_per_iteration": 3.4175233840942383 }, { "auxiliary_loss_clip": 0.01198194, "auxiliary_loss_mlp": 0.01028168, "balance_loss_clip": 1.057127, "balance_loss_mlp": 1.01967406, "epoch": 0.22713882041724284, "flos": 22492185770400.0, "grad_norm": 2.033014030277137, "language_loss": 0.83588046, "learning_rate": 3.6061395004926066e-06, "loss": 0.85814404, "num_input_tokens_seen": 40141680, "step": 1889, "time_per_iteration": 2.590337038040161 }, { "auxiliary_loss_clip": 0.01213602, "auxiliary_loss_mlp": 0.0103126, "balance_loss_clip": 1.05963111, "balance_loss_mlp": 1.02273655, "epoch": 0.22725906330788193, "flos": 20521551642720.0, "grad_norm": 3.7668744762426174, "language_loss": 0.84862125, "learning_rate": 3.605675201131129e-06, "loss": 0.87106985, "num_input_tokens_seen": 40160140, "step": 1890, "time_per_iteration": 2.522930383682251 }, { "auxiliary_loss_clip": 0.01204202, "auxiliary_loss_mlp": 0.01035049, "balance_loss_clip": 1.06179333, "balance_loss_mlp": 1.02655554, "epoch": 0.227379306198521, "flos": 18989925691200.0, "grad_norm": 2.1346333601818004, "language_loss": 0.79685116, "learning_rate": 3.60521065819e-06, "loss": 0.81924367, "num_input_tokens_seen": 40177450, "step": 1891, "time_per_iteration": 2.62439227104187 }, { "auxiliary_loss_clip": 0.01184964, "auxiliary_loss_mlp": 0.01036598, "balance_loss_clip": 1.05469227, "balance_loss_mlp": 1.02818155, "epoch": 0.2274995490891601, "flos": 21798355564320.0, "grad_norm": 1.8157603210585058, "language_loss": 0.87426823, "learning_rate": 3.60474587173969e-06, "loss": 0.89648384, "num_input_tokens_seen": 40195935, "step": 1892, "time_per_iteration": 3.529618978500366 }, { "auxiliary_loss_clip": 0.01196103, "auxiliary_loss_mlp": 0.01038866, "balance_loss_clip": 1.06005383, "balance_loss_mlp": 1.03010416, "epoch": 0.2276197919797992, "flos": 19058657976000.0, "grad_norm": 2.158546948023408, "language_loss": 0.83934414, "learning_rate": 3.6042808418507084e-06, "loss": 0.86169386, "num_input_tokens_seen": 40213620, "step": 1893, "time_per_iteration": 2.5429461002349854 }, { "auxiliary_loss_clip": 0.01201726, "auxiliary_loss_mlp": 0.01032761, "balance_loss_clip": 1.06151319, "balance_loss_mlp": 1.02369547, "epoch": 0.22774003487043828, "flos": 18806781136800.0, "grad_norm": 2.39335653960736, "language_loss": 0.77212906, "learning_rate": 3.6038155685935976e-06, "loss": 0.79447395, "num_input_tokens_seen": 40230190, "step": 1894, "time_per_iteration": 3.443523645401001 }, { "auxiliary_loss_clip": 0.01197925, "auxiliary_loss_mlp": 0.01036062, "balance_loss_clip": 1.06021261, "balance_loss_mlp": 1.02738953, "epoch": 0.22786027776107737, "flos": 23002548169440.0, "grad_norm": 1.9512683878605386, "language_loss": 0.70398045, "learning_rate": 3.6033500520389404e-06, "loss": 0.72632027, "num_input_tokens_seen": 40246860, "step": 1895, "time_per_iteration": 2.555091142654419 }, { "auxiliary_loss_clip": 0.01061093, "auxiliary_loss_mlp": 0.01008565, "balance_loss_clip": 1.01558518, "balance_loss_mlp": 1.00681305, "epoch": 0.22798052065171648, "flos": 66706883262240.0, "grad_norm": 0.7909671885793701, "language_loss": 0.64830995, "learning_rate": 3.6028842922573553e-06, "loss": 0.66900659, "num_input_tokens_seen": 40311005, "step": 1896, "time_per_iteration": 3.294739007949829 }, { "auxiliary_loss_clip": 0.01076167, "auxiliary_loss_mlp": 0.00755542, "balance_loss_clip": 1.01746106, "balance_loss_mlp": 0.99953502, "epoch": 0.22810076354235556, "flos": 62080907708640.0, "grad_norm": 0.8581820274201428, "language_loss": 0.62896383, "learning_rate": 3.602418289319497e-06, "loss": 0.64728093, "num_input_tokens_seen": 40369560, "step": 1897, "time_per_iteration": 3.176879405975342 }, { "auxiliary_loss_clip": 0.01156584, "auxiliary_loss_mlp": 0.010307, "balance_loss_clip": 1.0542779, "balance_loss_mlp": 1.02184844, "epoch": 0.22822100643299464, "flos": 23876362237440.0, "grad_norm": 1.6635799584426194, "language_loss": 0.73340809, "learning_rate": 3.601952043296059e-06, "loss": 0.75528091, "num_input_tokens_seen": 40389555, "step": 1898, "time_per_iteration": 2.701472043991089 }, { "auxiliary_loss_clip": 0.0118811, "auxiliary_loss_mlp": 0.01026038, "balance_loss_clip": 1.05535555, "balance_loss_mlp": 1.01778281, "epoch": 0.22834124932363373, "flos": 20991334265280.0, "grad_norm": 1.878544077119601, "language_loss": 0.80255389, "learning_rate": 3.6014855542577696e-06, "loss": 0.82469535, "num_input_tokens_seen": 40406765, "step": 1899, "time_per_iteration": 2.5855085849761963 }, { "auxiliary_loss_clip": 0.01184903, "auxiliary_loss_mlp": 0.01028181, "balance_loss_clip": 1.05826831, "balance_loss_mlp": 1.01933539, "epoch": 0.22846149221427284, "flos": 24901576655520.0, "grad_norm": 5.709325458815795, "language_loss": 0.84220642, "learning_rate": 3.6010188222753943e-06, "loss": 0.86433727, "num_input_tokens_seen": 40427535, "step": 1900, "time_per_iteration": 2.665693521499634 }, { "auxiliary_loss_clip": 0.01082597, "auxiliary_loss_mlp": 0.01003246, "balance_loss_clip": 1.01922739, "balance_loss_mlp": 1.00133824, "epoch": 0.22858173510491192, "flos": 56132304580800.0, "grad_norm": 0.8992523421617361, "language_loss": 0.64060438, "learning_rate": 3.6005518474197372e-06, "loss": 0.66146278, "num_input_tokens_seen": 40479580, "step": 1901, "time_per_iteration": 3.08784818649292 }, { "auxiliary_loss_clip": 0.01203373, "auxiliary_loss_mlp": 0.01034145, "balance_loss_clip": 1.06240714, "balance_loss_mlp": 1.02476954, "epoch": 0.228701977995551, "flos": 24170830202880.0, "grad_norm": 1.942699704536538, "language_loss": 0.78582764, "learning_rate": 3.6000846297616373e-06, "loss": 0.80820286, "num_input_tokens_seen": 40497880, "step": 1902, "time_per_iteration": 2.5865681171417236 }, { "auxiliary_loss_clip": 0.01219772, "auxiliary_loss_mlp": 0.01030412, "balance_loss_clip": 1.06494212, "balance_loss_mlp": 1.02026153, "epoch": 0.22882222088619011, "flos": 21387894983040.0, "grad_norm": 2.4154075019582972, "language_loss": 0.72302639, "learning_rate": 3.5996171693719717e-06, "loss": 0.74552822, "num_input_tokens_seen": 40513975, "step": 1903, "time_per_iteration": 2.538630962371826 }, { "auxiliary_loss_clip": 0.01095432, "auxiliary_loss_mlp": 0.01001977, "balance_loss_clip": 1.01891518, "balance_loss_mlp": 1.0001173, "epoch": 0.2289424637768292, "flos": 64589625740160.0, "grad_norm": 0.8458957247212121, "language_loss": 0.64860553, "learning_rate": 3.5991494663216528e-06, "loss": 0.66957963, "num_input_tokens_seen": 40576960, "step": 1904, "time_per_iteration": 3.1939265727996826 }, { "auxiliary_loss_clip": 0.01215464, "auxiliary_loss_mlp": 0.01033237, "balance_loss_clip": 1.06370425, "balance_loss_mlp": 1.02490377, "epoch": 0.22906270666746828, "flos": 22163423496480.0, "grad_norm": 2.0194907594703913, "language_loss": 0.87351966, "learning_rate": 3.5986815206816314e-06, "loss": 0.8960067, "num_input_tokens_seen": 40595780, "step": 1905, "time_per_iteration": 2.5558712482452393 }, { "auxiliary_loss_clip": 0.01212606, "auxiliary_loss_mlp": 0.01026309, "balance_loss_clip": 1.06030631, "balance_loss_mlp": 1.01737475, "epoch": 0.2291829495581074, "flos": 25772337781920.0, "grad_norm": 1.7222581491856985, "language_loss": 0.74311578, "learning_rate": 3.598213332522895e-06, "loss": 0.76550496, "num_input_tokens_seen": 40615810, "step": 1906, "time_per_iteration": 2.5716514587402344 }, { "auxiliary_loss_clip": 0.0119972, "auxiliary_loss_mlp": 0.01033864, "balance_loss_clip": 1.05899537, "balance_loss_mlp": 1.02542424, "epoch": 0.22930319244874647, "flos": 31172764340160.0, "grad_norm": 1.9475779921174714, "language_loss": 0.77776021, "learning_rate": 3.597744901916466e-06, "loss": 0.80009604, "num_input_tokens_seen": 40637095, "step": 1907, "time_per_iteration": 2.655810594558716 }, { "auxiliary_loss_clip": 0.01219842, "auxiliary_loss_mlp": 0.01032062, "balance_loss_clip": 1.06287682, "balance_loss_mlp": 1.02294207, "epoch": 0.22942343533938556, "flos": 23254101756000.0, "grad_norm": 2.066682969114109, "language_loss": 0.76959819, "learning_rate": 3.5972762289334058e-06, "loss": 0.79211724, "num_input_tokens_seen": 40656725, "step": 1908, "time_per_iteration": 2.553893804550171 }, { "auxiliary_loss_clip": 0.01133981, "auxiliary_loss_mlp": 0.0103761, "balance_loss_clip": 1.05243635, "balance_loss_mlp": 1.02875257, "epoch": 0.22954367823002464, "flos": 14610906353280.0, "grad_norm": 2.129605782257861, "language_loss": 0.84843421, "learning_rate": 3.5968073136448116e-06, "loss": 0.87015009, "num_input_tokens_seen": 40674745, "step": 1909, "time_per_iteration": 2.684384822845459 }, { "auxiliary_loss_clip": 0.01202907, "auxiliary_loss_mlp": 0.01034744, "balance_loss_clip": 1.05920291, "balance_loss_mlp": 1.02555275, "epoch": 0.22966392112066375, "flos": 16763607526560.0, "grad_norm": 1.7834676047987545, "language_loss": 0.91404837, "learning_rate": 3.596338156121818e-06, "loss": 0.93642485, "num_input_tokens_seen": 40693630, "step": 1910, "time_per_iteration": 2.545536994934082 }, { "auxiliary_loss_clip": 0.01080499, "auxiliary_loss_mlp": 0.01001215, "balance_loss_clip": 1.01707852, "balance_loss_mlp": 0.99931937, "epoch": 0.22978416401130283, "flos": 67474258625760.0, "grad_norm": 0.7416500600942699, "language_loss": 0.59353048, "learning_rate": 3.595868756435595e-06, "loss": 0.61434764, "num_input_tokens_seen": 40761310, "step": 1911, "time_per_iteration": 3.2905871868133545 }, { "auxiliary_loss_clip": 0.01175051, "auxiliary_loss_mlp": 0.01037811, "balance_loss_clip": 1.05950844, "balance_loss_mlp": 1.02922153, "epoch": 0.22990440690194192, "flos": 19865140520640.0, "grad_norm": 2.13956151629247, "language_loss": 0.80576587, "learning_rate": 3.5953991146573504e-06, "loss": 0.82789445, "num_input_tokens_seen": 40779955, "step": 1912, "time_per_iteration": 2.646497964859009 }, { "auxiliary_loss_clip": 0.01201892, "auxiliary_loss_mlp": 0.01030102, "balance_loss_clip": 1.05739307, "balance_loss_mlp": 1.0199635, "epoch": 0.23002464979258103, "flos": 13289248536960.0, "grad_norm": 2.3723057205344618, "language_loss": 0.83385932, "learning_rate": 3.5949292308583294e-06, "loss": 0.8561793, "num_input_tokens_seen": 40793200, "step": 1913, "time_per_iteration": 3.298058032989502 }, { "auxiliary_loss_clip": 0.01214722, "auxiliary_loss_mlp": 0.01031094, "balance_loss_clip": 1.06265092, "balance_loss_mlp": 1.02205765, "epoch": 0.2301448926832201, "flos": 22163782666080.0, "grad_norm": 1.9345737218502355, "language_loss": 0.80312037, "learning_rate": 3.594459105109811e-06, "loss": 0.82557845, "num_input_tokens_seen": 40812380, "step": 1914, "time_per_iteration": 3.2965147495269775 }, { "auxiliary_loss_clip": 0.01204463, "auxiliary_loss_mlp": 0.0103587, "balance_loss_clip": 1.06211925, "balance_loss_mlp": 1.02677476, "epoch": 0.2302651355738592, "flos": 20704480695360.0, "grad_norm": 3.425415464624299, "language_loss": 0.81167597, "learning_rate": 3.593988737483115e-06, "loss": 0.83407927, "num_input_tokens_seen": 40832320, "step": 1915, "time_per_iteration": 2.567873239517212 }, { "auxiliary_loss_clip": 0.01187012, "auxiliary_loss_mlp": 0.01032403, "balance_loss_clip": 1.05949962, "balance_loss_mlp": 1.02353394, "epoch": 0.23038537846449827, "flos": 18588947187360.0, "grad_norm": 1.9963546718404876, "language_loss": 0.78155959, "learning_rate": 3.5935181280495947e-06, "loss": 0.80375379, "num_input_tokens_seen": 40850900, "step": 1916, "time_per_iteration": 2.537890911102295 }, { "auxiliary_loss_clip": 0.01081568, "auxiliary_loss_mlp": 0.01012462, "balance_loss_clip": 1.0184741, "balance_loss_mlp": 1.01054311, "epoch": 0.23050562135513739, "flos": 64224270472320.0, "grad_norm": 0.801932317212394, "language_loss": 0.54335904, "learning_rate": 3.5930472768806412e-06, "loss": 0.56429935, "num_input_tokens_seen": 40909570, "step": 1917, "time_per_iteration": 3.136352300643921 }, { "auxiliary_loss_clip": 0.01220441, "auxiliary_loss_mlp": 0.01030388, "balance_loss_clip": 1.06673884, "balance_loss_mlp": 1.02119672, "epoch": 0.23062586424577647, "flos": 17313400359360.0, "grad_norm": 1.8661463894642252, "language_loss": 0.77124727, "learning_rate": 3.5925761840476826e-06, "loss": 0.79375547, "num_input_tokens_seen": 40928180, "step": 1918, "time_per_iteration": 2.549668550491333 }, { "auxiliary_loss_clip": 0.01181629, "auxiliary_loss_mlp": 0.01035503, "balance_loss_clip": 1.06011701, "balance_loss_mlp": 1.02686596, "epoch": 0.23074610713641555, "flos": 27855983417760.0, "grad_norm": 2.2318701907105174, "language_loss": 0.81578827, "learning_rate": 3.592104849622183e-06, "loss": 0.83795965, "num_input_tokens_seen": 40950435, "step": 1919, "time_per_iteration": 3.5460493564605713 }, { "auxiliary_loss_clip": 0.01146269, "auxiliary_loss_mlp": 0.01029077, "balance_loss_clip": 1.05347538, "balance_loss_mlp": 1.01989198, "epoch": 0.23086635002705466, "flos": 28841803319040.0, "grad_norm": 4.697275380802103, "language_loss": 0.73656416, "learning_rate": 3.591633273675644e-06, "loss": 0.75831771, "num_input_tokens_seen": 40972670, "step": 1920, "time_per_iteration": 3.6036436557769775 }, { "auxiliary_loss_clip": 0.01071275, "auxiliary_loss_mlp": 0.01022663, "balance_loss_clip": 1.03346312, "balance_loss_mlp": 1.02082729, "epoch": 0.23098659291769374, "flos": 62923689523200.0, "grad_norm": 0.9108483890883081, "language_loss": 0.58162928, "learning_rate": 3.591161456279602e-06, "loss": 0.60256869, "num_input_tokens_seen": 41018215, "step": 1921, "time_per_iteration": 3.0313615798950195 }, { "auxiliary_loss_clip": 0.01187772, "auxiliary_loss_mlp": 0.01030381, "balance_loss_clip": 1.05737281, "balance_loss_mlp": 1.02138078, "epoch": 0.23110683580833283, "flos": 23476820412000.0, "grad_norm": 1.5999911664076771, "language_loss": 0.80046773, "learning_rate": 3.590689397505633e-06, "loss": 0.82264924, "num_input_tokens_seen": 41039125, "step": 1922, "time_per_iteration": 2.6447701454162598 }, { "auxiliary_loss_clip": 0.01211474, "auxiliary_loss_mlp": 0.0103367, "balance_loss_clip": 1.06101024, "balance_loss_mlp": 1.02498579, "epoch": 0.2312270786989719, "flos": 27271070186400.0, "grad_norm": 1.68165932454291, "language_loss": 0.86583734, "learning_rate": 3.590217097425347e-06, "loss": 0.88828874, "num_input_tokens_seen": 41059025, "step": 1923, "time_per_iteration": 2.576550006866455 }, { "auxiliary_loss_clip": 0.0121799, "auxiliary_loss_mlp": 0.01035296, "balance_loss_clip": 1.06525731, "balance_loss_mlp": 1.02615309, "epoch": 0.23134732158961102, "flos": 13261347447360.0, "grad_norm": 1.995378700500369, "language_loss": 0.71284205, "learning_rate": 3.589744556110391e-06, "loss": 0.73537493, "num_input_tokens_seen": 41077015, "step": 1924, "time_per_iteration": 2.534431219100952 }, { "auxiliary_loss_clip": 0.01179512, "auxiliary_loss_mlp": 0.01035277, "balance_loss_clip": 1.05482984, "balance_loss_mlp": 1.02603805, "epoch": 0.2314675644802501, "flos": 36977653347360.0, "grad_norm": 1.691622201697321, "language_loss": 0.84528172, "learning_rate": 3.58927177363245e-06, "loss": 0.86742961, "num_input_tokens_seen": 41099840, "step": 1925, "time_per_iteration": 2.7172353267669678 }, { "auxiliary_loss_clip": 0.01166361, "auxiliary_loss_mlp": 0.01032517, "balance_loss_clip": 1.05509186, "balance_loss_mlp": 1.02311707, "epoch": 0.2315878073708892, "flos": 23842211596800.0, "grad_norm": 2.61108674651918, "language_loss": 0.72629172, "learning_rate": 3.5887987500632447e-06, "loss": 0.74828053, "num_input_tokens_seen": 41117845, "step": 1926, "time_per_iteration": 2.667046546936035 }, { "auxiliary_loss_clip": 0.01177298, "auxiliary_loss_mlp": 0.01026672, "balance_loss_clip": 1.05849826, "balance_loss_mlp": 1.0179522, "epoch": 0.2317080502615283, "flos": 23039428499040.0, "grad_norm": 1.6853545178263583, "language_loss": 0.84099901, "learning_rate": 3.5883254854745325e-06, "loss": 0.86303866, "num_input_tokens_seen": 41136235, "step": 1927, "time_per_iteration": 2.638756513595581 }, { "auxiliary_loss_clip": 0.01204668, "auxiliary_loss_mlp": 0.01035644, "balance_loss_clip": 1.05889595, "balance_loss_mlp": 1.02583861, "epoch": 0.23182829315216738, "flos": 11254659080160.0, "grad_norm": 3.832254323484343, "language_loss": 0.75294602, "learning_rate": 3.587851979938107e-06, "loss": 0.77534914, "num_input_tokens_seen": 41153125, "step": 1928, "time_per_iteration": 2.5494370460510254 }, { "auxiliary_loss_clip": 0.01201178, "auxiliary_loss_mlp": 0.01030459, "balance_loss_clip": 1.06249118, "balance_loss_mlp": 1.02168548, "epoch": 0.23194853604280646, "flos": 19828942613280.0, "grad_norm": 2.016440162298553, "language_loss": 0.7709893, "learning_rate": 3.5873782335257985e-06, "loss": 0.7933057, "num_input_tokens_seen": 41171290, "step": 1929, "time_per_iteration": 2.546767234802246 }, { "auxiliary_loss_clip": 0.0117107, "auxiliary_loss_mlp": 0.01030243, "balance_loss_clip": 1.06103802, "balance_loss_mlp": 1.02118301, "epoch": 0.23206877893344555, "flos": 15305023895040.0, "grad_norm": 1.8914406831467396, "language_loss": 0.78524053, "learning_rate": 3.5869042463094744e-06, "loss": 0.8072536, "num_input_tokens_seen": 41189005, "step": 1930, "time_per_iteration": 2.621192693710327 }, { "auxiliary_loss_clip": 0.01136271, "auxiliary_loss_mlp": 0.01032641, "balance_loss_clip": 1.05133796, "balance_loss_mlp": 1.02360535, "epoch": 0.23218902182408466, "flos": 22711492315200.0, "grad_norm": 3.7411208846595905, "language_loss": 0.77027059, "learning_rate": 3.586430018361038e-06, "loss": 0.79195976, "num_input_tokens_seen": 41208775, "step": 1931, "time_per_iteration": 2.6494340896606445 }, { "auxiliary_loss_clip": 0.01170625, "auxiliary_loss_mlp": 0.01033006, "balance_loss_clip": 1.05365086, "balance_loss_mlp": 1.02382064, "epoch": 0.23230926471472374, "flos": 22710738059040.0, "grad_norm": 2.074682701815147, "language_loss": 0.76526093, "learning_rate": 3.5859555497524283e-06, "loss": 0.78729725, "num_input_tokens_seen": 41226010, "step": 1932, "time_per_iteration": 2.60990571975708 }, { "auxiliary_loss_clip": 0.01200352, "auxiliary_loss_mlp": 0.01032926, "balance_loss_clip": 1.0619148, "balance_loss_mlp": 1.02396739, "epoch": 0.23242950760536282, "flos": 20375502919680.0, "grad_norm": 1.8265190236781594, "language_loss": 0.92292178, "learning_rate": 3.5854808405556237e-06, "loss": 0.9452545, "num_input_tokens_seen": 41245245, "step": 1933, "time_per_iteration": 2.6012604236602783 }, { "auxiliary_loss_clip": 0.01169459, "auxiliary_loss_mlp": 0.01036287, "balance_loss_clip": 1.05574739, "balance_loss_mlp": 1.02779961, "epoch": 0.23254975049600193, "flos": 16908327322080.0, "grad_norm": 2.5158432960409627, "language_loss": 0.74645042, "learning_rate": 3.5850058908426355e-06, "loss": 0.76850796, "num_input_tokens_seen": 41263795, "step": 1934, "time_per_iteration": 2.6181154251098633 }, { "auxiliary_loss_clip": 0.01187316, "auxiliary_loss_mlp": 0.01030844, "balance_loss_clip": 1.05685329, "balance_loss_mlp": 1.02252316, "epoch": 0.23266999338664102, "flos": 23294825200320.0, "grad_norm": 2.1040704714455996, "language_loss": 0.85747808, "learning_rate": 3.584530700685514e-06, "loss": 0.87965965, "num_input_tokens_seen": 41284055, "step": 1935, "time_per_iteration": 2.616023063659668 }, { "auxiliary_loss_clip": 0.01183648, "auxiliary_loss_mlp": 0.01036917, "balance_loss_clip": 1.06221437, "balance_loss_mlp": 1.02863777, "epoch": 0.2327902362772801, "flos": 19569990132960.0, "grad_norm": 2.270595028576268, "language_loss": 0.88689435, "learning_rate": 3.5840552701563448e-06, "loss": 0.90910006, "num_input_tokens_seen": 41300255, "step": 1936, "time_per_iteration": 2.5789151191711426 }, { "auxiliary_loss_clip": 0.01214007, "auxiliary_loss_mlp": 0.01033496, "balance_loss_clip": 1.06390715, "balance_loss_mlp": 1.02516961, "epoch": 0.2329104791679192, "flos": 16727517370080.0, "grad_norm": 5.732870007276157, "language_loss": 0.82012713, "learning_rate": 3.5835795993272513e-06, "loss": 0.84260225, "num_input_tokens_seen": 41318540, "step": 1937, "time_per_iteration": 2.5181186199188232 }, { "auxiliary_loss_clip": 0.01105106, "auxiliary_loss_mlp": 0.01028383, "balance_loss_clip": 1.04771161, "balance_loss_mlp": 1.02000272, "epoch": 0.2330307220585583, "flos": 22163746749120.0, "grad_norm": 1.8147969452120507, "language_loss": 0.70998639, "learning_rate": 3.583103688270391e-06, "loss": 0.73132133, "num_input_tokens_seen": 41338320, "step": 1938, "time_per_iteration": 2.8554861545562744 }, { "auxiliary_loss_clip": 0.01171354, "auxiliary_loss_mlp": 0.01030199, "balance_loss_clip": 1.05723357, "balance_loss_mlp": 1.02046013, "epoch": 0.23315096494919738, "flos": 19317323120640.0, "grad_norm": 2.1308606223272415, "language_loss": 0.89847291, "learning_rate": 3.58262753705796e-06, "loss": 0.92048842, "num_input_tokens_seen": 41353210, "step": 1939, "time_per_iteration": 3.6052048206329346 }, { "auxiliary_loss_clip": 0.01073472, "auxiliary_loss_mlp": 0.01010721, "balance_loss_clip": 1.01558113, "balance_loss_mlp": 1.00898063, "epoch": 0.23327120783983646, "flos": 53031058922400.0, "grad_norm": 0.7548236848778813, "language_loss": 0.55536264, "learning_rate": 3.5821511457621902e-06, "loss": 0.5762046, "num_input_tokens_seen": 41410510, "step": 1940, "time_per_iteration": 4.111621618270874 }, { "auxiliary_loss_clip": 0.01180107, "auxiliary_loss_mlp": 0.01040066, "balance_loss_clip": 1.05883741, "balance_loss_mlp": 1.03009403, "epoch": 0.23339145073047557, "flos": 17126987361600.0, "grad_norm": 3.257144724071296, "language_loss": 0.81273603, "learning_rate": 3.5816745144553497e-06, "loss": 0.83493775, "num_input_tokens_seen": 41425830, "step": 1941, "time_per_iteration": 2.550218343734741 }, { "auxiliary_loss_clip": 0.01150118, "auxiliary_loss_mlp": 0.01031113, "balance_loss_clip": 1.05607271, "balance_loss_mlp": 1.02194023, "epoch": 0.23351169362111465, "flos": 13078921232160.0, "grad_norm": 1.9036183360715284, "language_loss": 0.75135982, "learning_rate": 3.5811976432097424e-06, "loss": 0.7731722, "num_input_tokens_seen": 41443500, "step": 1942, "time_per_iteration": 2.6446704864501953 }, { "auxiliary_loss_clip": 0.0120162, "auxiliary_loss_mlp": 0.00764249, "balance_loss_clip": 1.0639019, "balance_loss_mlp": 1.00026178, "epoch": 0.23363193651175373, "flos": 15851260948800.0, "grad_norm": 1.9293015214531146, "language_loss": 0.84652627, "learning_rate": 3.58072053209771e-06, "loss": 0.86618495, "num_input_tokens_seen": 41460055, "step": 1943, "time_per_iteration": 2.5836341381073 }, { "auxiliary_loss_clip": 0.01175698, "auxiliary_loss_mlp": 0.01032314, "balance_loss_clip": 1.05519843, "balance_loss_mlp": 1.02239561, "epoch": 0.23375217940239285, "flos": 21025772241600.0, "grad_norm": 2.3109567838054295, "language_loss": 0.7890805, "learning_rate": 3.5802431811916296e-06, "loss": 0.81116062, "num_input_tokens_seen": 41476665, "step": 1944, "time_per_iteration": 3.492157220840454 }, { "auxiliary_loss_clip": 0.01182193, "auxiliary_loss_mlp": 0.01028484, "balance_loss_clip": 1.05878723, "balance_loss_mlp": 1.01949525, "epoch": 0.23387242229303193, "flos": 20594701713600.0, "grad_norm": 1.7688149666961233, "language_loss": 0.8068409, "learning_rate": 3.579765590563916e-06, "loss": 0.82894766, "num_input_tokens_seen": 41496065, "step": 1945, "time_per_iteration": 2.594571352005005 }, { "auxiliary_loss_clip": 0.01189044, "auxiliary_loss_mlp": 0.01037952, "balance_loss_clip": 1.05819488, "balance_loss_mlp": 1.02896404, "epoch": 0.233992665183671, "flos": 24279495758880.0, "grad_norm": 1.8634765401961577, "language_loss": 0.81797522, "learning_rate": 3.579287760287017e-06, "loss": 0.84024513, "num_input_tokens_seen": 41516815, "step": 1946, "time_per_iteration": 3.412222385406494 }, { "auxiliary_loss_clip": 0.01198139, "auxiliary_loss_mlp": 0.01026334, "balance_loss_clip": 1.06227362, "balance_loss_mlp": 1.01751876, "epoch": 0.2341129080743101, "flos": 30154625563200.0, "grad_norm": 1.7857035601177604, "language_loss": 0.73029864, "learning_rate": 3.578809690433421e-06, "loss": 0.75254339, "num_input_tokens_seen": 41538525, "step": 1947, "time_per_iteration": 2.6385769844055176 }, { "auxiliary_loss_clip": 0.01220422, "auxiliary_loss_mlp": 0.01042574, "balance_loss_clip": 1.06591439, "balance_loss_mlp": 1.03259015, "epoch": 0.2342331509649492, "flos": 22784139548640.0, "grad_norm": 2.495276075152083, "language_loss": 0.80992687, "learning_rate": 3.578331381075651e-06, "loss": 0.83255678, "num_input_tokens_seen": 41559025, "step": 1948, "time_per_iteration": 2.5523314476013184 }, { "auxiliary_loss_clip": 0.01198615, "auxiliary_loss_mlp": 0.01030822, "balance_loss_clip": 1.05885231, "balance_loss_mlp": 1.02147055, "epoch": 0.2343533938555883, "flos": 23623156470720.0, "grad_norm": 2.070144878120189, "language_loss": 0.69903231, "learning_rate": 3.5778528322862646e-06, "loss": 0.72132671, "num_input_tokens_seen": 41577845, "step": 1949, "time_per_iteration": 2.5923609733581543 }, { "auxiliary_loss_clip": 0.01202598, "auxiliary_loss_mlp": 0.01029537, "balance_loss_clip": 1.06021333, "balance_loss_mlp": 1.02033412, "epoch": 0.23447363674622737, "flos": 24570336111360.0, "grad_norm": 1.4807703012658446, "language_loss": 0.86721176, "learning_rate": 3.5773740441378585e-06, "loss": 0.88953304, "num_input_tokens_seen": 41598600, "step": 1950, "time_per_iteration": 2.581263542175293 }, { "auxiliary_loss_clip": 0.01197067, "auxiliary_loss_mlp": 0.01030075, "balance_loss_clip": 1.06112409, "balance_loss_mlp": 1.02159929, "epoch": 0.23459387963686648, "flos": 53140328678400.0, "grad_norm": 1.7795961249067638, "language_loss": 0.7399286, "learning_rate": 3.5768950167030633e-06, "loss": 0.7622, "num_input_tokens_seen": 41623300, "step": 1951, "time_per_iteration": 2.8676235675811768 }, { "auxiliary_loss_clip": 0.01170103, "auxiliary_loss_mlp": 0.01032982, "balance_loss_clip": 1.05271995, "balance_loss_mlp": 1.02382088, "epoch": 0.23471412252750556, "flos": 23951415907200.0, "grad_norm": 1.6959320050470283, "language_loss": 0.78336275, "learning_rate": 3.576415750054548e-06, "loss": 0.80539358, "num_input_tokens_seen": 41643420, "step": 1952, "time_per_iteration": 2.61751127243042 }, { "auxiliary_loss_clip": 0.0117448, "auxiliary_loss_mlp": 0.01031428, "balance_loss_clip": 1.05617356, "balance_loss_mlp": 1.02304721, "epoch": 0.23483436541814465, "flos": 15706577070240.0, "grad_norm": 1.8348085700269212, "language_loss": 0.85696721, "learning_rate": 3.5759362442650172e-06, "loss": 0.87902629, "num_input_tokens_seen": 41660170, "step": 1953, "time_per_iteration": 2.574432134628296 }, { "auxiliary_loss_clip": 0.0119694, "auxiliary_loss_mlp": 0.01032436, "balance_loss_clip": 1.05978751, "balance_loss_mlp": 1.02347147, "epoch": 0.23495460830878373, "flos": 24936266050560.0, "grad_norm": 2.13669071631956, "language_loss": 0.85426366, "learning_rate": 3.5754564994072113e-06, "loss": 0.87655747, "num_input_tokens_seen": 41679010, "step": 1954, "time_per_iteration": 2.5692942142486572 }, { "auxiliary_loss_clip": 0.01180994, "auxiliary_loss_mlp": 0.01030544, "balance_loss_clip": 1.05743897, "balance_loss_mlp": 1.02181172, "epoch": 0.23507485119942284, "flos": 30482669497920.0, "grad_norm": 2.7880306081770745, "language_loss": 0.59755301, "learning_rate": 3.5749765155539067e-06, "loss": 0.61966836, "num_input_tokens_seen": 41699495, "step": 1955, "time_per_iteration": 2.6943399906158447 }, { "auxiliary_loss_clip": 0.01169046, "auxiliary_loss_mlp": 0.01030949, "balance_loss_clip": 1.05756593, "balance_loss_mlp": 1.02203226, "epoch": 0.23519509409006192, "flos": 18329132700000.0, "grad_norm": 2.2767299752106056, "language_loss": 0.92365247, "learning_rate": 3.574496292777917e-06, "loss": 0.94565248, "num_input_tokens_seen": 41717705, "step": 1956, "time_per_iteration": 2.6162195205688477 }, { "auxiliary_loss_clip": 0.01194066, "auxiliary_loss_mlp": 0.01039464, "balance_loss_clip": 1.06213164, "balance_loss_mlp": 1.02983737, "epoch": 0.235315336980701, "flos": 29643221572320.0, "grad_norm": 1.89547059915367, "language_loss": 0.7142241, "learning_rate": 3.574015831152092e-06, "loss": 0.73655939, "num_input_tokens_seen": 41738120, "step": 1957, "time_per_iteration": 2.6896555423736572 }, { "auxiliary_loss_clip": 0.01173646, "auxiliary_loss_mlp": 0.01035877, "balance_loss_clip": 1.05790472, "balance_loss_mlp": 1.02669764, "epoch": 0.23543557987134012, "flos": 18551707688160.0, "grad_norm": 2.1480573343659155, "language_loss": 0.83413339, "learning_rate": 3.573535130749316e-06, "loss": 0.85622859, "num_input_tokens_seen": 41756070, "step": 1958, "time_per_iteration": 2.566638469696045 }, { "auxiliary_loss_clip": 0.0117596, "auxiliary_loss_mlp": 0.01045383, "balance_loss_clip": 1.05938959, "balance_loss_mlp": 1.03597713, "epoch": 0.2355558227619792, "flos": 24679037584320.0, "grad_norm": 1.6479436934906082, "language_loss": 0.7369321, "learning_rate": 3.5730541916425127e-06, "loss": 0.7591455, "num_input_tokens_seen": 41777550, "step": 1959, "time_per_iteration": 2.649810552597046 }, { "auxiliary_loss_clip": 0.01170424, "auxiliary_loss_mlp": 0.01031351, "balance_loss_clip": 1.05842912, "balance_loss_mlp": 1.02239275, "epoch": 0.23567606565261828, "flos": 21944799373920.0, "grad_norm": 1.7712695006134018, "language_loss": 0.8577559, "learning_rate": 3.572573013904639e-06, "loss": 0.87977362, "num_input_tokens_seen": 41797460, "step": 1960, "time_per_iteration": 2.623741626739502 }, { "auxiliary_loss_clip": 0.01213061, "auxiliary_loss_mlp": 0.01028879, "balance_loss_clip": 1.06188393, "balance_loss_mlp": 1.01992083, "epoch": 0.2357963085432574, "flos": 13589355465120.0, "grad_norm": 2.08110625573721, "language_loss": 0.91822946, "learning_rate": 3.572091597608689e-06, "loss": 0.94064885, "num_input_tokens_seen": 41815585, "step": 1961, "time_per_iteration": 2.5389702320098877 }, { "auxiliary_loss_clip": 0.01188905, "auxiliary_loss_mlp": 0.01036679, "balance_loss_clip": 1.05779791, "balance_loss_mlp": 1.02686787, "epoch": 0.23591655143389648, "flos": 22088693079360.0, "grad_norm": 2.1898777239211844, "language_loss": 0.73499954, "learning_rate": 3.571609942827694e-06, "loss": 0.75725532, "num_input_tokens_seen": 41834700, "step": 1962, "time_per_iteration": 2.5968563556671143 }, { "auxiliary_loss_clip": 0.0117937, "auxiliary_loss_mlp": 0.01036727, "balance_loss_clip": 1.05724466, "balance_loss_mlp": 1.02748263, "epoch": 0.23603679432453556, "flos": 17017352047680.0, "grad_norm": 1.7023043207798019, "language_loss": 0.88608932, "learning_rate": 3.57112804963472e-06, "loss": 0.90825027, "num_input_tokens_seen": 41852915, "step": 1963, "time_per_iteration": 2.588135242462158 }, { "auxiliary_loss_clip": 0.01163815, "auxiliary_loss_mlp": 0.0103133, "balance_loss_clip": 1.05819106, "balance_loss_mlp": 1.0222342, "epoch": 0.23615703721517464, "flos": 19171310314560.0, "grad_norm": 1.7525168495462229, "language_loss": 0.76057661, "learning_rate": 3.57064591810287e-06, "loss": 0.78252804, "num_input_tokens_seen": 41870415, "step": 1964, "time_per_iteration": 2.583162546157837 }, { "auxiliary_loss_clip": 0.01215482, "auxiliary_loss_mlp": 0.00763703, "balance_loss_clip": 1.06439579, "balance_loss_mlp": 1.00036705, "epoch": 0.23627728010581375, "flos": 19098806748960.0, "grad_norm": 2.1363973885366483, "language_loss": 0.80371213, "learning_rate": 3.570163548305284e-06, "loss": 0.82350391, "num_input_tokens_seen": 41889345, "step": 1965, "time_per_iteration": 3.297057628631592 }, { "auxiliary_loss_clip": 0.01182173, "auxiliary_loss_mlp": 0.01029466, "balance_loss_clip": 1.05871594, "balance_loss_mlp": 1.02006638, "epoch": 0.23639752299645284, "flos": 14282215913280.0, "grad_norm": 2.0708935104063784, "language_loss": 0.69681287, "learning_rate": 3.569680940315135e-06, "loss": 0.71892923, "num_input_tokens_seen": 41905745, "step": 1966, "time_per_iteration": 2.581327438354492 }, { "auxiliary_loss_clip": 0.01173412, "auxiliary_loss_mlp": 0.01031, "balance_loss_clip": 1.05674314, "balance_loss_mlp": 1.02204704, "epoch": 0.23651776588709192, "flos": 22893415692960.0, "grad_norm": 1.8027795144732124, "language_loss": 0.81943095, "learning_rate": 3.5691980942056356e-06, "loss": 0.84147501, "num_input_tokens_seen": 41925115, "step": 1967, "time_per_iteration": 3.386340856552124 }, { "auxiliary_loss_clip": 0.01201784, "auxiliary_loss_mlp": 0.01028149, "balance_loss_clip": 1.05860996, "balance_loss_mlp": 1.01929748, "epoch": 0.23663800877773103, "flos": 18624534506400.0, "grad_norm": 1.6658265095825957, "language_loss": 0.79451907, "learning_rate": 3.5687150100500332e-06, "loss": 0.81681842, "num_input_tokens_seen": 41944815, "step": 1968, "time_per_iteration": 2.577725887298584 }, { "auxiliary_loss_clip": 0.01201198, "auxiliary_loss_mlp": 0.01035818, "balance_loss_clip": 1.06018555, "balance_loss_mlp": 1.02697253, "epoch": 0.2367582516683701, "flos": 25555832760000.0, "grad_norm": 1.952527366987107, "language_loss": 0.74367058, "learning_rate": 3.568231687921611e-06, "loss": 0.76604068, "num_input_tokens_seen": 41964990, "step": 1969, "time_per_iteration": 2.6042182445526123 }, { "auxiliary_loss_clip": 0.0121241, "auxiliary_loss_mlp": 0.01038718, "balance_loss_clip": 1.06366825, "balance_loss_mlp": 1.03020072, "epoch": 0.2368784945590092, "flos": 23295076619040.0, "grad_norm": 1.7303087371833274, "language_loss": 0.80517483, "learning_rate": 3.5677481278936883e-06, "loss": 0.82768607, "num_input_tokens_seen": 41984570, "step": 1970, "time_per_iteration": 3.4225587844848633 }, { "auxiliary_loss_clip": 0.01074589, "auxiliary_loss_mlp": 0.01000267, "balance_loss_clip": 1.01686454, "balance_loss_mlp": 0.99840766, "epoch": 0.23699873744964828, "flos": 69859304200320.0, "grad_norm": 0.8259492559196989, "language_loss": 0.57813811, "learning_rate": 3.5672643300396214e-06, "loss": 0.59888661, "num_input_tokens_seen": 42053715, "step": 1971, "time_per_iteration": 3.2238588333129883 }, { "auxiliary_loss_clip": 0.01164795, "auxiliary_loss_mlp": 0.0102994, "balance_loss_clip": 1.05399704, "balance_loss_mlp": 1.02120781, "epoch": 0.2371189803402874, "flos": 21835056309120.0, "grad_norm": 2.166103664696092, "language_loss": 0.67506957, "learning_rate": 3.566780294432802e-06, "loss": 0.69701695, "num_input_tokens_seen": 42070890, "step": 1972, "time_per_iteration": 3.5154824256896973 }, { "auxiliary_loss_clip": 0.01213512, "auxiliary_loss_mlp": 0.01037546, "balance_loss_clip": 1.06220007, "balance_loss_mlp": 1.02913022, "epoch": 0.23723922323092647, "flos": 21908493715680.0, "grad_norm": 2.338263106681044, "language_loss": 0.7441799, "learning_rate": 3.566296021146657e-06, "loss": 0.76669049, "num_input_tokens_seen": 42090270, "step": 1973, "time_per_iteration": 2.552421808242798 }, { "auxiliary_loss_clip": 0.01217528, "auxiliary_loss_mlp": 0.01032738, "balance_loss_clip": 1.06411505, "balance_loss_mlp": 1.02349329, "epoch": 0.23735946612156555, "flos": 32708808077760.0, "grad_norm": 1.5925834241948587, "language_loss": 0.72937977, "learning_rate": 3.565811510254652e-06, "loss": 0.75188243, "num_input_tokens_seen": 42111150, "step": 1974, "time_per_iteration": 2.6112060546875 }, { "auxiliary_loss_clip": 0.0109129, "auxiliary_loss_mlp": 0.01007313, "balance_loss_clip": 1.02611637, "balance_loss_mlp": 1.00545335, "epoch": 0.23747970901220466, "flos": 70546956689280.0, "grad_norm": 0.8322042425936939, "language_loss": 0.58229935, "learning_rate": 3.5653267618302845e-06, "loss": 0.60328531, "num_input_tokens_seen": 42178730, "step": 1975, "time_per_iteration": 3.1861345767974854 }, { "auxiliary_loss_clip": 0.01211357, "auxiliary_loss_mlp": 0.01035514, "balance_loss_clip": 1.0601809, "balance_loss_mlp": 1.02647805, "epoch": 0.23759995190284375, "flos": 20849811079200.0, "grad_norm": 1.7790509936299967, "language_loss": 0.8586992, "learning_rate": 3.564841775947093e-06, "loss": 0.88116789, "num_input_tokens_seen": 42199620, "step": 1976, "time_per_iteration": 2.5669732093811035 }, { "auxiliary_loss_clip": 0.01166678, "auxiliary_loss_mlp": 0.01040186, "balance_loss_clip": 1.05423868, "balance_loss_mlp": 1.03176355, "epoch": 0.23772019479348283, "flos": 32921649569760.0, "grad_norm": 2.0977808532743913, "language_loss": 0.75928855, "learning_rate": 3.5643565526786475e-06, "loss": 0.78135717, "num_input_tokens_seen": 42219560, "step": 1977, "time_per_iteration": 2.718893527984619 }, { "auxiliary_loss_clip": 0.01214569, "auxiliary_loss_mlp": 0.01032963, "balance_loss_clip": 1.06274986, "balance_loss_mlp": 1.0242188, "epoch": 0.2378404376841219, "flos": 32342770387680.0, "grad_norm": 1.6054185740252347, "language_loss": 0.77379656, "learning_rate": 3.5638710920985574e-06, "loss": 0.7962718, "num_input_tokens_seen": 42241020, "step": 1978, "time_per_iteration": 2.724346399307251 }, { "auxiliary_loss_clip": 0.01203952, "auxiliary_loss_mlp": 0.00764301, "balance_loss_clip": 1.05922472, "balance_loss_mlp": 1.00027955, "epoch": 0.23796068057476102, "flos": 22997627545920.0, "grad_norm": 2.512208996149022, "language_loss": 0.82001519, "learning_rate": 3.5633853942804655e-06, "loss": 0.83969766, "num_input_tokens_seen": 42259345, "step": 1979, "time_per_iteration": 2.575051784515381 }, { "auxiliary_loss_clip": 0.01171019, "auxiliary_loss_mlp": 0.01027565, "balance_loss_clip": 1.05566502, "balance_loss_mlp": 1.01786113, "epoch": 0.2380809234654001, "flos": 13480941327840.0, "grad_norm": 2.3952832322441155, "language_loss": 0.76818818, "learning_rate": 3.5628994592980527e-06, "loss": 0.79017401, "num_input_tokens_seen": 42277250, "step": 1980, "time_per_iteration": 2.641367197036743 }, { "auxiliary_loss_clip": 0.01216055, "auxiliary_loss_mlp": 0.01035146, "balance_loss_clip": 1.06334662, "balance_loss_mlp": 1.02564502, "epoch": 0.2382011663560392, "flos": 16871806162080.0, "grad_norm": 1.8468291370313046, "language_loss": 0.70401657, "learning_rate": 3.562413287225034e-06, "loss": 0.72652853, "num_input_tokens_seen": 42295360, "step": 1981, "time_per_iteration": 2.51650071144104 }, { "auxiliary_loss_clip": 0.01193052, "auxiliary_loss_mlp": 0.01032953, "balance_loss_clip": 1.05964744, "balance_loss_mlp": 1.02400017, "epoch": 0.2383214092466783, "flos": 18441138533280.0, "grad_norm": 2.4517360432746114, "language_loss": 0.89209425, "learning_rate": 3.5619268781351623e-06, "loss": 0.91435432, "num_input_tokens_seen": 42313430, "step": 1982, "time_per_iteration": 2.561699390411377 }, { "auxiliary_loss_clip": 0.01180722, "auxiliary_loss_mlp": 0.01026671, "balance_loss_clip": 1.06087995, "balance_loss_mlp": 1.01794469, "epoch": 0.23844165213731738, "flos": 19755720708480.0, "grad_norm": 1.8020223964168123, "language_loss": 0.76725578, "learning_rate": 3.5614402321022256e-06, "loss": 0.78932977, "num_input_tokens_seen": 42331260, "step": 1983, "time_per_iteration": 2.576770782470703 }, { "auxiliary_loss_clip": 0.01142452, "auxiliary_loss_mlp": 0.01028816, "balance_loss_clip": 1.05190444, "balance_loss_mlp": 1.02019775, "epoch": 0.23856189502795647, "flos": 23367364682880.0, "grad_norm": 2.227474489305841, "language_loss": 0.87047869, "learning_rate": 3.5609533492000463e-06, "loss": 0.89219141, "num_input_tokens_seen": 42350150, "step": 1984, "time_per_iteration": 2.697490692138672 }, { "auxiliary_loss_clip": 0.01177654, "auxiliary_loss_mlp": 0.01032568, "balance_loss_clip": 1.05713308, "balance_loss_mlp": 1.02312672, "epoch": 0.23868213791859555, "flos": 23475060480960.0, "grad_norm": 1.9380401029308953, "language_loss": 0.78462273, "learning_rate": 3.560466229502485e-06, "loss": 0.80672503, "num_input_tokens_seen": 42369495, "step": 1985, "time_per_iteration": 2.601940631866455 }, { "auxiliary_loss_clip": 0.01180619, "auxiliary_loss_mlp": 0.00764128, "balance_loss_clip": 1.06150842, "balance_loss_mlp": 1.00032735, "epoch": 0.23880238080923466, "flos": 16617343301760.0, "grad_norm": 2.167393242179471, "language_loss": 0.89469069, "learning_rate": 3.5599788730834384e-06, "loss": 0.9141382, "num_input_tokens_seen": 42387455, "step": 1986, "time_per_iteration": 2.624553680419922 }, { "auxiliary_loss_clip": 0.01200869, "auxiliary_loss_mlp": 0.01031917, "balance_loss_clip": 1.05879569, "balance_loss_mlp": 1.02233219, "epoch": 0.23892262369987374, "flos": 17348413007040.0, "grad_norm": 2.8816167280547482, "language_loss": 0.785896, "learning_rate": 3.559491280016836e-06, "loss": 0.80822384, "num_input_tokens_seen": 42405400, "step": 1987, "time_per_iteration": 2.5535175800323486 }, { "auxiliary_loss_clip": 0.01185347, "auxiliary_loss_mlp": 0.01034748, "balance_loss_clip": 1.06043816, "balance_loss_mlp": 1.02532434, "epoch": 0.23904286659051283, "flos": 22309903223040.0, "grad_norm": 1.7477535029945253, "language_loss": 0.70781475, "learning_rate": 3.5590034503766465e-06, "loss": 0.73001564, "num_input_tokens_seen": 42425065, "step": 1988, "time_per_iteration": 2.624988079071045 }, { "auxiliary_loss_clip": 0.01215821, "auxiliary_loss_mlp": 0.01021147, "balance_loss_clip": 1.06422174, "balance_loss_mlp": 1.012784, "epoch": 0.23916310948115194, "flos": 21178250100480.0, "grad_norm": 2.268705391080057, "language_loss": 0.8179481, "learning_rate": 3.558515384236874e-06, "loss": 0.84031779, "num_input_tokens_seen": 42442495, "step": 1989, "time_per_iteration": 2.5334885120391846 }, { "auxiliary_loss_clip": 0.01156527, "auxiliary_loss_mlp": 0.00764246, "balance_loss_clip": 1.0556488, "balance_loss_mlp": 1.00038981, "epoch": 0.23928335237179102, "flos": 14137352449920.0, "grad_norm": 1.8206081428006853, "language_loss": 0.8372311, "learning_rate": 3.558027081671556e-06, "loss": 0.85643882, "num_input_tokens_seen": 42459480, "step": 1990, "time_per_iteration": 2.6124699115753174 }, { "auxiliary_loss_clip": 0.01201227, "auxiliary_loss_mlp": 0.01027228, "balance_loss_clip": 1.05984354, "balance_loss_mlp": 1.01825178, "epoch": 0.2394035952624301, "flos": 23769600280320.0, "grad_norm": 1.8943779418211608, "language_loss": 0.6883238, "learning_rate": 3.557538542754769e-06, "loss": 0.71060836, "num_input_tokens_seen": 42479175, "step": 1991, "time_per_iteration": 3.358929395675659 }, { "auxiliary_loss_clip": 0.01213726, "auxiliary_loss_mlp": 0.01030451, "balance_loss_clip": 1.06341839, "balance_loss_mlp": 1.02097392, "epoch": 0.2395238381530692, "flos": 24206202020160.0, "grad_norm": 1.860997055339057, "language_loss": 0.66530693, "learning_rate": 3.557049767560623e-06, "loss": 0.68774867, "num_input_tokens_seen": 42498090, "step": 1992, "time_per_iteration": 3.2763874530792236 }, { "auxiliary_loss_clip": 0.01160488, "auxiliary_loss_mlp": 0.01029569, "balance_loss_clip": 1.06127739, "balance_loss_mlp": 1.02050924, "epoch": 0.2396440810437083, "flos": 25295766853920.0, "grad_norm": 2.0655914032391705, "language_loss": 0.85798633, "learning_rate": 3.5565607561632655e-06, "loss": 0.87988687, "num_input_tokens_seen": 42516930, "step": 1993, "time_per_iteration": 2.7155702114105225 }, { "auxiliary_loss_clip": 0.01180972, "auxiliary_loss_mlp": 0.01025777, "balance_loss_clip": 1.05788088, "balance_loss_mlp": 1.01614535, "epoch": 0.23976432393434738, "flos": 28543097152320.0, "grad_norm": 2.13957518253739, "language_loss": 0.79062444, "learning_rate": 3.5560715086368787e-06, "loss": 0.81269199, "num_input_tokens_seen": 42534800, "step": 1994, "time_per_iteration": 2.6454036235809326 }, { "auxiliary_loss_clip": 0.01178292, "auxiliary_loss_mlp": 0.01036423, "balance_loss_clip": 1.05739629, "balance_loss_mlp": 1.0271064, "epoch": 0.23988456682498646, "flos": 19494361791840.0, "grad_norm": 2.185222769582013, "language_loss": 0.82047105, "learning_rate": 3.5555820250556816e-06, "loss": 0.84261823, "num_input_tokens_seen": 42552000, "step": 1995, "time_per_iteration": 2.5417380332946777 }, { "auxiliary_loss_clip": 0.01188841, "auxiliary_loss_mlp": 0.01033788, "balance_loss_clip": 1.06087935, "balance_loss_mlp": 1.02436459, "epoch": 0.24000480971562557, "flos": 20266334526240.0, "grad_norm": 3.2408733480677254, "language_loss": 0.69131929, "learning_rate": 3.5550923054939278e-06, "loss": 0.71354556, "num_input_tokens_seen": 42571455, "step": 1996, "time_per_iteration": 3.3497040271759033 }, { "auxiliary_loss_clip": 0.01144557, "auxiliary_loss_mlp": 0.01028763, "balance_loss_clip": 1.04999268, "balance_loss_mlp": 1.02007818, "epoch": 0.24012505260626466, "flos": 25443180421440.0, "grad_norm": 2.0051166602693784, "language_loss": 0.74078858, "learning_rate": 3.5546023500259083e-06, "loss": 0.76252174, "num_input_tokens_seen": 42592550, "step": 1997, "time_per_iteration": 2.660399913787842 }, { "auxiliary_loss_clip": 0.01157692, "auxiliary_loss_mlp": 0.01030113, "balance_loss_clip": 1.05347812, "balance_loss_mlp": 1.02125573, "epoch": 0.24024529549690374, "flos": 15553344955200.0, "grad_norm": 2.205505568789749, "language_loss": 0.80828518, "learning_rate": 3.5541121587259477e-06, "loss": 0.83016324, "num_input_tokens_seen": 42610385, "step": 1998, "time_per_iteration": 3.5066778659820557 }, { "auxiliary_loss_clip": 0.01084864, "auxiliary_loss_mlp": 0.01009235, "balance_loss_clip": 1.01812387, "balance_loss_mlp": 1.00739884, "epoch": 0.24036553838754285, "flos": 57122362683360.0, "grad_norm": 0.832946148533733, "language_loss": 0.57951754, "learning_rate": 3.553621731668408e-06, "loss": 0.6004585, "num_input_tokens_seen": 42673595, "step": 1999, "time_per_iteration": 3.1167495250701904 }, { "auxiliary_loss_clip": 0.01190061, "auxiliary_loss_mlp": 0.01034346, "balance_loss_clip": 1.05700445, "balance_loss_mlp": 1.02527976, "epoch": 0.24048578127818193, "flos": 24969949770720.0, "grad_norm": 1.7196248997678778, "language_loss": 0.83511817, "learning_rate": 3.553131068927688e-06, "loss": 0.85736227, "num_input_tokens_seen": 42692000, "step": 2000, "time_per_iteration": 2.6144278049468994 }, { "auxiliary_loss_clip": 0.01169302, "auxiliary_loss_mlp": 0.01029345, "balance_loss_clip": 1.05905616, "balance_loss_mlp": 1.02105451, "epoch": 0.24060602416882101, "flos": 23330951273760.0, "grad_norm": 1.7486728799941849, "language_loss": 0.80284846, "learning_rate": 3.552640170578219e-06, "loss": 0.82483494, "num_input_tokens_seen": 42712250, "step": 2001, "time_per_iteration": 2.6461551189422607 }, { "auxiliary_loss_clip": 0.01183701, "auxiliary_loss_mlp": 0.0103171, "balance_loss_clip": 1.05887079, "balance_loss_mlp": 1.02292478, "epoch": 0.2407262670594601, "flos": 14173263021600.0, "grad_norm": 2.8543913917672326, "language_loss": 0.77984917, "learning_rate": 3.5521490366944703e-06, "loss": 0.80200326, "num_input_tokens_seen": 42729900, "step": 2002, "time_per_iteration": 2.5950112342834473 }, { "auxiliary_loss_clip": 0.01167383, "auxiliary_loss_mlp": 0.01036222, "balance_loss_clip": 1.05677366, "balance_loss_mlp": 1.02689946, "epoch": 0.2408465099500992, "flos": 13663116124320.0, "grad_norm": 2.0524199373643057, "language_loss": 0.80479133, "learning_rate": 3.5516576673509474e-06, "loss": 0.82682735, "num_input_tokens_seen": 42747900, "step": 2003, "time_per_iteration": 2.613477945327759 }, { "auxiliary_loss_clip": 0.01212451, "auxiliary_loss_mlp": 0.01031574, "balance_loss_clip": 1.06053698, "balance_loss_mlp": 1.02175725, "epoch": 0.2409667528407383, "flos": 31248033511680.0, "grad_norm": 1.6284442525630964, "language_loss": 0.8623684, "learning_rate": 3.5511660626221896e-06, "loss": 0.8848086, "num_input_tokens_seen": 42768540, "step": 2004, "time_per_iteration": 2.6366732120513916 }, { "auxiliary_loss_clip": 0.01183269, "auxiliary_loss_mlp": 0.00764589, "balance_loss_clip": 1.05930519, "balance_loss_mlp": 1.00043726, "epoch": 0.24108699573137737, "flos": 22199944656480.0, "grad_norm": 4.447554588220294, "language_loss": 0.89148623, "learning_rate": 3.5506742225827744e-06, "loss": 0.91096485, "num_input_tokens_seen": 42785395, "step": 2005, "time_per_iteration": 2.5955722332000732 }, { "auxiliary_loss_clip": 0.0116511, "auxiliary_loss_mlp": 0.01033857, "balance_loss_clip": 1.05594182, "balance_loss_mlp": 1.02430797, "epoch": 0.24120723862201648, "flos": 26103039571680.0, "grad_norm": 2.4139109157211576, "language_loss": 0.89802539, "learning_rate": 3.5501821473073116e-06, "loss": 0.92001504, "num_input_tokens_seen": 42801980, "step": 2006, "time_per_iteration": 2.6650006771087646 }, { "auxiliary_loss_clip": 0.01159989, "auxiliary_loss_mlp": 0.01037678, "balance_loss_clip": 1.05476236, "balance_loss_mlp": 1.02877903, "epoch": 0.24132748151265557, "flos": 18624929592960.0, "grad_norm": 2.1220220518338233, "language_loss": 0.86548805, "learning_rate": 3.54968983687045e-06, "loss": 0.88746476, "num_input_tokens_seen": 42818850, "step": 2007, "time_per_iteration": 2.6399970054626465 }, { "auxiliary_loss_clip": 0.01183761, "auxiliary_loss_mlp": 0.01036973, "balance_loss_clip": 1.05745685, "balance_loss_mlp": 1.0280447, "epoch": 0.24144772440329465, "flos": 15267676644960.0, "grad_norm": 2.826646041645952, "language_loss": 0.89670122, "learning_rate": 3.549197291346872e-06, "loss": 0.9189086, "num_input_tokens_seen": 42835375, "step": 2008, "time_per_iteration": 2.5947484970092773 }, { "auxiliary_loss_clip": 0.0119945, "auxiliary_loss_mlp": 0.01036885, "balance_loss_clip": 1.06052542, "balance_loss_mlp": 1.02781296, "epoch": 0.24156796729393373, "flos": 24024278642400.0, "grad_norm": 2.0065986835525256, "language_loss": 0.79618961, "learning_rate": 3.548704510811297e-06, "loss": 0.81855297, "num_input_tokens_seen": 42854570, "step": 2009, "time_per_iteration": 2.6231231689453125 }, { "auxiliary_loss_clip": 0.01159097, "auxiliary_loss_mlp": 0.01031745, "balance_loss_clip": 1.05439913, "balance_loss_mlp": 1.02207136, "epoch": 0.24168821018457284, "flos": 26286794714400.0, "grad_norm": 2.700642894494451, "language_loss": 0.74511874, "learning_rate": 3.5482114953384787e-06, "loss": 0.76702714, "num_input_tokens_seen": 42873800, "step": 2010, "time_per_iteration": 2.6840665340423584 }, { "auxiliary_loss_clip": 0.01202094, "auxiliary_loss_mlp": 0.01038255, "balance_loss_clip": 1.06077576, "balance_loss_mlp": 1.02921939, "epoch": 0.24180845307521193, "flos": 18223196832960.0, "grad_norm": 2.043893169474589, "language_loss": 0.84221953, "learning_rate": 3.5477182450032077e-06, "loss": 0.86462301, "num_input_tokens_seen": 42892400, "step": 2011, "time_per_iteration": 2.57647967338562 }, { "auxiliary_loss_clip": 0.01196638, "auxiliary_loss_mlp": 0.01036458, "balance_loss_clip": 1.0597477, "balance_loss_mlp": 1.0275774, "epoch": 0.241928695965851, "flos": 20449263578880.0, "grad_norm": 2.3236936752064232, "language_loss": 0.8298344, "learning_rate": 3.5472247598803097e-06, "loss": 0.8521654, "num_input_tokens_seen": 42911745, "step": 2012, "time_per_iteration": 2.558157444000244 }, { "auxiliary_loss_clip": 0.01213579, "auxiliary_loss_mlp": 0.0103829, "balance_loss_clip": 1.06184304, "balance_loss_mlp": 1.02844977, "epoch": 0.24204893885649012, "flos": 25556479265280.0, "grad_norm": 2.489854983168211, "language_loss": 0.85500544, "learning_rate": 3.546731040044645e-06, "loss": 0.87752414, "num_input_tokens_seen": 42926915, "step": 2013, "time_per_iteration": 2.563832998275757 }, { "auxiliary_loss_clip": 0.01212583, "auxiliary_loss_mlp": 0.01029068, "balance_loss_clip": 1.06169009, "balance_loss_mlp": 1.02050328, "epoch": 0.2421691817471292, "flos": 30660211006560.0, "grad_norm": 1.817403386109655, "language_loss": 0.75460279, "learning_rate": 3.546237085571112e-06, "loss": 0.77701926, "num_input_tokens_seen": 42945350, "step": 2014, "time_per_iteration": 2.5998313426971436 }, { "auxiliary_loss_clip": 0.0120085, "auxiliary_loss_mlp": 0.01037206, "balance_loss_clip": 1.06281745, "balance_loss_mlp": 1.02850389, "epoch": 0.24228942463776829, "flos": 21945014875680.0, "grad_norm": 2.063622897665687, "language_loss": 0.72174609, "learning_rate": 3.5457428965346425e-06, "loss": 0.74412668, "num_input_tokens_seen": 42964290, "step": 2015, "time_per_iteration": 2.6001241207122803 }, { "auxiliary_loss_clip": 0.01136816, "auxiliary_loss_mlp": 0.01038629, "balance_loss_clip": 1.05218351, "balance_loss_mlp": 1.02946162, "epoch": 0.2424096675284074, "flos": 33984498573600.0, "grad_norm": 1.616976089819781, "language_loss": 0.74628246, "learning_rate": 3.545248473010205e-06, "loss": 0.7680369, "num_input_tokens_seen": 42987095, "step": 2016, "time_per_iteration": 2.79744815826416 }, { "auxiliary_loss_clip": 0.01216026, "auxiliary_loss_mlp": 0.00764547, "balance_loss_clip": 1.06096315, "balance_loss_mlp": 1.0002054, "epoch": 0.24252991041904648, "flos": 21653420267040.0, "grad_norm": 2.3091876714994917, "language_loss": 0.87578142, "learning_rate": 3.544753815072802e-06, "loss": 0.89558715, "num_input_tokens_seen": 43005750, "step": 2017, "time_per_iteration": 2.558704137802124 }, { "auxiliary_loss_clip": 0.01111199, "auxiliary_loss_mlp": 0.01033638, "balance_loss_clip": 1.0436492, "balance_loss_mlp": 1.02443504, "epoch": 0.24265015330968556, "flos": 21870068956800.0, "grad_norm": 1.894506985957687, "language_loss": 0.88284171, "learning_rate": 3.544258922797474e-06, "loss": 0.90429002, "num_input_tokens_seen": 43023870, "step": 2018, "time_per_iteration": 3.5594675540924072 }, { "auxiliary_loss_clip": 0.01213403, "auxiliary_loss_mlp": 0.01034807, "balance_loss_clip": 1.06233442, "balance_loss_mlp": 1.02589655, "epoch": 0.24277039620032465, "flos": 25628264491680.0, "grad_norm": 1.588151440311033, "language_loss": 0.77962625, "learning_rate": 3.543763796259295e-06, "loss": 0.80210829, "num_input_tokens_seen": 43043825, "step": 2019, "time_per_iteration": 2.596151113510132 }, { "auxiliary_loss_clip": 0.01196233, "auxiliary_loss_mlp": 0.01028312, "balance_loss_clip": 1.06009638, "balance_loss_mlp": 1.01940751, "epoch": 0.24289063909096376, "flos": 26286579212640.0, "grad_norm": 1.8106280246223139, "language_loss": 0.90801197, "learning_rate": 3.5432684355333754e-06, "loss": 0.93025744, "num_input_tokens_seen": 43062480, "step": 2020, "time_per_iteration": 2.612396478652954 }, { "auxiliary_loss_clip": 0.01197735, "auxiliary_loss_mlp": 0.01029, "balance_loss_clip": 1.05842996, "balance_loss_mlp": 1.02019668, "epoch": 0.24301088198160284, "flos": 25075059547680.0, "grad_norm": 1.9793451285320287, "language_loss": 0.76766598, "learning_rate": 3.5427728406948613e-06, "loss": 0.78993332, "num_input_tokens_seen": 43081595, "step": 2021, "time_per_iteration": 3.5069453716278076 }, { "auxiliary_loss_clip": 0.0107294, "auxiliary_loss_mlp": 0.01002813, "balance_loss_clip": 1.01189852, "balance_loss_mlp": 1.00107288, "epoch": 0.24313112487224192, "flos": 69900961485600.0, "grad_norm": 0.7588911951531586, "language_loss": 0.57888633, "learning_rate": 3.542277011818934e-06, "loss": 0.59964383, "num_input_tokens_seen": 43145430, "step": 2022, "time_per_iteration": 3.3212130069732666 }, { "auxiliary_loss_clip": 0.01184626, "auxiliary_loss_mlp": 0.0103072, "balance_loss_clip": 1.06103086, "balance_loss_mlp": 1.02231622, "epoch": 0.24325136776288103, "flos": 40662339641760.0, "grad_norm": 1.9064527082133724, "language_loss": 0.74086225, "learning_rate": 3.5417809489808104e-06, "loss": 0.76301569, "num_input_tokens_seen": 43167040, "step": 2023, "time_per_iteration": 2.7385671138763428 }, { "auxiliary_loss_clip": 0.01201481, "auxiliary_loss_mlp": 0.01037088, "balance_loss_clip": 1.06233263, "balance_loss_mlp": 1.02824283, "epoch": 0.24337161065352012, "flos": 25046404201920.0, "grad_norm": 1.8670681568757956, "language_loss": 0.72079599, "learning_rate": 3.5412846522557422e-06, "loss": 0.74318171, "num_input_tokens_seen": 43187930, "step": 2024, "time_per_iteration": 2.622413396835327 }, { "auxiliary_loss_clip": 0.01212232, "auxiliary_loss_mlp": 0.01030824, "balance_loss_clip": 1.06223845, "balance_loss_mlp": 1.02232456, "epoch": 0.2434918535441592, "flos": 18661163417280.0, "grad_norm": 2.4474747086141133, "language_loss": 0.7383948, "learning_rate": 3.540788121719018e-06, "loss": 0.76082534, "num_input_tokens_seen": 43206350, "step": 2025, "time_per_iteration": 3.4186997413635254 }, { "auxiliary_loss_clip": 0.01161483, "auxiliary_loss_mlp": 0.01029987, "balance_loss_clip": 1.05927014, "balance_loss_mlp": 1.02154636, "epoch": 0.24361209643479828, "flos": 23915146165920.0, "grad_norm": 1.8874922042210498, "language_loss": 0.8218199, "learning_rate": 3.5402913574459604e-06, "loss": 0.84373456, "num_input_tokens_seen": 43226255, "step": 2026, "time_per_iteration": 2.6300435066223145 }, { "auxiliary_loss_clip": 0.01131902, "auxiliary_loss_mlp": 0.01037004, "balance_loss_clip": 1.04808927, "balance_loss_mlp": 1.02867699, "epoch": 0.2437323393254374, "flos": 28657509421920.0, "grad_norm": 2.1474102081825084, "language_loss": 0.85734737, "learning_rate": 3.5397943595119297e-06, "loss": 0.87903643, "num_input_tokens_seen": 43247675, "step": 2027, "time_per_iteration": 2.7693846225738525 }, { "auxiliary_loss_clip": 0.01178833, "auxiliary_loss_mlp": 0.01032436, "balance_loss_clip": 1.05950475, "balance_loss_mlp": 1.02315569, "epoch": 0.24385258221607647, "flos": 23550329652480.0, "grad_norm": 2.6767747060117117, "language_loss": 0.77377808, "learning_rate": 3.5392971279923177e-06, "loss": 0.79589081, "num_input_tokens_seen": 43265895, "step": 2028, "time_per_iteration": 2.5863234996795654 }, { "auxiliary_loss_clip": 0.01162525, "auxiliary_loss_mlp": 0.01035533, "balance_loss_clip": 1.05421233, "balance_loss_mlp": 1.02653313, "epoch": 0.24397282510671556, "flos": 25336095211680.0, "grad_norm": 2.0577016360022142, "language_loss": 0.8288362, "learning_rate": 3.5387996629625557e-06, "loss": 0.85081685, "num_input_tokens_seen": 43283485, "step": 2029, "time_per_iteration": 2.6613457202911377 }, { "auxiliary_loss_clip": 0.01096974, "auxiliary_loss_mlp": 0.00999916, "balance_loss_clip": 1.01440668, "balance_loss_mlp": 0.99810427, "epoch": 0.24409306799735467, "flos": 65187433160160.0, "grad_norm": 0.8097511785554095, "language_loss": 0.55030012, "learning_rate": 3.5383019644981083e-06, "loss": 0.57126904, "num_input_tokens_seen": 43347180, "step": 2030, "time_per_iteration": 3.1398468017578125 }, { "auxiliary_loss_clip": 0.01183152, "auxiliary_loss_mlp": 0.01034585, "balance_loss_clip": 1.0597688, "balance_loss_mlp": 1.02519107, "epoch": 0.24421331088799375, "flos": 19537096585920.0, "grad_norm": 2.393295835748951, "language_loss": 0.73339349, "learning_rate": 3.5378040326744763e-06, "loss": 0.75557089, "num_input_tokens_seen": 43366665, "step": 2031, "time_per_iteration": 2.6139631271362305 }, { "auxiliary_loss_clip": 0.01168891, "auxiliary_loss_mlp": 0.01033289, "balance_loss_clip": 1.06008959, "balance_loss_mlp": 1.02515912, "epoch": 0.24433355377863283, "flos": 21068578869600.0, "grad_norm": 2.0022435532643796, "language_loss": 0.85654545, "learning_rate": 3.5373058675671946e-06, "loss": 0.87856722, "num_input_tokens_seen": 43384670, "step": 2032, "time_per_iteration": 2.6332550048828125 }, { "auxiliary_loss_clip": 0.01144719, "auxiliary_loss_mlp": 0.0103834, "balance_loss_clip": 1.05290079, "balance_loss_mlp": 1.02938187, "epoch": 0.24445379666927192, "flos": 22637192901600.0, "grad_norm": 1.9296185730336428, "language_loss": 0.72140658, "learning_rate": 3.536807469251836e-06, "loss": 0.74323714, "num_input_tokens_seen": 43403825, "step": 2033, "time_per_iteration": 2.6876471042633057 }, { "auxiliary_loss_clip": 0.01172648, "auxiliary_loss_mlp": 0.01030557, "balance_loss_clip": 1.05558181, "balance_loss_mlp": 1.02177083, "epoch": 0.24457403955991103, "flos": 21251615673120.0, "grad_norm": 1.8712238700691524, "language_loss": 0.83004749, "learning_rate": 3.5363088378040055e-06, "loss": 0.85207951, "num_input_tokens_seen": 43422715, "step": 2034, "time_per_iteration": 2.6584086418151855 }, { "auxiliary_loss_clip": 0.01098081, "auxiliary_loss_mlp": 0.00755053, "balance_loss_clip": 1.01585567, "balance_loss_mlp": 0.99983019, "epoch": 0.2446942824505501, "flos": 66997831365600.0, "grad_norm": 0.7598230118152768, "language_loss": 0.64413214, "learning_rate": 3.5358099732993463e-06, "loss": 0.66266346, "num_input_tokens_seen": 43481825, "step": 2035, "time_per_iteration": 3.038926839828491 }, { "auxiliary_loss_clip": 0.01188865, "auxiliary_loss_mlp": 0.0103735, "balance_loss_clip": 1.05913734, "balance_loss_mlp": 1.02848625, "epoch": 0.2448145253411892, "flos": 20411126155680.0, "grad_norm": 1.8906254285608195, "language_loss": 0.89773685, "learning_rate": 3.535310875813535e-06, "loss": 0.919999, "num_input_tokens_seen": 43500220, "step": 2036, "time_per_iteration": 2.6189284324645996 }, { "auxiliary_loss_clip": 0.01194981, "auxiliary_loss_mlp": 0.01034951, "balance_loss_clip": 1.06016946, "balance_loss_mlp": 1.0265944, "epoch": 0.2449347682318283, "flos": 28804743404640.0, "grad_norm": 3.20606875282527, "language_loss": 0.81484866, "learning_rate": 3.5348115454222843e-06, "loss": 0.83714795, "num_input_tokens_seen": 43522805, "step": 2037, "time_per_iteration": 2.6143555641174316 }, { "auxiliary_loss_clip": 0.01176532, "auxiliary_loss_mlp": 0.01040239, "balance_loss_clip": 1.05540931, "balance_loss_mlp": 1.03113103, "epoch": 0.2450550111224674, "flos": 22528994266080.0, "grad_norm": 1.8299108114957103, "language_loss": 0.86409807, "learning_rate": 3.5343119822013425e-06, "loss": 0.88626575, "num_input_tokens_seen": 43541915, "step": 2038, "time_per_iteration": 2.6144821643829346 }, { "auxiliary_loss_clip": 0.01203563, "auxiliary_loss_mlp": 0.01037165, "balance_loss_clip": 1.06073618, "balance_loss_mlp": 1.02728295, "epoch": 0.24517525401310647, "flos": 21759140632320.0, "grad_norm": 4.167029219735465, "language_loss": 0.77852321, "learning_rate": 3.533812186226493e-06, "loss": 0.80093056, "num_input_tokens_seen": 43562625, "step": 2039, "time_per_iteration": 2.603196144104004 }, { "auxiliary_loss_clip": 0.01207795, "auxiliary_loss_mlp": 0.01029462, "balance_loss_clip": 1.05988455, "balance_loss_mlp": 1.02168322, "epoch": 0.24529549690374555, "flos": 25043315343360.0, "grad_norm": 1.7068980562076257, "language_loss": 0.75823683, "learning_rate": 3.5333121575735545e-06, "loss": 0.78060937, "num_input_tokens_seen": 43582265, "step": 2040, "time_per_iteration": 2.594317674636841 }, { "auxiliary_loss_clip": 0.01180913, "auxiliary_loss_mlp": 0.0103707, "balance_loss_clip": 1.05975676, "balance_loss_mlp": 1.02870786, "epoch": 0.24541573979438466, "flos": 32123643427680.0, "grad_norm": 2.469367176589629, "language_loss": 0.75241077, "learning_rate": 3.532811896318381e-06, "loss": 0.77459061, "num_input_tokens_seen": 43604335, "step": 2041, "time_per_iteration": 2.6961569786071777 }, { "auxiliary_loss_clip": 0.01170991, "auxiliary_loss_mlp": 0.01038572, "balance_loss_clip": 1.05741382, "balance_loss_mlp": 1.02881503, "epoch": 0.24553598268502375, "flos": 31357561074720.0, "grad_norm": 2.5662860311568574, "language_loss": 0.81593925, "learning_rate": 3.5323114025368615e-06, "loss": 0.83803487, "num_input_tokens_seen": 43619400, "step": 2042, "time_per_iteration": 2.7145280838012695 }, { "auxiliary_loss_clip": 0.01192234, "auxiliary_loss_mlp": 0.01033056, "balance_loss_clip": 1.05656362, "balance_loss_mlp": 1.02469921, "epoch": 0.24565622557566283, "flos": 14027465717280.0, "grad_norm": 3.249247188911264, "language_loss": 0.82091677, "learning_rate": 3.53181067630492e-06, "loss": 0.84316957, "num_input_tokens_seen": 43636870, "step": 2043, "time_per_iteration": 2.5486438274383545 }, { "auxiliary_loss_clip": 0.01173584, "auxiliary_loss_mlp": 0.01034709, "balance_loss_clip": 1.0570755, "balance_loss_mlp": 1.02616203, "epoch": 0.24577646846630194, "flos": 16581468647040.0, "grad_norm": 1.8275456097477312, "language_loss": 0.75966084, "learning_rate": 3.5313097176985175e-06, "loss": 0.78174376, "num_input_tokens_seen": 43655180, "step": 2044, "time_per_iteration": 3.3196959495544434 }, { "auxiliary_loss_clip": 0.01196587, "auxiliary_loss_mlp": 0.01038195, "balance_loss_clip": 1.05980492, "balance_loss_mlp": 1.03012419, "epoch": 0.24589671135694102, "flos": 18807427642080.0, "grad_norm": 1.8040224700645013, "language_loss": 0.812271, "learning_rate": 3.5308085267936482e-06, "loss": 0.83461887, "num_input_tokens_seen": 43672895, "step": 2045, "time_per_iteration": 3.596735954284668 }, { "auxiliary_loss_clip": 0.01140883, "auxiliary_loss_mlp": 0.00762837, "balance_loss_clip": 1.05632389, "balance_loss_mlp": 1.00014162, "epoch": 0.2460169542475801, "flos": 19938542010240.0, "grad_norm": 1.7799864880002676, "language_loss": 0.90137118, "learning_rate": 3.530307103666342e-06, "loss": 0.92040843, "num_input_tokens_seen": 43691975, "step": 2046, "time_per_iteration": 2.7161662578582764 }, { "auxiliary_loss_clip": 0.01172669, "auxiliary_loss_mlp": 0.01029533, "balance_loss_clip": 1.05844557, "balance_loss_mlp": 1.02100992, "epoch": 0.24613719713821922, "flos": 24171225289440.0, "grad_norm": 1.6071581268932642, "language_loss": 0.8044402, "learning_rate": 3.5298054483926658e-06, "loss": 0.82646221, "num_input_tokens_seen": 43712670, "step": 2047, "time_per_iteration": 3.562032461166382 }, { "auxiliary_loss_clip": 0.01201889, "auxiliary_loss_mlp": 0.01036323, "balance_loss_clip": 1.05945134, "balance_loss_mlp": 1.0274663, "epoch": 0.2462574400288583, "flos": 30221059162560.0, "grad_norm": 1.8524231792605945, "language_loss": 0.82528752, "learning_rate": 3.5293035610487187e-06, "loss": 0.8476696, "num_input_tokens_seen": 43732035, "step": 2048, "time_per_iteration": 2.645321846008301 }, { "auxiliary_loss_clip": 0.01067812, "auxiliary_loss_mlp": 0.01017094, "balance_loss_clip": 1.01320386, "balance_loss_mlp": 1.01529372, "epoch": 0.24637768291949738, "flos": 68943042673920.0, "grad_norm": 0.7353580516908212, "language_loss": 0.62044501, "learning_rate": 3.5288014417106374e-06, "loss": 0.64129412, "num_input_tokens_seen": 43798055, "step": 2049, "time_per_iteration": 3.1867380142211914 }, { "auxiliary_loss_clip": 0.0116718, "auxiliary_loss_mlp": 0.01026404, "balance_loss_clip": 1.05834401, "balance_loss_mlp": 1.01804781, "epoch": 0.24649792581013646, "flos": 34384004482080.0, "grad_norm": 1.8957909846422345, "language_loss": 0.75578284, "learning_rate": 3.528299090454593e-06, "loss": 0.77771866, "num_input_tokens_seen": 43818590, "step": 2050, "time_per_iteration": 2.7364296913146973 }, { "auxiliary_loss_clip": 0.01198594, "auxiliary_loss_mlp": 0.01028669, "balance_loss_clip": 1.05811977, "balance_loss_mlp": 1.01995492, "epoch": 0.24661816870077558, "flos": 19680451536960.0, "grad_norm": 2.1792153324858363, "language_loss": 0.82681018, "learning_rate": 3.527796507356792e-06, "loss": 0.84908283, "num_input_tokens_seen": 43832480, "step": 2051, "time_per_iteration": 3.411237955093384 }, { "auxiliary_loss_clip": 0.01196824, "auxiliary_loss_mlp": 0.01032731, "balance_loss_clip": 1.0570817, "balance_loss_mlp": 1.02402318, "epoch": 0.24673841159141466, "flos": 20002281837600.0, "grad_norm": 2.317952677623147, "language_loss": 0.89914513, "learning_rate": 3.527293692493475e-06, "loss": 0.92144072, "num_input_tokens_seen": 43848345, "step": 2052, "time_per_iteration": 2.547451972961426 }, { "auxiliary_loss_clip": 0.01197821, "auxiliary_loss_mlp": 0.01031122, "balance_loss_clip": 1.05724525, "balance_loss_mlp": 1.02233636, "epoch": 0.24685865448205374, "flos": 21646596044640.0, "grad_norm": 2.8646195380508224, "language_loss": 0.7335375, "learning_rate": 3.52679064594092e-06, "loss": 0.75582695, "num_input_tokens_seen": 43865685, "step": 2053, "time_per_iteration": 2.564563274383545 }, { "auxiliary_loss_clip": 0.01136398, "auxiliary_loss_mlp": 0.01030947, "balance_loss_clip": 1.04473221, "balance_loss_mlp": 1.02269793, "epoch": 0.24697889737269285, "flos": 17960473071840.0, "grad_norm": 2.145163252970941, "language_loss": 0.75104034, "learning_rate": 3.5262873677754375e-06, "loss": 0.77271378, "num_input_tokens_seen": 43883690, "step": 2054, "time_per_iteration": 2.656907081604004 }, { "auxiliary_loss_clip": 0.01207866, "auxiliary_loss_mlp": 0.01030713, "balance_loss_clip": 1.06042433, "balance_loss_mlp": 1.02233899, "epoch": 0.24709914026333193, "flos": 27344615343840.0, "grad_norm": 1.9122552793116818, "language_loss": 0.804887, "learning_rate": 3.5257838580733745e-06, "loss": 0.82727283, "num_input_tokens_seen": 43903295, "step": 2055, "time_per_iteration": 2.5734972953796387 }, { "auxiliary_loss_clip": 0.01200324, "auxiliary_loss_mlp": 0.01024904, "balance_loss_clip": 1.06069982, "balance_loss_mlp": 1.01629162, "epoch": 0.24721938315397102, "flos": 19275522167520.0, "grad_norm": 2.206668194658621, "language_loss": 0.87344825, "learning_rate": 3.5252801169111138e-06, "loss": 0.89570057, "num_input_tokens_seen": 43920960, "step": 2056, "time_per_iteration": 2.5740020275115967 }, { "auxiliary_loss_clip": 0.01179366, "auxiliary_loss_mlp": 0.01029884, "balance_loss_clip": 1.06040335, "balance_loss_mlp": 1.02115798, "epoch": 0.2473396260446101, "flos": 23185800474720.0, "grad_norm": 1.7801545615658134, "language_loss": 0.79854357, "learning_rate": 3.524776144365072e-06, "loss": 0.82063609, "num_input_tokens_seen": 43939415, "step": 2057, "time_per_iteration": 2.6008706092834473 }, { "auxiliary_loss_clip": 0.01174226, "auxiliary_loss_mlp": 0.01029792, "balance_loss_clip": 1.05937433, "balance_loss_mlp": 1.02171576, "epoch": 0.2474598689352492, "flos": 21142447279680.0, "grad_norm": 1.7714399965944299, "language_loss": 0.78878641, "learning_rate": 3.5242719405117016e-06, "loss": 0.8108266, "num_input_tokens_seen": 43959220, "step": 2058, "time_per_iteration": 2.628354072570801 }, { "auxiliary_loss_clip": 0.01180893, "auxiliary_loss_mlp": 0.00763846, "balance_loss_clip": 1.05652249, "balance_loss_mlp": 1.00027478, "epoch": 0.2475801118258883, "flos": 21648355975680.0, "grad_norm": 3.057360085107292, "language_loss": 0.75127435, "learning_rate": 3.5237675054274893e-06, "loss": 0.77072173, "num_input_tokens_seen": 43978420, "step": 2059, "time_per_iteration": 2.6232004165649414 }, { "auxiliary_loss_clip": 0.01195877, "auxiliary_loss_mlp": 0.01035781, "balance_loss_clip": 1.05928147, "balance_loss_mlp": 1.02705503, "epoch": 0.24770035471652738, "flos": 22674504234720.0, "grad_norm": 1.9425318861173049, "language_loss": 0.80311608, "learning_rate": 3.5232628391889584e-06, "loss": 0.82543278, "num_input_tokens_seen": 43996710, "step": 2060, "time_per_iteration": 2.5679478645324707 }, { "auxiliary_loss_clip": 0.01146989, "auxiliary_loss_mlp": 0.01029227, "balance_loss_clip": 1.05414033, "balance_loss_mlp": 1.02091217, "epoch": 0.2478205976071665, "flos": 22163818583040.0, "grad_norm": 2.3994123591129717, "language_loss": 0.64239365, "learning_rate": 3.522757941872666e-06, "loss": 0.66415584, "num_input_tokens_seen": 44014865, "step": 2061, "time_per_iteration": 2.7032577991485596 }, { "auxiliary_loss_clip": 0.01210666, "auxiliary_loss_mlp": 0.00763213, "balance_loss_clip": 1.06188428, "balance_loss_mlp": 1.00032055, "epoch": 0.24794084049780557, "flos": 24973110463200.0, "grad_norm": 1.5222832634440024, "language_loss": 0.82439888, "learning_rate": 3.5222528135552042e-06, "loss": 0.84413767, "num_input_tokens_seen": 44036325, "step": 2062, "time_per_iteration": 2.5792977809906006 }, { "auxiliary_loss_clip": 0.01191112, "auxiliary_loss_mlp": 0.01030583, "balance_loss_clip": 1.05856252, "balance_loss_mlp": 1.02197599, "epoch": 0.24806108338844465, "flos": 18296382820800.0, "grad_norm": 1.8354331657416278, "language_loss": 0.80625421, "learning_rate": 3.521747454313201e-06, "loss": 0.82847118, "num_input_tokens_seen": 44055005, "step": 2063, "time_per_iteration": 2.5568180084228516 }, { "auxiliary_loss_clip": 0.01157391, "auxiliary_loss_mlp": 0.0103101, "balance_loss_clip": 1.05156446, "balance_loss_mlp": 1.02225447, "epoch": 0.24818132627908374, "flos": 19282166805120.0, "grad_norm": 3.977800633466725, "language_loss": 0.66898417, "learning_rate": 3.521241864223319e-06, "loss": 0.6908682, "num_input_tokens_seen": 44073965, "step": 2064, "time_per_iteration": 2.609078884124756 }, { "auxiliary_loss_clip": 0.01074215, "auxiliary_loss_mlp": 0.01004484, "balance_loss_clip": 1.01266348, "balance_loss_mlp": 1.00262451, "epoch": 0.24830156916972285, "flos": 70285849191360.0, "grad_norm": 0.7928975099202358, "language_loss": 0.619928, "learning_rate": 3.5207360433622552e-06, "loss": 0.640715, "num_input_tokens_seen": 44135965, "step": 2065, "time_per_iteration": 3.174250364303589 }, { "auxiliary_loss_clip": 0.01175679, "auxiliary_loss_mlp": 0.01029398, "balance_loss_clip": 1.05693102, "balance_loss_mlp": 1.02094615, "epoch": 0.24842181206036193, "flos": 40409133875040.0, "grad_norm": 1.7087652052103117, "language_loss": 0.74499607, "learning_rate": 3.5202299918067437e-06, "loss": 0.76704687, "num_input_tokens_seen": 44159560, "step": 2066, "time_per_iteration": 2.766172409057617 }, { "auxiliary_loss_clip": 0.01193948, "auxiliary_loss_mlp": 0.0102811, "balance_loss_clip": 1.05901742, "balance_loss_mlp": 1.01974797, "epoch": 0.248542054951001, "flos": 20082435715680.0, "grad_norm": 2.8507093550482634, "language_loss": 0.69345117, "learning_rate": 3.519723709633551e-06, "loss": 0.71567178, "num_input_tokens_seen": 44178320, "step": 2067, "time_per_iteration": 2.5592219829559326 }, { "auxiliary_loss_clip": 0.01170664, "auxiliary_loss_mlp": 0.0103265, "balance_loss_clip": 1.05592275, "balance_loss_mlp": 1.02430534, "epoch": 0.24866229784164012, "flos": 23513952160320.0, "grad_norm": 1.8566311771958206, "language_loss": 0.83314884, "learning_rate": 3.519217196919479e-06, "loss": 0.85518193, "num_input_tokens_seen": 44197305, "step": 2068, "time_per_iteration": 2.5990219116210938 }, { "auxiliary_loss_clip": 0.01182407, "auxiliary_loss_mlp": 0.01035336, "balance_loss_clip": 1.05870914, "balance_loss_mlp": 1.02703309, "epoch": 0.2487825407322792, "flos": 19865104603680.0, "grad_norm": 1.8596041739653946, "language_loss": 0.73301202, "learning_rate": 3.518710453741367e-06, "loss": 0.75518954, "num_input_tokens_seen": 44216505, "step": 2069, "time_per_iteration": 2.599724054336548 }, { "auxiliary_loss_clip": 0.01167776, "auxiliary_loss_mlp": 0.00763792, "balance_loss_clip": 1.05408168, "balance_loss_mlp": 1.00020278, "epoch": 0.2489027836229183, "flos": 22017626192160.0, "grad_norm": 2.1523607338973902, "language_loss": 0.67782456, "learning_rate": 3.518203480176086e-06, "loss": 0.69714022, "num_input_tokens_seen": 44235435, "step": 2070, "time_per_iteration": 4.085761547088623 }, { "auxiliary_loss_clip": 0.01120648, "auxiliary_loss_mlp": 0.01034777, "balance_loss_clip": 1.0459137, "balance_loss_mlp": 1.02716565, "epoch": 0.2490230265135574, "flos": 23294358279840.0, "grad_norm": 1.7865198430289044, "language_loss": 0.80441856, "learning_rate": 3.517696276300545e-06, "loss": 0.8259728, "num_input_tokens_seen": 44256975, "step": 2071, "time_per_iteration": 2.75650691986084 }, { "auxiliary_loss_clip": 0.01191986, "auxiliary_loss_mlp": 0.01031017, "balance_loss_clip": 1.05970705, "balance_loss_mlp": 1.02223682, "epoch": 0.24914326940419648, "flos": 19826787595680.0, "grad_norm": 2.411590862005898, "language_loss": 0.69132602, "learning_rate": 3.517188842191685e-06, "loss": 0.71355611, "num_input_tokens_seen": 44275125, "step": 2072, "time_per_iteration": 2.5609891414642334 }, { "auxiliary_loss_clip": 0.01189843, "auxiliary_loss_mlp": 0.01026536, "balance_loss_clip": 1.05559993, "balance_loss_mlp": 1.0181855, "epoch": 0.24926351229483557, "flos": 20229274611840.0, "grad_norm": 1.5646081832552727, "language_loss": 0.73796266, "learning_rate": 3.5166811779264837e-06, "loss": 0.76012647, "num_input_tokens_seen": 44295445, "step": 2073, "time_per_iteration": 3.492379903793335 }, { "auxiliary_loss_clip": 0.01207486, "auxiliary_loss_mlp": 0.01037566, "balance_loss_clip": 1.05788577, "balance_loss_mlp": 1.02849984, "epoch": 0.24938375518547465, "flos": 23294573781600.0, "grad_norm": 1.839829247276346, "language_loss": 0.7775721, "learning_rate": 3.5161732835819545e-06, "loss": 0.8000226, "num_input_tokens_seen": 44314755, "step": 2074, "time_per_iteration": 2.5886528491973877 }, { "auxiliary_loss_clip": 0.01206449, "auxiliary_loss_mlp": 0.01029853, "balance_loss_clip": 1.05902171, "balance_loss_mlp": 1.02166319, "epoch": 0.24950399807611376, "flos": 17311676345280.0, "grad_norm": 2.865853262976597, "language_loss": 0.82732964, "learning_rate": 3.515665159235143e-06, "loss": 0.84969264, "num_input_tokens_seen": 44333640, "step": 2075, "time_per_iteration": 2.5303165912628174 }, { "auxiliary_loss_clip": 0.01174267, "auxiliary_loss_mlp": 0.0102897, "balance_loss_clip": 1.05202925, "balance_loss_mlp": 1.02089381, "epoch": 0.24962424096675284, "flos": 19024866504960.0, "grad_norm": 1.6017588868659385, "language_loss": 0.74804568, "learning_rate": 3.5151568049631318e-06, "loss": 0.77007806, "num_input_tokens_seen": 44352355, "step": 2076, "time_per_iteration": 2.548715114593506 }, { "auxiliary_loss_clip": 0.01209287, "auxiliary_loss_mlp": 0.01031751, "balance_loss_clip": 1.05945706, "balance_loss_mlp": 1.02331722, "epoch": 0.24974448385739192, "flos": 33398795169120.0, "grad_norm": 2.131901642674451, "language_loss": 0.8045758, "learning_rate": 3.5146482208430385e-06, "loss": 0.82698619, "num_input_tokens_seen": 44374185, "step": 2077, "time_per_iteration": 3.37200665473938 }, { "auxiliary_loss_clip": 0.01121846, "auxiliary_loss_mlp": 0.01030499, "balance_loss_clip": 1.04619479, "balance_loss_mlp": 1.02113533, "epoch": 0.24986472674803104, "flos": 30007283829600.0, "grad_norm": 6.679029157127327, "language_loss": 0.67700207, "learning_rate": 3.514139406952014e-06, "loss": 0.69852555, "num_input_tokens_seen": 44396210, "step": 2078, "time_per_iteration": 2.7170028686523438 }, { "auxiliary_loss_clip": 0.01190299, "auxiliary_loss_mlp": 0.01031511, "balance_loss_clip": 1.05725777, "balance_loss_mlp": 1.02314818, "epoch": 0.24998496963867012, "flos": 26613078718080.0, "grad_norm": 1.8596352597430954, "language_loss": 0.83626008, "learning_rate": 3.5136303633672454e-06, "loss": 0.85847819, "num_input_tokens_seen": 44416340, "step": 2079, "time_per_iteration": 2.621140480041504 }, { "auxiliary_loss_clip": 0.0117407, "auxiliary_loss_mlp": 0.00763301, "balance_loss_clip": 1.05722523, "balance_loss_mlp": 1.00015616, "epoch": 0.25010521252930923, "flos": 23553993182400.0, "grad_norm": 2.18361547061319, "language_loss": 0.74753797, "learning_rate": 3.5131210901659544e-06, "loss": 0.76691169, "num_input_tokens_seen": 44438095, "step": 2080, "time_per_iteration": 2.6894783973693848 }, { "auxiliary_loss_clip": 0.01159349, "auxiliary_loss_mlp": 0.01031594, "balance_loss_clip": 1.0516355, "balance_loss_mlp": 1.02317834, "epoch": 0.2502254554199483, "flos": 23441197176000.0, "grad_norm": 2.3842179013325486, "language_loss": 0.82068312, "learning_rate": 3.5126115874253967e-06, "loss": 0.8425926, "num_input_tokens_seen": 44457650, "step": 2081, "time_per_iteration": 2.652304172515869 }, { "auxiliary_loss_clip": 0.01167722, "auxiliary_loss_mlp": 0.01034255, "balance_loss_clip": 1.05731833, "balance_loss_mlp": 1.02521336, "epoch": 0.2503456983105874, "flos": 28761685357920.0, "grad_norm": 2.0822875994740015, "language_loss": 0.80838919, "learning_rate": 3.5121018552228644e-06, "loss": 0.83040893, "num_input_tokens_seen": 44476155, "step": 2082, "time_per_iteration": 2.7007436752319336 }, { "auxiliary_loss_clip": 0.01161305, "auxiliary_loss_mlp": 0.01028458, "balance_loss_clip": 1.05109, "balance_loss_mlp": 1.02007198, "epoch": 0.2504659412012265, "flos": 18770259976800.0, "grad_norm": 2.5876094124697113, "language_loss": 0.7578119, "learning_rate": 3.5115918936356827e-06, "loss": 0.77970958, "num_input_tokens_seen": 44492910, "step": 2083, "time_per_iteration": 2.6180591583251953 }, { "auxiliary_loss_clip": 0.01143394, "auxiliary_loss_mlp": 0.0103173, "balance_loss_clip": 1.05081487, "balance_loss_mlp": 1.02368629, "epoch": 0.25058618409186556, "flos": 16873386508320.0, "grad_norm": 1.780651592587906, "language_loss": 0.78887141, "learning_rate": 3.5110817027412123e-06, "loss": 0.81062263, "num_input_tokens_seen": 44512000, "step": 2084, "time_per_iteration": 2.6164298057556152 }, { "auxiliary_loss_clip": 0.01156755, "auxiliary_loss_mlp": 0.01030856, "balance_loss_clip": 1.04834771, "balance_loss_mlp": 1.02308393, "epoch": 0.25070642698250467, "flos": 24425544481920.0, "grad_norm": 4.099563317780119, "language_loss": 0.68267155, "learning_rate": 3.5105712826168493e-06, "loss": 0.7045477, "num_input_tokens_seen": 44531650, "step": 2085, "time_per_iteration": 2.6537933349609375 }, { "auxiliary_loss_clip": 0.01192151, "auxiliary_loss_mlp": 0.00763013, "balance_loss_clip": 1.05603147, "balance_loss_mlp": 1.00017691, "epoch": 0.2508266698731437, "flos": 20260946982240.0, "grad_norm": 1.9041456084223478, "language_loss": 0.70668989, "learning_rate": 3.5100606333400235e-06, "loss": 0.72624153, "num_input_tokens_seen": 44548785, "step": 2086, "time_per_iteration": 2.551044225692749 }, { "auxiliary_loss_clip": 0.01187353, "auxiliary_loss_mlp": 0.0102971, "balance_loss_clip": 1.05648506, "balance_loss_mlp": 1.02128172, "epoch": 0.25094691276378284, "flos": 19245322392480.0, "grad_norm": 3.1335162696746943, "language_loss": 0.77254015, "learning_rate": 3.5095497549882006e-06, "loss": 0.79471081, "num_input_tokens_seen": 44567230, "step": 2087, "time_per_iteration": 2.5882441997528076 }, { "auxiliary_loss_clip": 0.01193503, "auxiliary_loss_mlp": 0.01033643, "balance_loss_clip": 1.0600673, "balance_loss_mlp": 1.0248754, "epoch": 0.25106715565442195, "flos": 26943241753440.0, "grad_norm": 1.9617617824462767, "language_loss": 0.72061694, "learning_rate": 3.50903864763888e-06, "loss": 0.74288833, "num_input_tokens_seen": 44588020, "step": 2088, "time_per_iteration": 2.5941717624664307 }, { "auxiliary_loss_clip": 0.01196024, "auxiliary_loss_mlp": 0.01034837, "balance_loss_clip": 1.05595195, "balance_loss_mlp": 1.02627134, "epoch": 0.251187398545061, "flos": 48359576580480.0, "grad_norm": 1.8933845035215868, "language_loss": 0.75643194, "learning_rate": 3.5085273113695965e-06, "loss": 0.77874053, "num_input_tokens_seen": 44612590, "step": 2089, "time_per_iteration": 2.792238473892212 }, { "auxiliary_loss_clip": 0.01207432, "auxiliary_loss_mlp": 0.01026939, "balance_loss_clip": 1.05911589, "balance_loss_mlp": 1.01845706, "epoch": 0.2513076414357001, "flos": 27016104488640.0, "grad_norm": 1.804656420674964, "language_loss": 0.78303879, "learning_rate": 3.508015746257919e-06, "loss": 0.80538249, "num_input_tokens_seen": 44631630, "step": 2090, "time_per_iteration": 2.575904130935669 }, { "auxiliary_loss_clip": 0.01169045, "auxiliary_loss_mlp": 0.01036746, "balance_loss_clip": 1.0548327, "balance_loss_mlp": 1.02805543, "epoch": 0.2514278843263392, "flos": 19463623262400.0, "grad_norm": 2.232007793325033, "language_loss": 0.82820272, "learning_rate": 3.5075039523814518e-06, "loss": 0.85026067, "num_input_tokens_seen": 44650820, "step": 2091, "time_per_iteration": 2.609722852706909 }, { "auxiliary_loss_clip": 0.01197886, "auxiliary_loss_mlp": 0.01026148, "balance_loss_clip": 1.05539823, "balance_loss_mlp": 1.01751113, "epoch": 0.2515481272169783, "flos": 16866095365440.0, "grad_norm": 1.8574087554373306, "language_loss": 0.81324178, "learning_rate": 3.506991929817834e-06, "loss": 0.83548212, "num_input_tokens_seen": 44667540, "step": 2092, "time_per_iteration": 2.554352283477783 }, { "auxiliary_loss_clip": 0.01206258, "auxiliary_loss_mlp": 0.01032255, "balance_loss_clip": 1.05987787, "balance_loss_mlp": 1.02445865, "epoch": 0.2516683701076174, "flos": 23732468532000.0, "grad_norm": 1.7589343984636545, "language_loss": 0.82573801, "learning_rate": 3.506479678644738e-06, "loss": 0.84812319, "num_input_tokens_seen": 44687935, "step": 2093, "time_per_iteration": 2.5484209060668945 }, { "auxiliary_loss_clip": 0.01140491, "auxiliary_loss_mlp": 0.01032401, "balance_loss_clip": 1.05082679, "balance_loss_mlp": 1.02315688, "epoch": 0.2517886129982565, "flos": 27635958533760.0, "grad_norm": 2.5407817163884743, "language_loss": 0.73661923, "learning_rate": 3.505967198939873e-06, "loss": 0.75834823, "num_input_tokens_seen": 44704975, "step": 2094, "time_per_iteration": 2.693310260772705 }, { "auxiliary_loss_clip": 0.01174416, "auxiliary_loss_mlp": 0.01032841, "balance_loss_clip": 1.05100489, "balance_loss_mlp": 1.0242579, "epoch": 0.25190885588889556, "flos": 38104601348160.0, "grad_norm": 1.8854189994118986, "language_loss": 0.78070748, "learning_rate": 3.5054544907809813e-06, "loss": 0.80278003, "num_input_tokens_seen": 44725475, "step": 2095, "time_per_iteration": 2.75070858001709 }, { "auxiliary_loss_clip": 0.01175333, "auxiliary_loss_mlp": 0.00763114, "balance_loss_clip": 1.05577207, "balance_loss_mlp": 1.0002178, "epoch": 0.25202909877953467, "flos": 22269898117920.0, "grad_norm": 2.0397222761515104, "language_loss": 0.80576038, "learning_rate": 3.50494155424584e-06, "loss": 0.82514489, "num_input_tokens_seen": 44744380, "step": 2096, "time_per_iteration": 4.169250965118408 }, { "auxiliary_loss_clip": 0.01194028, "auxiliary_loss_mlp": 0.01033069, "balance_loss_clip": 1.05612874, "balance_loss_mlp": 1.02454555, "epoch": 0.2521493416701738, "flos": 21761762570400.0, "grad_norm": 1.7736756457608058, "language_loss": 0.83301109, "learning_rate": 3.504428389412262e-06, "loss": 0.85528207, "num_input_tokens_seen": 44765190, "step": 2097, "time_per_iteration": 2.6052684783935547 }, { "auxiliary_loss_clip": 0.01190679, "auxiliary_loss_mlp": 0.01026459, "balance_loss_clip": 1.05532598, "balance_loss_mlp": 1.01804316, "epoch": 0.25226958456081283, "flos": 27746743190400.0, "grad_norm": 2.1956951769634645, "language_loss": 0.7312156, "learning_rate": 3.5039149963580927e-06, "loss": 0.75338691, "num_input_tokens_seen": 44785210, "step": 2098, "time_per_iteration": 2.614959955215454 }, { "auxiliary_loss_clip": 0.01173342, "auxiliary_loss_mlp": 0.0103105, "balance_loss_clip": 1.0575316, "balance_loss_mlp": 1.0221746, "epoch": 0.25238982745145194, "flos": 30732175817760.0, "grad_norm": 2.0810635495740377, "language_loss": 0.70219117, "learning_rate": 3.503401375161215e-06, "loss": 0.72423512, "num_input_tokens_seen": 44804955, "step": 2099, "time_per_iteration": 3.5472891330718994 }, { "auxiliary_loss_clip": 0.01205532, "auxiliary_loss_mlp": 0.01034597, "balance_loss_clip": 1.05725479, "balance_loss_mlp": 1.02672386, "epoch": 0.252510070342091, "flos": 20266334526240.0, "grad_norm": 1.53768213870322, "language_loss": 0.83522761, "learning_rate": 3.502887525899544e-06, "loss": 0.85762882, "num_input_tokens_seen": 44823935, "step": 2100, "time_per_iteration": 2.534320831298828 }, { "auxiliary_loss_clip": 0.01177184, "auxiliary_loss_mlp": 0.01032258, "balance_loss_clip": 1.05575788, "balance_loss_mlp": 1.02329373, "epoch": 0.2526303132327301, "flos": 22747402886880.0, "grad_norm": 1.8175677880164112, "language_loss": 0.82868373, "learning_rate": 3.50237344865103e-06, "loss": 0.85077816, "num_input_tokens_seen": 44844935, "step": 2101, "time_per_iteration": 2.648876667022705 }, { "auxiliary_loss_clip": 0.01208209, "auxiliary_loss_mlp": 0.01031475, "balance_loss_clip": 1.05964732, "balance_loss_mlp": 1.02301121, "epoch": 0.2527505561233692, "flos": 30263470704000.0, "grad_norm": 13.09616258351244, "language_loss": 0.76146281, "learning_rate": 3.501859143493658e-06, "loss": 0.78385961, "num_input_tokens_seen": 44865565, "step": 2102, "time_per_iteration": 2.6281769275665283 }, { "auxiliary_loss_clip": 0.010993, "auxiliary_loss_mlp": 0.01003575, "balance_loss_clip": 1.01919246, "balance_loss_mlp": 1.00154865, "epoch": 0.2528707990140083, "flos": 58492926622560.0, "grad_norm": 0.923396248127977, "language_loss": 0.60571861, "learning_rate": 3.5013446105054488e-06, "loss": 0.62674737, "num_input_tokens_seen": 44918485, "step": 2103, "time_per_iteration": 3.6986327171325684 }, { "auxiliary_loss_clip": 0.01146645, "auxiliary_loss_mlp": 0.01035374, "balance_loss_clip": 1.05102253, "balance_loss_mlp": 1.02657676, "epoch": 0.2529910419046474, "flos": 24645138362400.0, "grad_norm": 1.6809929767916967, "language_loss": 0.74877781, "learning_rate": 3.5008298497644555e-06, "loss": 0.77059799, "num_input_tokens_seen": 44937530, "step": 2104, "time_per_iteration": 2.656222105026245 }, { "auxiliary_loss_clip": 0.01163744, "auxiliary_loss_mlp": 0.01027492, "balance_loss_clip": 1.05387735, "balance_loss_mlp": 1.01886082, "epoch": 0.2531112847952865, "flos": 23842139762880.0, "grad_norm": 1.6663792609742263, "language_loss": 0.88067079, "learning_rate": 3.500314861348767e-06, "loss": 0.90258312, "num_input_tokens_seen": 44958165, "step": 2105, "time_per_iteration": 2.64290189743042 }, { "auxiliary_loss_clip": 0.01153837, "auxiliary_loss_mlp": 0.01035735, "balance_loss_clip": 1.05249047, "balance_loss_mlp": 1.02715147, "epoch": 0.25323152768592555, "flos": 16143825315360.0, "grad_norm": 1.9420759725888197, "language_loss": 0.77478647, "learning_rate": 3.499799645336507e-06, "loss": 0.79668224, "num_input_tokens_seen": 44975060, "step": 2106, "time_per_iteration": 2.6076040267944336 }, { "auxiliary_loss_clip": 0.01193855, "auxiliary_loss_mlp": 0.01035544, "balance_loss_clip": 1.0577662, "balance_loss_mlp": 1.02754557, "epoch": 0.25335177057656466, "flos": 28405165662240.0, "grad_norm": 1.4955697248521003, "language_loss": 0.86555743, "learning_rate": 3.4992842018058336e-06, "loss": 0.88785148, "num_input_tokens_seen": 44997960, "step": 2107, "time_per_iteration": 2.644575595855713 }, { "auxiliary_loss_clip": 0.01170864, "auxiliary_loss_mlp": 0.0103237, "balance_loss_clip": 1.05368805, "balance_loss_mlp": 1.02416229, "epoch": 0.25347201346720377, "flos": 18799669578720.0, "grad_norm": 2.49527969990012, "language_loss": 0.88311833, "learning_rate": 3.4987685308349384e-06, "loss": 0.90515071, "num_input_tokens_seen": 45015690, "step": 2108, "time_per_iteration": 2.650160312652588 }, { "auxiliary_loss_clip": 0.01159902, "auxiliary_loss_mlp": 0.01027294, "balance_loss_clip": 1.04894781, "balance_loss_mlp": 1.01865768, "epoch": 0.2535922563578428, "flos": 15815494044960.0, "grad_norm": 2.058191675767595, "language_loss": 0.61176777, "learning_rate": 3.4982526325020497e-06, "loss": 0.63363969, "num_input_tokens_seen": 45032660, "step": 2109, "time_per_iteration": 2.598522186279297 }, { "auxiliary_loss_clip": 0.01179488, "auxiliary_loss_mlp": 0.01029921, "balance_loss_clip": 1.05500042, "balance_loss_mlp": 1.02089667, "epoch": 0.25371249924848194, "flos": 16318924470720.0, "grad_norm": 2.1177789839786345, "language_loss": 0.82108951, "learning_rate": 3.4977365068854273e-06, "loss": 0.84318358, "num_input_tokens_seen": 45048280, "step": 2110, "time_per_iteration": 2.56534743309021 }, { "auxiliary_loss_clip": 0.01166489, "auxiliary_loss_mlp": 0.01033461, "balance_loss_clip": 1.05219173, "balance_loss_mlp": 1.02466929, "epoch": 0.25383274213912105, "flos": 21761618902560.0, "grad_norm": 1.816578727319156, "language_loss": 0.73679292, "learning_rate": 3.4972201540633676e-06, "loss": 0.7587924, "num_input_tokens_seen": 45067635, "step": 2111, "time_per_iteration": 2.6000876426696777 }, { "auxiliary_loss_clip": 0.01171353, "auxiliary_loss_mlp": 0.01039133, "balance_loss_clip": 1.05578327, "balance_loss_mlp": 1.03059149, "epoch": 0.2539529850297601, "flos": 21396874223040.0, "grad_norm": 1.8976755560223388, "language_loss": 0.85459483, "learning_rate": 3.4967035741142008e-06, "loss": 0.87669969, "num_input_tokens_seen": 45086455, "step": 2112, "time_per_iteration": 2.6263978481292725 }, { "auxiliary_loss_clip": 0.01169948, "auxiliary_loss_mlp": 0.01024744, "balance_loss_clip": 1.05956697, "balance_loss_mlp": 1.01691175, "epoch": 0.2540732279203992, "flos": 25228471247520.0, "grad_norm": 2.0043666862377254, "language_loss": 0.82047504, "learning_rate": 3.4961867671162917e-06, "loss": 0.84242189, "num_input_tokens_seen": 45106385, "step": 2113, "time_per_iteration": 2.6128551959991455 }, { "auxiliary_loss_clip": 0.01207424, "auxiliary_loss_mlp": 0.01030916, "balance_loss_clip": 1.05771756, "balance_loss_mlp": 1.02195692, "epoch": 0.2541934708110383, "flos": 19427389438080.0, "grad_norm": 2.4969164161276116, "language_loss": 0.77336985, "learning_rate": 3.4956697331480402e-06, "loss": 0.79575324, "num_input_tokens_seen": 45124955, "step": 2114, "time_per_iteration": 2.542452812194824 }, { "auxiliary_loss_clip": 0.01165141, "auxiliary_loss_mlp": 0.01036124, "balance_loss_clip": 1.0507946, "balance_loss_mlp": 1.02746987, "epoch": 0.2543137137016774, "flos": 23949440474400.0, "grad_norm": 1.6079942873170014, "language_loss": 0.79862595, "learning_rate": 3.495152472287879e-06, "loss": 0.82063854, "num_input_tokens_seen": 45145665, "step": 2115, "time_per_iteration": 2.6404402256011963 }, { "auxiliary_loss_clip": 0.01166654, "auxiliary_loss_mlp": 0.01040108, "balance_loss_clip": 1.05728102, "balance_loss_mlp": 1.03193653, "epoch": 0.2544339565923165, "flos": 25593287760960.0, "grad_norm": 1.7184937181070756, "language_loss": 0.7386269, "learning_rate": 3.4946349846142766e-06, "loss": 0.7606945, "num_input_tokens_seen": 45164805, "step": 2116, "time_per_iteration": 2.656748056411743 }, { "auxiliary_loss_clip": 0.0120627, "auxiliary_loss_mlp": 0.01038324, "balance_loss_clip": 1.05824769, "balance_loss_mlp": 1.02945471, "epoch": 0.25455419948295555, "flos": 21689474506560.0, "grad_norm": 2.9648320047170165, "language_loss": 0.75580144, "learning_rate": 3.4941172702057353e-06, "loss": 0.7782473, "num_input_tokens_seen": 45184865, "step": 2117, "time_per_iteration": 2.5249316692352295 }, { "auxiliary_loss_clip": 0.01179575, "auxiliary_loss_mlp": 0.01029013, "balance_loss_clip": 1.05684471, "balance_loss_mlp": 1.02064419, "epoch": 0.25467444237359466, "flos": 26250345388320.0, "grad_norm": 1.7307830410747944, "language_loss": 0.80244893, "learning_rate": 3.4935993291407924e-06, "loss": 0.82453477, "num_input_tokens_seen": 45203690, "step": 2118, "time_per_iteration": 2.6368765830993652 }, { "auxiliary_loss_clip": 0.01175233, "auxiliary_loss_mlp": 0.0103108, "balance_loss_clip": 1.05550027, "balance_loss_mlp": 1.02228808, "epoch": 0.25479468526423377, "flos": 26979691079520.0, "grad_norm": 5.608397511187589, "language_loss": 0.71286523, "learning_rate": 3.4930811614980183e-06, "loss": 0.73492837, "num_input_tokens_seen": 45225385, "step": 2119, "time_per_iteration": 2.643099069595337 }, { "auxiliary_loss_clip": 0.01185679, "auxiliary_loss_mlp": 0.01041496, "balance_loss_clip": 1.05552483, "balance_loss_mlp": 1.03295493, "epoch": 0.2549149281548728, "flos": 23475814737120.0, "grad_norm": 1.8680432584510995, "language_loss": 0.79445338, "learning_rate": 3.4925627673560198e-06, "loss": 0.81672513, "num_input_tokens_seen": 45246045, "step": 2120, "time_per_iteration": 2.601935625076294 }, { "auxiliary_loss_clip": 0.01164564, "auxiliary_loss_mlp": 0.01030476, "balance_loss_clip": 1.0554595, "balance_loss_mlp": 1.02182698, "epoch": 0.25503517104551193, "flos": 25812306970080.0, "grad_norm": 1.9588802309030606, "language_loss": 0.88147008, "learning_rate": 3.4920441467934357e-06, "loss": 0.90342045, "num_input_tokens_seen": 45266560, "step": 2121, "time_per_iteration": 2.684194803237915 }, { "auxiliary_loss_clip": 0.01152961, "auxiliary_loss_mlp": 0.01031812, "balance_loss_clip": 1.05098331, "balance_loss_mlp": 1.02350938, "epoch": 0.25515541393615104, "flos": 26645110258080.0, "grad_norm": 2.2326296438265625, "language_loss": 0.83047265, "learning_rate": 3.491525299888941e-06, "loss": 0.85232043, "num_input_tokens_seen": 45285405, "step": 2122, "time_per_iteration": 3.512667655944824 }, { "auxiliary_loss_clip": 0.01068295, "auxiliary_loss_mlp": 0.00755714, "balance_loss_clip": 1.02021563, "balance_loss_mlp": 0.99987048, "epoch": 0.2552756568267901, "flos": 65955957866400.0, "grad_norm": 0.9017105073992749, "language_loss": 0.62728751, "learning_rate": 3.491006226721244e-06, "loss": 0.6455276, "num_input_tokens_seen": 45349615, "step": 2123, "time_per_iteration": 3.2352209091186523 }, { "auxiliary_loss_clip": 0.01182273, "auxiliary_loss_mlp": 0.00763201, "balance_loss_clip": 1.05915153, "balance_loss_mlp": 1.00014615, "epoch": 0.2553958997174292, "flos": 17931099386880.0, "grad_norm": 2.63044995319465, "language_loss": 0.77905083, "learning_rate": 3.4904869273690882e-06, "loss": 0.7985056, "num_input_tokens_seen": 45367505, "step": 2124, "time_per_iteration": 2.576979875564575 }, { "auxiliary_loss_clip": 0.01196983, "auxiliary_loss_mlp": 0.01034326, "balance_loss_clip": 1.05794668, "balance_loss_mlp": 1.02541482, "epoch": 0.2555161426080683, "flos": 23367795686400.0, "grad_norm": 2.0896399978870988, "language_loss": 0.89393938, "learning_rate": 3.489967401911251e-06, "loss": 0.91625237, "num_input_tokens_seen": 45386805, "step": 2125, "time_per_iteration": 3.472270965576172 }, { "auxiliary_loss_clip": 0.01211069, "auxiliary_loss_mlp": 0.01036855, "balance_loss_clip": 1.0603441, "balance_loss_mlp": 1.02806973, "epoch": 0.2556363854987074, "flos": 40625136059520.0, "grad_norm": 1.7907231945380326, "language_loss": 0.69400722, "learning_rate": 3.4894476504265428e-06, "loss": 0.71648645, "num_input_tokens_seen": 45411045, "step": 2126, "time_per_iteration": 2.6995201110839844 }, { "auxiliary_loss_clip": 0.01080848, "auxiliary_loss_mlp": 0.01006758, "balance_loss_clip": 1.0167141, "balance_loss_mlp": 1.0047549, "epoch": 0.2557566283893465, "flos": 68019454087680.0, "grad_norm": 2.974057745911107, "language_loss": 0.54485381, "learning_rate": 3.4889276729938104e-06, "loss": 0.56572986, "num_input_tokens_seen": 45469575, "step": 2127, "time_per_iteration": 3.056272268295288 }, { "auxiliary_loss_clip": 0.01175435, "auxiliary_loss_mlp": 0.01028996, "balance_loss_clip": 1.05724692, "balance_loss_mlp": 1.0204134, "epoch": 0.2558768712799856, "flos": 22635648472320.0, "grad_norm": 3.6428453462646675, "language_loss": 0.80057013, "learning_rate": 3.488407469691934e-06, "loss": 0.82261443, "num_input_tokens_seen": 45490270, "step": 2128, "time_per_iteration": 2.6199886798858643 }, { "auxiliary_loss_clip": 0.01177054, "auxiliary_loss_mlp": 0.01032325, "balance_loss_clip": 1.05358815, "balance_loss_mlp": 1.02331281, "epoch": 0.25599711417062465, "flos": 26396358194400.0, "grad_norm": 2.241345746023726, "language_loss": 0.80917817, "learning_rate": 3.487887040599828e-06, "loss": 0.83127195, "num_input_tokens_seen": 45510070, "step": 2129, "time_per_iteration": 3.5392823219299316 }, { "auxiliary_loss_clip": 0.01211266, "auxiliary_loss_mlp": 0.01044841, "balance_loss_clip": 1.06127095, "balance_loss_mlp": 1.03536415, "epoch": 0.25611735706126376, "flos": 22852045743360.0, "grad_norm": 3.8386299482701514, "language_loss": 0.7624746, "learning_rate": 3.4873663857964407e-06, "loss": 0.78503567, "num_input_tokens_seen": 45527285, "step": 2130, "time_per_iteration": 2.542951822280884 }, { "auxiliary_loss_clip": 0.01148504, "auxiliary_loss_mlp": 0.01035122, "balance_loss_clip": 1.05269361, "balance_loss_mlp": 1.02646136, "epoch": 0.2562375999519028, "flos": 23367867520320.0, "grad_norm": 1.9863494347452026, "language_loss": 0.66424316, "learning_rate": 3.4868455053607556e-06, "loss": 0.68607938, "num_input_tokens_seen": 45546900, "step": 2131, "time_per_iteration": 2.6698086261749268 }, { "auxiliary_loss_clip": 0.01197649, "auxiliary_loss_mlp": 0.01039461, "balance_loss_clip": 1.05739665, "balance_loss_mlp": 1.03035927, "epoch": 0.2563578428425419, "flos": 22856966366880.0, "grad_norm": 2.7489614928321036, "language_loss": 0.72349983, "learning_rate": 3.486324399371789e-06, "loss": 0.74587089, "num_input_tokens_seen": 45566200, "step": 2132, "time_per_iteration": 2.5790069103240967 }, { "auxiliary_loss_clip": 0.01160055, "auxiliary_loss_mlp": 0.01029255, "balance_loss_clip": 1.05525768, "balance_loss_mlp": 1.02157795, "epoch": 0.25647808573318104, "flos": 21653887187520.0, "grad_norm": 1.9614201062166532, "language_loss": 0.78401625, "learning_rate": 3.485803067908593e-06, "loss": 0.80590934, "num_input_tokens_seen": 45585710, "step": 2133, "time_per_iteration": 2.6403045654296875 }, { "auxiliary_loss_clip": 0.01110814, "auxiliary_loss_mlp": 0.01037266, "balance_loss_clip": 1.04469514, "balance_loss_mlp": 1.02869534, "epoch": 0.2565983286238201, "flos": 33730574467680.0, "grad_norm": 1.6884643714693222, "language_loss": 0.79682839, "learning_rate": 3.485281511050253e-06, "loss": 0.81830919, "num_input_tokens_seen": 45607845, "step": 2134, "time_per_iteration": 2.7995009422302246 }, { "auxiliary_loss_clip": 0.01193797, "auxiliary_loss_mlp": 0.01032227, "balance_loss_clip": 1.05660915, "balance_loss_mlp": 1.02413917, "epoch": 0.2567185715144592, "flos": 16216005628320.0, "grad_norm": 2.269282142026353, "language_loss": 0.90243471, "learning_rate": 3.484759728875889e-06, "loss": 0.9246949, "num_input_tokens_seen": 45623210, "step": 2135, "time_per_iteration": 2.528534173965454 }, { "auxiliary_loss_clip": 0.01138302, "auxiliary_loss_mlp": 0.01027055, "balance_loss_clip": 1.04987824, "balance_loss_mlp": 1.0185194, "epoch": 0.2568388144050983, "flos": 17458479324480.0, "grad_norm": 1.8619525989845056, "language_loss": 0.8104533, "learning_rate": 3.4842377214646543e-06, "loss": 0.83210689, "num_input_tokens_seen": 45641505, "step": 2136, "time_per_iteration": 2.6354899406433105 }, { "auxiliary_loss_clip": 0.01205404, "auxiliary_loss_mlp": 0.01032641, "balance_loss_clip": 1.05956411, "balance_loss_mlp": 1.02435017, "epoch": 0.25695905729573737, "flos": 20887445664960.0, "grad_norm": 1.6347879495508895, "language_loss": 0.66921341, "learning_rate": 3.483715488895737e-06, "loss": 0.69159389, "num_input_tokens_seen": 45661835, "step": 2137, "time_per_iteration": 2.561782121658325 }, { "auxiliary_loss_clip": 0.01144895, "auxiliary_loss_mlp": 0.01029413, "balance_loss_clip": 1.04715562, "balance_loss_mlp": 1.02059197, "epoch": 0.2570793001863765, "flos": 24717282758400.0, "grad_norm": 1.7801565080601232, "language_loss": 0.78625441, "learning_rate": 3.48319303124836e-06, "loss": 0.80799747, "num_input_tokens_seen": 45682215, "step": 2138, "time_per_iteration": 2.678187847137451 }, { "auxiliary_loss_clip": 0.01171453, "auxiliary_loss_mlp": 0.01030748, "balance_loss_clip": 1.0555476, "balance_loss_mlp": 1.02237368, "epoch": 0.2571995430770156, "flos": 26906900178240.0, "grad_norm": 2.242413285592266, "language_loss": 0.6699034, "learning_rate": 3.4826703486017798e-06, "loss": 0.69192541, "num_input_tokens_seen": 45701840, "step": 2139, "time_per_iteration": 2.633190870285034 }, { "auxiliary_loss_clip": 0.01191023, "auxiliary_loss_mlp": 0.01036055, "balance_loss_clip": 1.05927658, "balance_loss_mlp": 1.02804983, "epoch": 0.25731978596765465, "flos": 19792565121120.0, "grad_norm": 2.027133283018911, "language_loss": 0.76532435, "learning_rate": 3.4821474410352867e-06, "loss": 0.78759509, "num_input_tokens_seen": 45720500, "step": 2140, "time_per_iteration": 2.5625669956207275 }, { "auxiliary_loss_clip": 0.01060793, "auxiliary_loss_mlp": 0.01001138, "balance_loss_clip": 1.01691246, "balance_loss_mlp": 0.99908727, "epoch": 0.25744002885829376, "flos": 70564980614880.0, "grad_norm": 0.8946876735260637, "language_loss": 0.62696469, "learning_rate": 3.481624308628205e-06, "loss": 0.64758396, "num_input_tokens_seen": 45781870, "step": 2141, "time_per_iteration": 3.2974894046783447 }, { "auxiliary_loss_clip": 0.01179207, "auxiliary_loss_mlp": 0.01031644, "balance_loss_clip": 1.05676794, "balance_loss_mlp": 1.02253079, "epoch": 0.25756027174893287, "flos": 18038184596640.0, "grad_norm": 2.7551156823269234, "language_loss": 1.00488579, "learning_rate": 3.481100951459893e-06, "loss": 1.02699435, "num_input_tokens_seen": 45794890, "step": 2142, "time_per_iteration": 2.5464015007019043 }, { "auxiliary_loss_clip": 0.01189245, "auxiliary_loss_mlp": 0.01031637, "balance_loss_clip": 1.05645633, "balance_loss_mlp": 1.02279198, "epoch": 0.2576805146395719, "flos": 22674073231200.0, "grad_norm": 1.752747791218205, "language_loss": 0.78672695, "learning_rate": 3.4805773696097453e-06, "loss": 0.80893576, "num_input_tokens_seen": 45815780, "step": 2143, "time_per_iteration": 2.5990395545959473 }, { "auxiliary_loss_clip": 0.01172015, "auxiliary_loss_mlp": 0.01032163, "balance_loss_clip": 1.05736434, "balance_loss_mlp": 1.02368176, "epoch": 0.25780075753021103, "flos": 16472228419680.0, "grad_norm": 1.9347988236849303, "language_loss": 0.87789965, "learning_rate": 3.4800535631571874e-06, "loss": 0.89994144, "num_input_tokens_seen": 45831310, "step": 2144, "time_per_iteration": 2.532817840576172 }, { "auxiliary_loss_clip": 0.01183383, "auxiliary_loss_mlp": 0.01032191, "balance_loss_clip": 1.05655253, "balance_loss_mlp": 1.02355993, "epoch": 0.25792100042085014, "flos": 22820301539040.0, "grad_norm": 2.5281085033324913, "language_loss": 0.76311994, "learning_rate": 3.4795295321816804e-06, "loss": 0.7852757, "num_input_tokens_seen": 45850135, "step": 2145, "time_per_iteration": 2.6030538082122803 }, { "auxiliary_loss_clip": 0.01166437, "auxiliary_loss_mlp": 0.01029212, "balance_loss_clip": 1.05389762, "balance_loss_mlp": 1.02042031, "epoch": 0.2580412433114892, "flos": 18697289490720.0, "grad_norm": 2.0652427953568915, "language_loss": 0.90864444, "learning_rate": 3.47900527676272e-06, "loss": 0.930601, "num_input_tokens_seen": 45868470, "step": 2146, "time_per_iteration": 2.5721795558929443 }, { "auxiliary_loss_clip": 0.01210852, "auxiliary_loss_mlp": 0.01033603, "balance_loss_clip": 1.0624764, "balance_loss_mlp": 1.02497208, "epoch": 0.2581614862021283, "flos": 14283149754240.0, "grad_norm": 1.934928866995249, "language_loss": 0.8841821, "learning_rate": 3.478480796979835e-06, "loss": 0.90662664, "num_input_tokens_seen": 45886355, "step": 2147, "time_per_iteration": 2.50945782661438 }, { "auxiliary_loss_clip": 0.01173052, "auxiliary_loss_mlp": 0.01033322, "balance_loss_clip": 1.05554152, "balance_loss_mlp": 1.02469087, "epoch": 0.25828172909276736, "flos": 29498286275040.0, "grad_norm": 1.4980942975214284, "language_loss": 0.78092515, "learning_rate": 3.4779560929125894e-06, "loss": 0.80298889, "num_input_tokens_seen": 45907900, "step": 2148, "time_per_iteration": 3.4962518215179443 }, { "auxiliary_loss_clip": 0.01058043, "auxiliary_loss_mlp": 0.01002359, "balance_loss_clip": 1.01703548, "balance_loss_mlp": 1.00038052, "epoch": 0.2584019719834065, "flos": 67114398652800.0, "grad_norm": 0.6631828462905072, "language_loss": 0.56948245, "learning_rate": 3.4774311646405783e-06, "loss": 0.59008646, "num_input_tokens_seen": 45977805, "step": 2149, "time_per_iteration": 3.303210735321045 }, { "auxiliary_loss_clip": 0.01151658, "auxiliary_loss_mlp": 0.01030546, "balance_loss_clip": 1.04940557, "balance_loss_mlp": 1.0217011, "epoch": 0.2585222148740456, "flos": 22893559360800.0, "grad_norm": 2.100137026078624, "language_loss": 0.83888459, "learning_rate": 3.476906012243435e-06, "loss": 0.86070657, "num_input_tokens_seen": 45996715, "step": 2150, "time_per_iteration": 2.6283159255981445 }, { "auxiliary_loss_clip": 0.01181509, "auxiliary_loss_mlp": 0.01035012, "balance_loss_clip": 1.05525649, "balance_loss_mlp": 1.0262804, "epoch": 0.25864245776468464, "flos": 28909206676320.0, "grad_norm": 3.197231793785045, "language_loss": 0.81419069, "learning_rate": 3.476380635800824e-06, "loss": 0.83635587, "num_input_tokens_seen": 46017915, "step": 2151, "time_per_iteration": 3.542245626449585 }, { "auxiliary_loss_clip": 0.01175088, "auxiliary_loss_mlp": 0.01029259, "balance_loss_clip": 1.05686498, "balance_loss_mlp": 1.02021658, "epoch": 0.25876270065532375, "flos": 14793188900640.0, "grad_norm": 2.040839156242211, "language_loss": 0.85932696, "learning_rate": 3.475855035392444e-06, "loss": 0.88137043, "num_input_tokens_seen": 46033235, "step": 2152, "time_per_iteration": 2.5579400062561035 }, { "auxiliary_loss_clip": 0.01131969, "auxiliary_loss_mlp": 0.01029678, "balance_loss_clip": 1.05234599, "balance_loss_mlp": 1.02114248, "epoch": 0.25888294354596286, "flos": 60467828480160.0, "grad_norm": 1.8451286075707558, "language_loss": 0.71520257, "learning_rate": 3.475329211098029e-06, "loss": 0.73681903, "num_input_tokens_seen": 46056390, "step": 2153, "time_per_iteration": 3.0328521728515625 }, { "auxiliary_loss_clip": 0.01156078, "auxiliary_loss_mlp": 0.01029563, "balance_loss_clip": 1.05739915, "balance_loss_mlp": 1.020962, "epoch": 0.2590031864366019, "flos": 27851170545120.0, "grad_norm": 1.6121185319272802, "language_loss": 0.82234263, "learning_rate": 3.4748031629973453e-06, "loss": 0.84419906, "num_input_tokens_seen": 46077120, "step": 2154, "time_per_iteration": 2.709589958190918 }, { "auxiliary_loss_clip": 0.01047058, "auxiliary_loss_mlp": 0.01014178, "balance_loss_clip": 1.02024865, "balance_loss_mlp": 1.01256871, "epoch": 0.25912342932724103, "flos": 62422456420320.0, "grad_norm": 0.9154958656953744, "language_loss": 0.56697226, "learning_rate": 3.4742768911701944e-06, "loss": 0.58758461, "num_input_tokens_seen": 46139815, "step": 2155, "time_per_iteration": 3.3274264335632324 }, { "auxiliary_loss_clip": 0.01198888, "auxiliary_loss_mlp": 0.0102997, "balance_loss_clip": 1.06082726, "balance_loss_mlp": 1.02122569, "epoch": 0.25924367221788014, "flos": 12378841475040.0, "grad_norm": 2.777807090789425, "language_loss": 0.70091504, "learning_rate": 3.4737503956964113e-06, "loss": 0.72320366, "num_input_tokens_seen": 46152120, "step": 2156, "time_per_iteration": 3.344147205352783 }, { "auxiliary_loss_clip": 0.01169711, "auxiliary_loss_mlp": 0.01033628, "balance_loss_clip": 1.05298948, "balance_loss_mlp": 1.02447248, "epoch": 0.2593639151085192, "flos": 14575211283360.0, "grad_norm": 2.0864293006561496, "language_loss": 0.67281294, "learning_rate": 3.473223676655865e-06, "loss": 0.69484639, "num_input_tokens_seen": 46170120, "step": 2157, "time_per_iteration": 2.5621423721313477 }, { "auxiliary_loss_clip": 0.01170968, "auxiliary_loss_mlp": 0.01037262, "balance_loss_clip": 1.05291533, "balance_loss_mlp": 1.02788639, "epoch": 0.2594841579991583, "flos": 15230939983200.0, "grad_norm": 2.413259261426167, "language_loss": 0.79537231, "learning_rate": 3.472696734128459e-06, "loss": 0.81745458, "num_input_tokens_seen": 46187985, "step": 2158, "time_per_iteration": 2.562638521194458 }, { "auxiliary_loss_clip": 0.01197983, "auxiliary_loss_mlp": 0.0103964, "balance_loss_clip": 1.06160843, "balance_loss_mlp": 1.03094339, "epoch": 0.2596044008897974, "flos": 23623587474240.0, "grad_norm": 1.63260406600009, "language_loss": 0.76028347, "learning_rate": 3.4721695681941286e-06, "loss": 0.78265971, "num_input_tokens_seen": 46207025, "step": 2159, "time_per_iteration": 2.549671173095703 }, { "auxiliary_loss_clip": 0.01176902, "auxiliary_loss_mlp": 0.00763493, "balance_loss_clip": 1.05460358, "balance_loss_mlp": 1.0001471, "epoch": 0.25972464378043647, "flos": 13772284517760.0, "grad_norm": 2.2995199188977344, "language_loss": 0.82349265, "learning_rate": 3.471642178932845e-06, "loss": 0.84289658, "num_input_tokens_seen": 46225670, "step": 2160, "time_per_iteration": 2.6515653133392334 }, { "auxiliary_loss_clip": 0.01179611, "auxiliary_loss_mlp": 0.01024151, "balance_loss_clip": 1.05563247, "balance_loss_mlp": 1.01581252, "epoch": 0.2598448866710756, "flos": 19573581828960.0, "grad_norm": 3.0001240294811766, "language_loss": 0.89798963, "learning_rate": 3.471114566424613e-06, "loss": 0.92002726, "num_input_tokens_seen": 46244130, "step": 2161, "time_per_iteration": 2.614915370941162 }, { "auxiliary_loss_clip": 0.01175079, "auxiliary_loss_mlp": 0.0102772, "balance_loss_clip": 1.05699384, "balance_loss_mlp": 1.01885092, "epoch": 0.25996512956171464, "flos": 21653240682240.0, "grad_norm": 1.8452647852618198, "language_loss": 0.75447536, "learning_rate": 3.4705867307494715e-06, "loss": 0.77650332, "num_input_tokens_seen": 46263200, "step": 2162, "time_per_iteration": 2.632369041442871 }, { "auxiliary_loss_clip": 0.01194033, "auxiliary_loss_mlp": 0.0103286, "balance_loss_clip": 1.05582345, "balance_loss_mlp": 1.02375865, "epoch": 0.26008537245235375, "flos": 18223484168640.0, "grad_norm": 2.522968130013709, "language_loss": 0.84318721, "learning_rate": 3.470058671987492e-06, "loss": 0.8654561, "num_input_tokens_seen": 46281465, "step": 2163, "time_per_iteration": 2.5371780395507812 }, { "auxiliary_loss_clip": 0.01201334, "auxiliary_loss_mlp": 0.01034263, "balance_loss_clip": 1.06109476, "balance_loss_mlp": 1.02470803, "epoch": 0.26020561534299286, "flos": 24645389781120.0, "grad_norm": 1.7534735007654008, "language_loss": 0.84018451, "learning_rate": 3.4695303902187805e-06, "loss": 0.86254048, "num_input_tokens_seen": 46301020, "step": 2164, "time_per_iteration": 2.604464530944824 }, { "auxiliary_loss_clip": 0.01156202, "auxiliary_loss_mlp": 0.0103568, "balance_loss_clip": 1.04993141, "balance_loss_mlp": 1.02686405, "epoch": 0.2603258582336319, "flos": 25773666709440.0, "grad_norm": 1.7885000786401104, "language_loss": 0.783867, "learning_rate": 3.469001885523478e-06, "loss": 0.80578589, "num_input_tokens_seen": 46321740, "step": 2165, "time_per_iteration": 2.6634809970855713 }, { "auxiliary_loss_clip": 0.01207425, "auxiliary_loss_mlp": 0.01025718, "balance_loss_clip": 1.0589813, "balance_loss_mlp": 1.01749218, "epoch": 0.260446101124271, "flos": 28766318645760.0, "grad_norm": 1.7511738686390066, "language_loss": 0.80745721, "learning_rate": 3.4684731579817568e-06, "loss": 0.82978863, "num_input_tokens_seen": 46342730, "step": 2166, "time_per_iteration": 2.6035168170928955 }, { "auxiliary_loss_clip": 0.01131693, "auxiliary_loss_mlp": 0.01033109, "balance_loss_clip": 1.05379581, "balance_loss_mlp": 1.02477002, "epoch": 0.26056634401491013, "flos": 25666761084480.0, "grad_norm": 1.5259346883065943, "language_loss": 0.76717293, "learning_rate": 3.4679442076738247e-06, "loss": 0.78882098, "num_input_tokens_seen": 46362445, "step": 2167, "time_per_iteration": 2.7077443599700928 }, { "auxiliary_loss_clip": 0.01207982, "auxiliary_loss_mlp": 0.01032411, "balance_loss_clip": 1.05871761, "balance_loss_mlp": 1.02327979, "epoch": 0.2606865869055492, "flos": 27052769316480.0, "grad_norm": 1.843830403664379, "language_loss": 0.83467555, "learning_rate": 3.4674150346799245e-06, "loss": 0.85707951, "num_input_tokens_seen": 46382145, "step": 2168, "time_per_iteration": 2.5645503997802734 }, { "auxiliary_loss_clip": 0.01176574, "auxiliary_loss_mlp": 0.01037866, "balance_loss_clip": 1.0554316, "balance_loss_mlp": 1.02844262, "epoch": 0.2608068297961883, "flos": 17712618932160.0, "grad_norm": 2.5450535250338158, "language_loss": 0.79403484, "learning_rate": 3.4668856390803295e-06, "loss": 0.81617928, "num_input_tokens_seen": 46400025, "step": 2169, "time_per_iteration": 2.5655694007873535 }, { "auxiliary_loss_clip": 0.0117711, "auxiliary_loss_mlp": 0.01027285, "balance_loss_clip": 1.0528481, "balance_loss_mlp": 1.01903546, "epoch": 0.2609270726868274, "flos": 18551635854240.0, "grad_norm": 2.024818743490959, "language_loss": 0.89867127, "learning_rate": 3.4663560209553495e-06, "loss": 0.92071521, "num_input_tokens_seen": 46418090, "step": 2170, "time_per_iteration": 2.55823016166687 }, { "auxiliary_loss_clip": 0.0116771, "auxiliary_loss_mlp": 0.01030534, "balance_loss_clip": 1.05247986, "balance_loss_mlp": 1.02199864, "epoch": 0.26104731557746647, "flos": 21835702814400.0, "grad_norm": 1.6753775623816876, "language_loss": 0.79333043, "learning_rate": 3.4658261803853267e-06, "loss": 0.81531286, "num_input_tokens_seen": 46436015, "step": 2171, "time_per_iteration": 2.588521957397461 }, { "auxiliary_loss_clip": 0.01174859, "auxiliary_loss_mlp": 0.01039917, "balance_loss_clip": 1.0575434, "balance_loss_mlp": 1.03113151, "epoch": 0.2611675584681056, "flos": 21689654091360.0, "grad_norm": 2.2815408197668727, "language_loss": 0.80612016, "learning_rate": 3.4652961174506383e-06, "loss": 0.82826787, "num_input_tokens_seen": 46455885, "step": 2172, "time_per_iteration": 2.594520330429077 }, { "auxiliary_loss_clip": 0.01080386, "auxiliary_loss_mlp": 0.01002342, "balance_loss_clip": 1.01661134, "balance_loss_mlp": 1.00074446, "epoch": 0.2612878013587447, "flos": 71862113535840.0, "grad_norm": 0.9739618173444675, "language_loss": 0.58203071, "learning_rate": 3.464765832231694e-06, "loss": 0.60285801, "num_input_tokens_seen": 46510050, "step": 2173, "time_per_iteration": 3.1737592220306396 }, { "auxiliary_loss_clip": 0.0119257, "auxiliary_loss_mlp": 0.01039934, "balance_loss_clip": 1.05774486, "balance_loss_mlp": 1.03141069, "epoch": 0.26140804424938374, "flos": 20227514680800.0, "grad_norm": 2.3124939544112966, "language_loss": 0.70316029, "learning_rate": 3.4642353248089373e-06, "loss": 0.72548532, "num_input_tokens_seen": 46528810, "step": 2174, "time_per_iteration": 4.081676483154297 }, { "auxiliary_loss_clip": 0.01170223, "auxiliary_loss_mlp": 0.01029533, "balance_loss_clip": 1.05407643, "balance_loss_mlp": 1.02080667, "epoch": 0.26152828714002285, "flos": 25557089853600.0, "grad_norm": 1.748249348254914, "language_loss": 0.80285233, "learning_rate": 3.463704595262846e-06, "loss": 0.8248499, "num_input_tokens_seen": 46549690, "step": 2175, "time_per_iteration": 2.6429059505462646 }, { "auxiliary_loss_clip": 0.01158421, "auxiliary_loss_mlp": 0.01032757, "balance_loss_clip": 1.05272722, "balance_loss_mlp": 1.02444172, "epoch": 0.26164853003066196, "flos": 25446520698720.0, "grad_norm": 1.992482097993461, "language_loss": 0.70754671, "learning_rate": 3.463173643673931e-06, "loss": 0.72945851, "num_input_tokens_seen": 46572215, "step": 2176, "time_per_iteration": 2.689079761505127 }, { "auxiliary_loss_clip": 0.01086479, "auxiliary_loss_mlp": 0.01001694, "balance_loss_clip": 1.01591313, "balance_loss_mlp": 0.99996537, "epoch": 0.261768772921301, "flos": 53944590759840.0, "grad_norm": 0.8953302387004393, "language_loss": 0.63594759, "learning_rate": 3.4626424701227387e-06, "loss": 0.65682936, "num_input_tokens_seen": 46627275, "step": 2177, "time_per_iteration": 3.911675214767456 }, { "auxiliary_loss_clip": 0.01097513, "auxiliary_loss_mlp": 0.01000679, "balance_loss_clip": 1.01671171, "balance_loss_mlp": 0.99896222, "epoch": 0.26188901581194013, "flos": 70687617868320.0, "grad_norm": 0.8262293880897639, "language_loss": 0.55823803, "learning_rate": 3.4621110746898452e-06, "loss": 0.57921994, "num_input_tokens_seen": 46695135, "step": 2178, "time_per_iteration": 3.2322757244110107 }, { "auxiliary_loss_clip": 0.01195317, "auxiliary_loss_mlp": 0.0102896, "balance_loss_clip": 1.05849767, "balance_loss_mlp": 1.01981103, "epoch": 0.2620092587025792, "flos": 21069584544480.0, "grad_norm": 1.807421035836227, "language_loss": 0.74663055, "learning_rate": 3.4615794574558654e-06, "loss": 0.76887333, "num_input_tokens_seen": 46714145, "step": 2179, "time_per_iteration": 2.5684826374053955 }, { "auxiliary_loss_clip": 0.01178113, "auxiliary_loss_mlp": 0.0102813, "balance_loss_clip": 1.05555999, "balance_loss_mlp": 1.0196836, "epoch": 0.2621295015932183, "flos": 18369604725600.0, "grad_norm": 2.1773860909183527, "language_loss": 0.84053361, "learning_rate": 3.4610476185014436e-06, "loss": 0.86259604, "num_input_tokens_seen": 46731405, "step": 2180, "time_per_iteration": 2.589966297149658 }, { "auxiliary_loss_clip": 0.01207956, "auxiliary_loss_mlp": 0.01028559, "balance_loss_clip": 1.05761111, "balance_loss_mlp": 1.01949298, "epoch": 0.2622497444838574, "flos": 23660000883360.0, "grad_norm": 1.7600507091585147, "language_loss": 0.79432678, "learning_rate": 3.4605155579072597e-06, "loss": 0.81669194, "num_input_tokens_seen": 46751260, "step": 2181, "time_per_iteration": 2.5758285522460938 }, { "auxiliary_loss_clip": 0.011382, "auxiliary_loss_mlp": 0.01031106, "balance_loss_clip": 1.047647, "balance_loss_mlp": 1.02313054, "epoch": 0.26236998737449646, "flos": 22123813477920.0, "grad_norm": 1.7702773846002695, "language_loss": 0.71484089, "learning_rate": 3.459983275754027e-06, "loss": 0.73653388, "num_input_tokens_seen": 46770155, "step": 2182, "time_per_iteration": 3.558833122253418 }, { "auxiliary_loss_clip": 0.01205354, "auxiliary_loss_mlp": 0.01033602, "balance_loss_clip": 1.05796516, "balance_loss_mlp": 1.02495956, "epoch": 0.26249023026513557, "flos": 17895188815200.0, "grad_norm": 2.2310801177850452, "language_loss": 0.79710329, "learning_rate": 3.4594507721224918e-06, "loss": 0.81949282, "num_input_tokens_seen": 46788805, "step": 2183, "time_per_iteration": 2.5360584259033203 }, { "auxiliary_loss_clip": 0.01175541, "auxiliary_loss_mlp": 0.01030727, "balance_loss_clip": 1.05263615, "balance_loss_mlp": 1.02167928, "epoch": 0.2626104731557747, "flos": 18332939897760.0, "grad_norm": 1.7600067717933323, "language_loss": 0.81604958, "learning_rate": 3.4589180470934353e-06, "loss": 0.83811235, "num_input_tokens_seen": 46808670, "step": 2184, "time_per_iteration": 2.6157877445220947 }, { "auxiliary_loss_clip": 0.01197833, "auxiliary_loss_mlp": 0.01029746, "balance_loss_clip": 1.05518246, "balance_loss_mlp": 1.02072155, "epoch": 0.26273071604641374, "flos": 19317718207200.0, "grad_norm": 1.8403660563382316, "language_loss": 0.76817626, "learning_rate": 3.4583851007476713e-06, "loss": 0.79045206, "num_input_tokens_seen": 46827140, "step": 2185, "time_per_iteration": 2.562736749649048 }, { "auxiliary_loss_clip": 0.01165734, "auxiliary_loss_mlp": 0.01030936, "balance_loss_clip": 1.05504251, "balance_loss_mlp": 1.02191806, "epoch": 0.26285095893705285, "flos": 18327480519840.0, "grad_norm": 1.920238755671998, "language_loss": 0.68672872, "learning_rate": 3.4578519331660464e-06, "loss": 0.70869541, "num_input_tokens_seen": 46844135, "step": 2186, "time_per_iteration": 2.5906803607940674 }, { "auxiliary_loss_clip": 0.01190899, "auxiliary_loss_mlp": 0.01036145, "balance_loss_clip": 1.06081605, "balance_loss_mlp": 1.02790177, "epoch": 0.26297120182769196, "flos": 20193830960640.0, "grad_norm": 1.7103157145054457, "language_loss": 0.81534588, "learning_rate": 3.4573185444294426e-06, "loss": 0.83761632, "num_input_tokens_seen": 46862500, "step": 2187, "time_per_iteration": 2.5707955360412598 }, { "auxiliary_loss_clip": 0.01179184, "auxiliary_loss_mlp": 0.00763829, "balance_loss_clip": 1.05685067, "balance_loss_mlp": 1.00021291, "epoch": 0.263091444718331, "flos": 22418425111200.0, "grad_norm": 1.7284649995772459, "language_loss": 0.79019284, "learning_rate": 3.456784934618774e-06, "loss": 0.809623, "num_input_tokens_seen": 46883665, "step": 2188, "time_per_iteration": 2.621161460876465 }, { "auxiliary_loss_clip": 0.01174425, "auxiliary_loss_mlp": 0.0103094, "balance_loss_clip": 1.05378985, "balance_loss_mlp": 1.02165341, "epoch": 0.2632116876089701, "flos": 19024830588000.0, "grad_norm": 2.0491190746281354, "language_loss": 0.79875112, "learning_rate": 3.4562511038149897e-06, "loss": 0.82080483, "num_input_tokens_seen": 46899160, "step": 2189, "time_per_iteration": 2.5789248943328857 }, { "auxiliary_loss_clip": 0.0104639, "auxiliary_loss_mlp": 0.01000975, "balance_loss_clip": 1.01319897, "balance_loss_mlp": 0.99911547, "epoch": 0.26333193049960923, "flos": 67308066876480.0, "grad_norm": 0.8602437964802211, "language_loss": 0.57844222, "learning_rate": 3.4557170520990705e-06, "loss": 0.59891593, "num_input_tokens_seen": 46959835, "step": 2190, "time_per_iteration": 3.2560207843780518 }, { "auxiliary_loss_clip": 0.01188741, "auxiliary_loss_mlp": 0.01027116, "balance_loss_clip": 1.05781221, "balance_loss_mlp": 1.01938486, "epoch": 0.2634521733902483, "flos": 25048810638240.0, "grad_norm": 1.4621886603565732, "language_loss": 0.86393839, "learning_rate": 3.4551827795520324e-06, "loss": 0.88609695, "num_input_tokens_seen": 46982720, "step": 2191, "time_per_iteration": 2.619438886642456 }, { "auxiliary_loss_clip": 0.01192908, "auxiliary_loss_mlp": 0.01037196, "balance_loss_clip": 1.05696785, "balance_loss_mlp": 1.0287981, "epoch": 0.2635724162808874, "flos": 20594989049280.0, "grad_norm": 1.6730765717475655, "language_loss": 0.85136986, "learning_rate": 3.4546482862549226e-06, "loss": 0.87367088, "num_input_tokens_seen": 47003035, "step": 2192, "time_per_iteration": 2.5724360942840576 }, { "auxiliary_loss_clip": 0.01157006, "auxiliary_loss_mlp": 0.01035108, "balance_loss_clip": 1.05322814, "balance_loss_mlp": 1.02613139, "epoch": 0.2636926591715265, "flos": 19244639970240.0, "grad_norm": 2.9037587471385247, "language_loss": 0.78831249, "learning_rate": 3.4541135722888253e-06, "loss": 0.81023371, "num_input_tokens_seen": 47019625, "step": 2193, "time_per_iteration": 2.6139237880706787 }, { "auxiliary_loss_clip": 0.01206417, "auxiliary_loss_mlp": 0.01027847, "balance_loss_clip": 1.05874872, "balance_loss_mlp": 1.01896644, "epoch": 0.26381290206216557, "flos": 28804887072480.0, "grad_norm": 1.9067374750645, "language_loss": 0.80354691, "learning_rate": 3.453578637734854e-06, "loss": 0.82588959, "num_input_tokens_seen": 47040815, "step": 2194, "time_per_iteration": 2.5982651710510254 }, { "auxiliary_loss_clip": 0.01210743, "auxiliary_loss_mlp": 0.01033553, "balance_loss_clip": 1.06321871, "balance_loss_mlp": 1.02510107, "epoch": 0.2639331449528047, "flos": 25008913284000.0, "grad_norm": 1.8416752882001088, "language_loss": 0.78762436, "learning_rate": 3.4530434826741605e-06, "loss": 0.8100673, "num_input_tokens_seen": 47061755, "step": 2195, "time_per_iteration": 2.589970111846924 }, { "auxiliary_loss_clip": 0.01171476, "auxiliary_loss_mlp": 0.01031431, "balance_loss_clip": 1.0565114, "balance_loss_mlp": 1.02302718, "epoch": 0.26405338784344373, "flos": 46535781348960.0, "grad_norm": 2.1883122343281, "language_loss": 0.6866498, "learning_rate": 3.452508107187926e-06, "loss": 0.70867884, "num_input_tokens_seen": 47085130, "step": 2196, "time_per_iteration": 2.8077774047851562 }, { "auxiliary_loss_clip": 0.01133734, "auxiliary_loss_mlp": 0.0103842, "balance_loss_clip": 1.04957914, "balance_loss_mlp": 1.02936006, "epoch": 0.26417363073408284, "flos": 21179471277120.0, "grad_norm": 1.8709174386173817, "language_loss": 0.77490705, "learning_rate": 3.451972511357366e-06, "loss": 0.79662859, "num_input_tokens_seen": 47104675, "step": 2197, "time_per_iteration": 2.686136484146118 }, { "auxiliary_loss_clip": 0.01193136, "auxiliary_loss_mlp": 0.01035012, "balance_loss_clip": 1.06162214, "balance_loss_mlp": 1.02636921, "epoch": 0.26429387362472195, "flos": 22674755653440.0, "grad_norm": 2.0951145064112304, "language_loss": 0.85455149, "learning_rate": 3.45143669526373e-06, "loss": 0.87683302, "num_input_tokens_seen": 47124435, "step": 2198, "time_per_iteration": 2.593752861022949 }, { "auxiliary_loss_clip": 0.01077751, "auxiliary_loss_mlp": 0.00999264, "balance_loss_clip": 1.01501656, "balance_loss_mlp": 0.99755889, "epoch": 0.264414116515361, "flos": 67180544916480.0, "grad_norm": 0.7813803624895939, "language_loss": 0.63225871, "learning_rate": 3.450900658988302e-06, "loss": 0.65302891, "num_input_tokens_seen": 47185985, "step": 2199, "time_per_iteration": 3.887300729751587 }, { "auxiliary_loss_clip": 0.01168546, "auxiliary_loss_mlp": 0.01032724, "balance_loss_clip": 1.05572653, "balance_loss_mlp": 1.02388501, "epoch": 0.2645343594060001, "flos": 25664713817760.0, "grad_norm": 1.8068825675856173, "language_loss": 0.77588385, "learning_rate": 3.450364402612397e-06, "loss": 0.79789656, "num_input_tokens_seen": 47203140, "step": 2200, "time_per_iteration": 3.3535308837890625 }, { "auxiliary_loss_clip": 0.01175007, "auxiliary_loss_mlp": 0.01026681, "balance_loss_clip": 1.05630422, "balance_loss_mlp": 1.018116, "epoch": 0.26465460229663923, "flos": 22491826600800.0, "grad_norm": 2.2332045539217376, "language_loss": 0.83653802, "learning_rate": 3.449827926217366e-06, "loss": 0.8585549, "num_input_tokens_seen": 47222575, "step": 2201, "time_per_iteration": 2.596200704574585 }, { "auxiliary_loss_clip": 0.01180627, "auxiliary_loss_mlp": 0.01023841, "balance_loss_clip": 1.05361152, "balance_loss_mlp": 1.01523376, "epoch": 0.2647748451872783, "flos": 29388040372800.0, "grad_norm": 1.8547958540324814, "language_loss": 0.8051089, "learning_rate": 3.449291229884591e-06, "loss": 0.82715362, "num_input_tokens_seen": 47243815, "step": 2202, "time_per_iteration": 2.669959783554077 }, { "auxiliary_loss_clip": 0.01168436, "auxiliary_loss_mlp": 0.01032474, "balance_loss_clip": 1.0547688, "balance_loss_mlp": 1.023664, "epoch": 0.2648950880779174, "flos": 26797803618720.0, "grad_norm": 1.7309085111548521, "language_loss": 0.86704147, "learning_rate": 3.4487543136954887e-06, "loss": 0.88905048, "num_input_tokens_seen": 47263435, "step": 2203, "time_per_iteration": 3.5085885524749756 }, { "auxiliary_loss_clip": 0.01164048, "auxiliary_loss_mlp": 0.0103387, "balance_loss_clip": 1.05637717, "balance_loss_mlp": 1.02554929, "epoch": 0.2650153309685565, "flos": 28841013145920.0, "grad_norm": 4.10985682959584, "language_loss": 0.91006958, "learning_rate": 3.448217177731509e-06, "loss": 0.9320488, "num_input_tokens_seen": 47283920, "step": 2204, "time_per_iteration": 2.6709306240081787 }, { "auxiliary_loss_clip": 0.01176623, "auxiliary_loss_mlp": 0.01036487, "balance_loss_clip": 1.06096196, "balance_loss_mlp": 1.02805293, "epoch": 0.26513557385919556, "flos": 20303250772800.0, "grad_norm": 1.8297260616403503, "language_loss": 0.77754319, "learning_rate": 3.4476798220741348e-06, "loss": 0.79967427, "num_input_tokens_seen": 47302800, "step": 2205, "time_per_iteration": 2.5867056846618652 }, { "auxiliary_loss_clip": 0.01206974, "auxiliary_loss_mlp": 0.01028108, "balance_loss_clip": 1.06205988, "balance_loss_mlp": 1.02000141, "epoch": 0.26525581674983467, "flos": 17676277356960.0, "grad_norm": 1.6073264335556903, "language_loss": 0.78332376, "learning_rate": 3.4471422468048826e-06, "loss": 0.80567467, "num_input_tokens_seen": 47321525, "step": 2206, "time_per_iteration": 2.579375743865967 }, { "auxiliary_loss_clip": 0.01182187, "auxiliary_loss_mlp": 0.01027845, "balance_loss_clip": 1.05632889, "balance_loss_mlp": 1.01877928, "epoch": 0.2653760596404738, "flos": 26833750107360.0, "grad_norm": 2.4502818629225693, "language_loss": 0.72779799, "learning_rate": 3.4466044520053022e-06, "loss": 0.74989831, "num_input_tokens_seen": 47340530, "step": 2207, "time_per_iteration": 2.610440731048584 }, { "auxiliary_loss_clip": 0.01165817, "auxiliary_loss_mlp": 0.01031932, "balance_loss_clip": 1.05314302, "balance_loss_mlp": 1.02359951, "epoch": 0.26549630253111284, "flos": 22782164115840.0, "grad_norm": 1.7484832257239988, "language_loss": 0.60433269, "learning_rate": 3.446066437756977e-06, "loss": 0.62631017, "num_input_tokens_seen": 47359735, "step": 2208, "time_per_iteration": 3.4864532947540283 }, { "auxiliary_loss_clip": 0.01177223, "auxiliary_loss_mlp": 0.01027678, "balance_loss_clip": 1.05788064, "balance_loss_mlp": 1.01972628, "epoch": 0.26561654542175195, "flos": 23550006399840.0, "grad_norm": 2.0385806113638854, "language_loss": 0.75155354, "learning_rate": 3.4455282041415224e-06, "loss": 0.77360249, "num_input_tokens_seen": 47378945, "step": 2209, "time_per_iteration": 2.6180615425109863 }, { "auxiliary_loss_clip": 0.01165108, "auxiliary_loss_mlp": 0.01027242, "balance_loss_clip": 1.05568409, "balance_loss_mlp": 1.0189507, "epoch": 0.265736788312391, "flos": 26906684676480.0, "grad_norm": 3.993960197601789, "language_loss": 0.87237024, "learning_rate": 3.4449897512405894e-06, "loss": 0.89429384, "num_input_tokens_seen": 47398095, "step": 2210, "time_per_iteration": 2.67242693901062 }, { "auxiliary_loss_clip": 0.0112345, "auxiliary_loss_mlp": 0.00762861, "balance_loss_clip": 1.04576695, "balance_loss_mlp": 1.00010157, "epoch": 0.2658570312030301, "flos": 23477143664640.0, "grad_norm": 2.1457400123410726, "language_loss": 0.75399679, "learning_rate": 3.444451079135859e-06, "loss": 0.77285987, "num_input_tokens_seen": 47417605, "step": 2211, "time_per_iteration": 2.7140915393829346 }, { "auxiliary_loss_clip": 0.01140427, "auxiliary_loss_mlp": 0.00762808, "balance_loss_clip": 1.05131078, "balance_loss_mlp": 1.00000882, "epoch": 0.2659772740936692, "flos": 21866405426880.0, "grad_norm": 2.640738055421857, "language_loss": 0.73978567, "learning_rate": 3.4439121879090493e-06, "loss": 0.75881803, "num_input_tokens_seen": 47435385, "step": 2212, "time_per_iteration": 2.6820013523101807 }, { "auxiliary_loss_clip": 0.01184895, "auxiliary_loss_mlp": 0.01034467, "balance_loss_clip": 1.05853462, "balance_loss_mlp": 1.02608109, "epoch": 0.2660975169843083, "flos": 19793103875520.0, "grad_norm": 2.671936539714208, "language_loss": 0.83573961, "learning_rate": 3.4433730776419082e-06, "loss": 0.85793322, "num_input_tokens_seen": 47454310, "step": 2213, "time_per_iteration": 2.623704195022583 }, { "auxiliary_loss_clip": 0.01196755, "auxiliary_loss_mlp": 0.00763938, "balance_loss_clip": 1.05852151, "balance_loss_mlp": 1.0001421, "epoch": 0.2662177598749474, "flos": 29018985658080.0, "grad_norm": 2.1216567766677215, "language_loss": 0.80271316, "learning_rate": 3.4428337484162183e-06, "loss": 0.8223201, "num_input_tokens_seen": 47475120, "step": 2214, "time_per_iteration": 2.6497647762298584 }, { "auxiliary_loss_clip": 0.0117254, "auxiliary_loss_mlp": 0.01029604, "balance_loss_clip": 1.05387449, "balance_loss_mlp": 1.02132535, "epoch": 0.2663380027655865, "flos": 21762552743520.0, "grad_norm": 1.8610913022823796, "language_loss": 0.84847134, "learning_rate": 3.442294200313797e-06, "loss": 0.87049282, "num_input_tokens_seen": 47493150, "step": 2215, "time_per_iteration": 2.601527690887451 }, { "auxiliary_loss_clip": 0.01095734, "auxiliary_loss_mlp": 0.01003348, "balance_loss_clip": 1.01498461, "balance_loss_mlp": 1.0016669, "epoch": 0.26645824565622556, "flos": 66980346194400.0, "grad_norm": 0.7608055608804148, "language_loss": 0.52766502, "learning_rate": 3.4417544334164916e-06, "loss": 0.54865587, "num_input_tokens_seen": 47557295, "step": 2216, "time_per_iteration": 3.1475367546081543 }, { "auxiliary_loss_clip": 0.01162762, "auxiliary_loss_mlp": 0.0103196, "balance_loss_clip": 1.05818641, "balance_loss_mlp": 1.02411032, "epoch": 0.26657848854686467, "flos": 25264202234400.0, "grad_norm": 1.5593342261630256, "language_loss": 0.77436143, "learning_rate": 3.4412144478061854e-06, "loss": 0.7963087, "num_input_tokens_seen": 47579705, "step": 2217, "time_per_iteration": 2.690127372741699 }, { "auxiliary_loss_clip": 0.01102458, "auxiliary_loss_mlp": 0.01033913, "balance_loss_clip": 1.04710603, "balance_loss_mlp": 1.02465653, "epoch": 0.2666987314375038, "flos": 23696773462080.0, "grad_norm": 2.002229877440907, "language_loss": 0.75310999, "learning_rate": 3.4406742435647925e-06, "loss": 0.77447367, "num_input_tokens_seen": 47599770, "step": 2218, "time_per_iteration": 2.95649790763855 }, { "auxiliary_loss_clip": 0.01193538, "auxiliary_loss_mlp": 0.0102888, "balance_loss_clip": 1.06225264, "balance_loss_mlp": 1.02132821, "epoch": 0.26681897432814283, "flos": 27048962118720.0, "grad_norm": 2.0616909302398922, "language_loss": 0.78638935, "learning_rate": 3.440133820774263e-06, "loss": 0.80861354, "num_input_tokens_seen": 47619580, "step": 2219, "time_per_iteration": 2.8543541431427 }, { "auxiliary_loss_clip": 0.01180984, "auxiliary_loss_mlp": 0.0102923, "balance_loss_clip": 1.0560441, "balance_loss_mlp": 1.02066517, "epoch": 0.26693921721878194, "flos": 28985948443200.0, "grad_norm": 2.5633392681506573, "language_loss": 0.81569171, "learning_rate": 3.439593179516578e-06, "loss": 0.83779383, "num_input_tokens_seen": 47639490, "step": 2220, "time_per_iteration": 2.646719455718994 }, { "auxiliary_loss_clip": 0.0118101, "auxiliary_loss_mlp": 0.0102747, "balance_loss_clip": 1.05753779, "balance_loss_mlp": 1.01869059, "epoch": 0.26705946010942105, "flos": 21507838464480.0, "grad_norm": 2.0578327282180244, "language_loss": 0.811068, "learning_rate": 3.4390523198737524e-06, "loss": 0.83315283, "num_input_tokens_seen": 47658650, "step": 2221, "time_per_iteration": 2.600393295288086 }, { "auxiliary_loss_clip": 0.01208309, "auxiliary_loss_mlp": 0.00763462, "balance_loss_clip": 1.0608561, "balance_loss_mlp": 1.0000602, "epoch": 0.2671797030000601, "flos": 21471281387520.0, "grad_norm": 1.7796035760286832, "language_loss": 0.73442674, "learning_rate": 3.4385112419278333e-06, "loss": 0.75414449, "num_input_tokens_seen": 47679875, "step": 2222, "time_per_iteration": 2.5543301105499268 }, { "auxiliary_loss_clip": 0.01086591, "auxiliary_loss_mlp": 0.0100595, "balance_loss_clip": 1.01489806, "balance_loss_mlp": 1.00420916, "epoch": 0.2672999458906992, "flos": 64189940246880.0, "grad_norm": 0.7900519544321867, "language_loss": 0.64838916, "learning_rate": 3.4379699457609033e-06, "loss": 0.66931456, "num_input_tokens_seen": 47737700, "step": 2223, "time_per_iteration": 2.9878249168395996 }, { "auxiliary_loss_clip": 0.01167368, "auxiliary_loss_mlp": 0.01027899, "balance_loss_clip": 1.05351949, "balance_loss_mlp": 1.01936328, "epoch": 0.26742018878133833, "flos": 16909045661280.0, "grad_norm": 1.7224055863865728, "language_loss": 0.89888084, "learning_rate": 3.4374284314550755e-06, "loss": 0.92083347, "num_input_tokens_seen": 47756740, "step": 2224, "time_per_iteration": 2.588132619857788 }, { "auxiliary_loss_clip": 0.01204366, "auxiliary_loss_mlp": 0.0103164, "balance_loss_clip": 1.05855823, "balance_loss_mlp": 1.02324748, "epoch": 0.2675404316719774, "flos": 20667564448800.0, "grad_norm": 2.2798986549159093, "language_loss": 0.8103736, "learning_rate": 3.436886699092498e-06, "loss": 0.83273369, "num_input_tokens_seen": 47775255, "step": 2225, "time_per_iteration": 3.2968926429748535 }, { "auxiliary_loss_clip": 0.01211717, "auxiliary_loss_mlp": 0.0103348, "balance_loss_clip": 1.06182659, "balance_loss_mlp": 1.02490926, "epoch": 0.2676606745626165, "flos": 17485015569600.0, "grad_norm": 3.8725170393716115, "language_loss": 0.71340835, "learning_rate": 3.4363447487553502e-06, "loss": 0.73586035, "num_input_tokens_seen": 47788570, "step": 2226, "time_per_iteration": 3.2782115936279297 }, { "auxiliary_loss_clip": 0.01172609, "auxiliary_loss_mlp": 0.0103319, "balance_loss_clip": 1.05779076, "balance_loss_mlp": 1.02453589, "epoch": 0.26778091745325555, "flos": 27852679057440.0, "grad_norm": 1.929699620741972, "language_loss": 0.7792483, "learning_rate": 3.4358025805258455e-06, "loss": 0.80130625, "num_input_tokens_seen": 47808275, "step": 2227, "time_per_iteration": 2.6294198036193848 }, { "auxiliary_loss_clip": 0.01152326, "auxiliary_loss_mlp": 0.01026145, "balance_loss_clip": 1.0520618, "balance_loss_mlp": 1.01683497, "epoch": 0.26790116034389466, "flos": 20955998364960.0, "grad_norm": 2.0202296198907135, "language_loss": 0.83368182, "learning_rate": 3.435260194486232e-06, "loss": 0.85546649, "num_input_tokens_seen": 47826245, "step": 2228, "time_per_iteration": 3.5802695751190186 }, { "auxiliary_loss_clip": 0.0117564, "auxiliary_loss_mlp": 0.01028265, "balance_loss_clip": 1.05737638, "balance_loss_mlp": 1.01922321, "epoch": 0.2680214032345338, "flos": 18040662866880.0, "grad_norm": 2.1163672309926005, "language_loss": 0.81892776, "learning_rate": 3.4347175907187875e-06, "loss": 0.84096682, "num_input_tokens_seen": 47843235, "step": 2229, "time_per_iteration": 2.5778470039367676 }, { "auxiliary_loss_clip": 0.01191561, "auxiliary_loss_mlp": 0.01028856, "balance_loss_clip": 1.05981147, "balance_loss_mlp": 1.02051127, "epoch": 0.26814164612517283, "flos": 22419430786080.0, "grad_norm": 1.9037472406941551, "language_loss": 0.87766039, "learning_rate": 3.4341747693058254e-06, "loss": 0.89986455, "num_input_tokens_seen": 47861710, "step": 2230, "time_per_iteration": 2.579298973083496 }, { "auxiliary_loss_clip": 0.01097016, "auxiliary_loss_mlp": 0.01027294, "balance_loss_clip": 1.04840565, "balance_loss_mlp": 1.01928961, "epoch": 0.26826188901581194, "flos": 35627376102240.0, "grad_norm": 2.0676769295890565, "language_loss": 0.77060318, "learning_rate": 3.4336317303296916e-06, "loss": 0.79184633, "num_input_tokens_seen": 47882685, "step": 2231, "time_per_iteration": 2.881765842437744 }, { "auxiliary_loss_clip": 0.0118945, "auxiliary_loss_mlp": 0.01025938, "balance_loss_clip": 1.05884671, "balance_loss_mlp": 1.01814747, "epoch": 0.26838213190645105, "flos": 17639792113920.0, "grad_norm": 2.314099230984554, "language_loss": 0.74653143, "learning_rate": 3.4330884738727635e-06, "loss": 0.76868528, "num_input_tokens_seen": 47900860, "step": 2232, "time_per_iteration": 2.5625646114349365 }, { "auxiliary_loss_clip": 0.01137753, "auxiliary_loss_mlp": 0.01031671, "balance_loss_clip": 1.05284047, "balance_loss_mlp": 1.02321362, "epoch": 0.2685023747970901, "flos": 22674827487360.0, "grad_norm": 2.267115123378417, "language_loss": 0.70616949, "learning_rate": 3.4325450000174535e-06, "loss": 0.72786367, "num_input_tokens_seen": 47917500, "step": 2233, "time_per_iteration": 2.6785473823547363 }, { "auxiliary_loss_clip": 0.01135013, "auxiliary_loss_mlp": 0.0103122, "balance_loss_clip": 1.05011523, "balance_loss_mlp": 1.02310133, "epoch": 0.2686226176877292, "flos": 20120537221920.0, "grad_norm": 1.6728090281414143, "language_loss": 0.74160331, "learning_rate": 3.4320013088462067e-06, "loss": 0.76326567, "num_input_tokens_seen": 47934860, "step": 2234, "time_per_iteration": 3.507112741470337 }, { "auxiliary_loss_clip": 0.01165882, "auxiliary_loss_mlp": 0.01030203, "balance_loss_clip": 1.05452204, "balance_loss_mlp": 1.0222578, "epoch": 0.2687428605783683, "flos": 21872044389600.0, "grad_norm": 1.4308487942381978, "language_loss": 0.81791842, "learning_rate": 3.431457400441499e-06, "loss": 0.83987927, "num_input_tokens_seen": 47955255, "step": 2235, "time_per_iteration": 2.6453235149383545 }, { "auxiliary_loss_clip": 0.01031146, "auxiliary_loss_mlp": 0.01004095, "balance_loss_clip": 1.01799476, "balance_loss_mlp": 1.00254488, "epoch": 0.2688631034690074, "flos": 69943330721760.0, "grad_norm": 0.913772616725529, "language_loss": 0.61057925, "learning_rate": 3.4309132748858424e-06, "loss": 0.63093162, "num_input_tokens_seen": 48016245, "step": 2236, "time_per_iteration": 3.2660484313964844 }, { "auxiliary_loss_clip": 0.01189049, "auxiliary_loss_mlp": 0.01027217, "balance_loss_clip": 1.0603776, "balance_loss_mlp": 1.01869345, "epoch": 0.2689833463596465, "flos": 22856643114240.0, "grad_norm": 1.6781935546344684, "language_loss": 0.83846402, "learning_rate": 3.430368932261779e-06, "loss": 0.86062658, "num_input_tokens_seen": 48036600, "step": 2237, "time_per_iteration": 2.549621820449829 }, { "auxiliary_loss_clip": 0.01174337, "auxiliary_loss_mlp": 0.01032722, "balance_loss_clip": 1.05686939, "balance_loss_mlp": 1.02407908, "epoch": 0.2691035892502856, "flos": 17200245183360.0, "grad_norm": 2.198097592003219, "language_loss": 0.75358331, "learning_rate": 3.429824372651886e-06, "loss": 0.77565396, "num_input_tokens_seen": 48054750, "step": 2238, "time_per_iteration": 2.526038408279419 }, { "auxiliary_loss_clip": 0.01156994, "auxiliary_loss_mlp": 0.01028671, "balance_loss_clip": 1.05571437, "balance_loss_mlp": 1.02011204, "epoch": 0.26922383214092466, "flos": 17747487912000.0, "grad_norm": 1.7901939406771816, "language_loss": 0.82912821, "learning_rate": 3.4292795961387732e-06, "loss": 0.85098487, "num_input_tokens_seen": 48072650, "step": 2239, "time_per_iteration": 2.6083478927612305 }, { "auxiliary_loss_clip": 0.01204537, "auxiliary_loss_mlp": 0.01028818, "balance_loss_clip": 1.05900252, "balance_loss_mlp": 1.02013993, "epoch": 0.26934407503156377, "flos": 16173378585120.0, "grad_norm": 2.261548917016537, "language_loss": 0.87455428, "learning_rate": 3.4287346028050818e-06, "loss": 0.89688784, "num_input_tokens_seen": 48088720, "step": 2240, "time_per_iteration": 2.5268983840942383 }, { "auxiliary_loss_clip": 0.01175301, "auxiliary_loss_mlp": 0.01027775, "balance_loss_clip": 1.05607104, "balance_loss_mlp": 1.0198772, "epoch": 0.2694643179222028, "flos": 23732899535520.0, "grad_norm": 1.4999508092668647, "language_loss": 0.793302, "learning_rate": 3.4281893927334866e-06, "loss": 0.81533271, "num_input_tokens_seen": 48108630, "step": 2241, "time_per_iteration": 2.597935914993286 }, { "auxiliary_loss_clip": 0.01190564, "auxiliary_loss_mlp": 0.01031126, "balance_loss_clip": 1.05830503, "balance_loss_mlp": 1.02302599, "epoch": 0.26958456081284193, "flos": 24718144765440.0, "grad_norm": 2.628257324021802, "language_loss": 0.75241768, "learning_rate": 3.4276439660066963e-06, "loss": 0.7746346, "num_input_tokens_seen": 48128330, "step": 2242, "time_per_iteration": 2.6156346797943115 }, { "auxiliary_loss_clip": 0.01202583, "auxiliary_loss_mlp": 0.01031845, "balance_loss_clip": 1.05917335, "balance_loss_mlp": 1.02388763, "epoch": 0.26970480370348104, "flos": 18112591761120.0, "grad_norm": 2.30825952411156, "language_loss": 0.84169251, "learning_rate": 3.427098322707452e-06, "loss": 0.8640368, "num_input_tokens_seen": 48144295, "step": 2243, "time_per_iteration": 2.513989210128784 }, { "auxiliary_loss_clip": 0.01188476, "auxiliary_loss_mlp": 0.01034125, "balance_loss_clip": 1.05923378, "balance_loss_mlp": 1.02642441, "epoch": 0.2698250465941201, "flos": 10816548828000.0, "grad_norm": 1.8531594506147695, "language_loss": 0.89441389, "learning_rate": 3.426552462918526e-06, "loss": 0.91663986, "num_input_tokens_seen": 48162230, "step": 2244, "time_per_iteration": 2.5401451587677 }, { "auxiliary_loss_clip": 0.0120502, "auxiliary_loss_mlp": 0.01035168, "balance_loss_clip": 1.06202757, "balance_loss_mlp": 1.02729392, "epoch": 0.2699452894847592, "flos": 17308120566240.0, "grad_norm": 2.370098449649009, "language_loss": 0.73094338, "learning_rate": 3.426006386722726e-06, "loss": 0.75334525, "num_input_tokens_seen": 48180290, "step": 2245, "time_per_iteration": 2.5103466510772705 }, { "auxiliary_loss_clip": 0.01164605, "auxiliary_loss_mlp": 0.01042246, "balance_loss_clip": 1.0586797, "balance_loss_mlp": 1.03354371, "epoch": 0.2700655323753983, "flos": 18078153784800.0, "grad_norm": 3.1805099876298053, "language_loss": 0.91993296, "learning_rate": 3.4254600942028914e-06, "loss": 0.94200146, "num_input_tokens_seen": 48198165, "step": 2246, "time_per_iteration": 2.6034836769104004 }, { "auxiliary_loss_clip": 0.01173256, "auxiliary_loss_mlp": 0.01030316, "balance_loss_clip": 1.05891967, "balance_loss_mlp": 1.02227592, "epoch": 0.2701857752660374, "flos": 18186639756000.0, "grad_norm": 2.276658471917496, "language_loss": 0.82919639, "learning_rate": 3.424913585441893e-06, "loss": 0.85123205, "num_input_tokens_seen": 48216000, "step": 2247, "time_per_iteration": 2.5565619468688965 }, { "auxiliary_loss_clip": 0.01187514, "auxiliary_loss_mlp": 0.01034686, "balance_loss_clip": 1.05876398, "balance_loss_mlp": 1.02633584, "epoch": 0.2703060181566765, "flos": 16319499142080.0, "grad_norm": 1.7957834371695125, "language_loss": 0.87003684, "learning_rate": 3.4243668605226374e-06, "loss": 0.89225882, "num_input_tokens_seen": 48233025, "step": 2248, "time_per_iteration": 2.551453113555908 }, { "auxiliary_loss_clip": 0.01161231, "auxiliary_loss_mlp": 0.0103366, "balance_loss_clip": 1.05685234, "balance_loss_mlp": 1.02551782, "epoch": 0.2704262610473156, "flos": 19572360652320.0, "grad_norm": 3.635714159845727, "language_loss": 0.82536411, "learning_rate": 3.423819919528061e-06, "loss": 0.84731305, "num_input_tokens_seen": 48251110, "step": 2249, "time_per_iteration": 2.5828909873962402 }, { "auxiliary_loss_clip": 0.01151059, "auxiliary_loss_mlp": 0.01029008, "balance_loss_clip": 1.05093765, "balance_loss_mlp": 1.02060318, "epoch": 0.27054650393795465, "flos": 20740750436640.0, "grad_norm": 1.738572491153402, "language_loss": 0.78522831, "learning_rate": 3.4232727625411355e-06, "loss": 0.80702901, "num_input_tokens_seen": 48270215, "step": 2250, "time_per_iteration": 2.6843154430389404 }, { "auxiliary_loss_clip": 0.0112266, "auxiliary_loss_mlp": 0.01025849, "balance_loss_clip": 1.04875517, "balance_loss_mlp": 1.01829147, "epoch": 0.27066674682859376, "flos": 18658325977440.0, "grad_norm": 1.5718687180280688, "language_loss": 0.86228949, "learning_rate": 3.4227253896448626e-06, "loss": 0.88377464, "num_input_tokens_seen": 48288075, "step": 2251, "time_per_iteration": 2.6571297645568848 }, { "auxiliary_loss_clip": 0.0120288, "auxiliary_loss_mlp": 0.0102967, "balance_loss_clip": 1.05757666, "balance_loss_mlp": 1.02123046, "epoch": 0.2707869897192329, "flos": 23002763671200.0, "grad_norm": 2.6918939184502726, "language_loss": 0.82948256, "learning_rate": 3.42217780092228e-06, "loss": 0.85180795, "num_input_tokens_seen": 48306415, "step": 2252, "time_per_iteration": 4.035227298736572 }, { "auxiliary_loss_clip": 0.01070038, "auxiliary_loss_mlp": 0.01000359, "balance_loss_clip": 1.01884103, "balance_loss_mlp": 0.99841541, "epoch": 0.27090723260987193, "flos": 58323250928160.0, "grad_norm": 0.7959370558921464, "language_loss": 0.60409194, "learning_rate": 3.421629996456456e-06, "loss": 0.62479591, "num_input_tokens_seen": 48365035, "step": 2253, "time_per_iteration": 3.123652458190918 }, { "auxiliary_loss_clip": 0.01186911, "auxiliary_loss_mlp": 0.01024521, "balance_loss_clip": 1.05658138, "balance_loss_mlp": 1.01523495, "epoch": 0.27102747550051104, "flos": 11984543525760.0, "grad_norm": 1.9780504442182607, "language_loss": 0.82446629, "learning_rate": 3.421081976330491e-06, "loss": 0.84658062, "num_input_tokens_seen": 48383550, "step": 2254, "time_per_iteration": 3.4427056312561035 }, { "auxiliary_loss_clip": 0.01171927, "auxiliary_loss_mlp": 0.01031999, "balance_loss_clip": 1.05336249, "balance_loss_mlp": 1.02398217, "epoch": 0.27114771839115015, "flos": 19900404587040.0, "grad_norm": 1.9443387156032874, "language_loss": 0.87745041, "learning_rate": 3.4205337406275207e-06, "loss": 0.89948964, "num_input_tokens_seen": 48403670, "step": 2255, "time_per_iteration": 2.6263043880462646 }, { "auxiliary_loss_clip": 0.01200944, "auxiliary_loss_mlp": 0.01029003, "balance_loss_clip": 1.0554949, "balance_loss_mlp": 1.02099848, "epoch": 0.2712679612817892, "flos": 18331970139840.0, "grad_norm": 2.2562535949180673, "language_loss": 0.75482142, "learning_rate": 3.4199852894307114e-06, "loss": 0.77712095, "num_input_tokens_seen": 48420420, "step": 2256, "time_per_iteration": 2.510234832763672 }, { "auxiliary_loss_clip": 0.01128114, "auxiliary_loss_mlp": 0.0102832, "balance_loss_clip": 1.04973769, "balance_loss_mlp": 1.01982045, "epoch": 0.2713882041724283, "flos": 24460305710880.0, "grad_norm": 4.422611837267434, "language_loss": 0.78636724, "learning_rate": 3.419436622823262e-06, "loss": 0.80793154, "num_input_tokens_seen": 48441140, "step": 2257, "time_per_iteration": 2.70918345451355 }, { "auxiliary_loss_clip": 0.01173106, "auxiliary_loss_mlp": 0.01030993, "balance_loss_clip": 1.05690098, "balance_loss_mlp": 1.02349472, "epoch": 0.27150844706306737, "flos": 23039320748160.0, "grad_norm": 1.555834049982802, "language_loss": 0.74032086, "learning_rate": 3.4188877408884063e-06, "loss": 0.76236188, "num_input_tokens_seen": 48461845, "step": 2258, "time_per_iteration": 2.6279091835021973 }, { "auxiliary_loss_clip": 0.01166293, "auxiliary_loss_mlp": 0.01031081, "balance_loss_clip": 1.05369616, "balance_loss_mlp": 1.02278948, "epoch": 0.2716286899537065, "flos": 22563647744160.0, "grad_norm": 2.512532776601696, "language_loss": 0.65409893, "learning_rate": 3.4183386437094088e-06, "loss": 0.67607266, "num_input_tokens_seen": 48478510, "step": 2259, "time_per_iteration": 2.5961215496063232 }, { "auxiliary_loss_clip": 0.01174323, "auxiliary_loss_mlp": 0.01031409, "balance_loss_clip": 1.05285513, "balance_loss_mlp": 1.02248049, "epoch": 0.2717489328443456, "flos": 13115047305600.0, "grad_norm": 2.3807401525342904, "language_loss": 0.82333708, "learning_rate": 3.417789331369565e-06, "loss": 0.84539437, "num_input_tokens_seen": 48494300, "step": 2260, "time_per_iteration": 3.4198262691497803 }, { "auxiliary_loss_clip": 0.01207094, "auxiliary_loss_mlp": 0.01033318, "balance_loss_clip": 1.06014979, "balance_loss_mlp": 1.02396011, "epoch": 0.27186917573498465, "flos": 29278692394560.0, "grad_norm": 2.045742231822123, "language_loss": 0.90937793, "learning_rate": 3.4172398039522088e-06, "loss": 0.93178207, "num_input_tokens_seen": 48515585, "step": 2261, "time_per_iteration": 2.581037998199463 }, { "auxiliary_loss_clip": 0.01189866, "auxiliary_loss_mlp": 0.01028592, "balance_loss_clip": 1.0566256, "balance_loss_mlp": 1.02015197, "epoch": 0.27198941862562376, "flos": 26032224103200.0, "grad_norm": 1.6951407473991515, "language_loss": 0.7969166, "learning_rate": 3.4166900615407e-06, "loss": 0.81910121, "num_input_tokens_seen": 48533500, "step": 2262, "time_per_iteration": 2.60638689994812 }, { "auxiliary_loss_clip": 0.01187608, "auxiliary_loss_mlp": 0.01032105, "balance_loss_clip": 1.05807233, "balance_loss_mlp": 1.0237726, "epoch": 0.27210966151626287, "flos": 32780988390720.0, "grad_norm": 2.0158143289771817, "language_loss": 0.75200325, "learning_rate": 3.416140104218436e-06, "loss": 0.77420032, "num_input_tokens_seen": 48552865, "step": 2263, "time_per_iteration": 2.645735740661621 }, { "auxiliary_loss_clip": 0.0107329, "auxiliary_loss_mlp": 0.00754576, "balance_loss_clip": 1.01764441, "balance_loss_mlp": 0.9998793, "epoch": 0.2722299044069019, "flos": 65471054203680.0, "grad_norm": 0.8443465727474472, "language_loss": 0.69728422, "learning_rate": 3.4155899320688437e-06, "loss": 0.71556282, "num_input_tokens_seen": 48618940, "step": 2264, "time_per_iteration": 3.219132900238037 }, { "auxiliary_loss_clip": 0.01135699, "auxiliary_loss_mlp": 0.01036062, "balance_loss_clip": 1.05473411, "balance_loss_mlp": 1.02736568, "epoch": 0.27235014729754103, "flos": 15334146161280.0, "grad_norm": 2.2311565420338737, "language_loss": 0.74212384, "learning_rate": 3.415039545175384e-06, "loss": 0.76384139, "num_input_tokens_seen": 48634665, "step": 2265, "time_per_iteration": 2.658003330230713 }, { "auxiliary_loss_clip": 0.01191487, "auxiliary_loss_mlp": 0.01024942, "balance_loss_clip": 1.0569936, "balance_loss_mlp": 1.0166744, "epoch": 0.27247039018818014, "flos": 21872367642240.0, "grad_norm": 2.1011581626152034, "language_loss": 0.647627, "learning_rate": 3.414488943621551e-06, "loss": 0.66979134, "num_input_tokens_seen": 48653330, "step": 2266, "time_per_iteration": 2.5785951614379883 }, { "auxiliary_loss_clip": 0.01182543, "auxiliary_loss_mlp": 0.01028743, "balance_loss_clip": 1.05514109, "balance_loss_mlp": 1.02007651, "epoch": 0.2725906330788192, "flos": 18695493642720.0, "grad_norm": 2.2687829744480084, "language_loss": 0.73954105, "learning_rate": 3.41393812749087e-06, "loss": 0.7616539, "num_input_tokens_seen": 48671375, "step": 2267, "time_per_iteration": 2.54036808013916 }, { "auxiliary_loss_clip": 0.01175552, "auxiliary_loss_mlp": 0.01031909, "balance_loss_clip": 1.0589335, "balance_loss_mlp": 1.0232482, "epoch": 0.2727108759694583, "flos": 17886101824320.0, "grad_norm": 2.3297419315447323, "language_loss": 0.71875894, "learning_rate": 3.4133870968668984e-06, "loss": 0.74083352, "num_input_tokens_seen": 48686175, "step": 2268, "time_per_iteration": 2.564903974533081 }, { "auxiliary_loss_clip": 0.01177186, "auxiliary_loss_mlp": 0.01043675, "balance_loss_clip": 1.05663657, "balance_loss_mlp": 1.03478742, "epoch": 0.2728311188600974, "flos": 24461670555360.0, "grad_norm": 1.6592633950311682, "language_loss": 0.78405094, "learning_rate": 3.412835851833229e-06, "loss": 0.80625951, "num_input_tokens_seen": 48708370, "step": 2269, "time_per_iteration": 2.6314237117767334 }, { "auxiliary_loss_clip": 0.01184956, "auxiliary_loss_mlp": 0.01032595, "balance_loss_clip": 1.05718756, "balance_loss_mlp": 1.02475679, "epoch": 0.2729513617507365, "flos": 30993319232640.0, "grad_norm": 1.9366965237834086, "language_loss": 0.78182471, "learning_rate": 3.4122843924734834e-06, "loss": 0.80400026, "num_input_tokens_seen": 48730670, "step": 2270, "time_per_iteration": 2.6440577507019043 }, { "auxiliary_loss_clip": 0.01169066, "auxiliary_loss_mlp": 0.01026512, "balance_loss_clip": 1.0541805, "balance_loss_mlp": 1.01829803, "epoch": 0.2730716046413756, "flos": 19094640381600.0, "grad_norm": 1.8854849192520406, "language_loss": 0.8774879, "learning_rate": 3.411732718871319e-06, "loss": 0.89944363, "num_input_tokens_seen": 48746510, "step": 2271, "time_per_iteration": 2.595777750015259 }, { "auxiliary_loss_clip": 0.01201993, "auxiliary_loss_mlp": 0.01035173, "balance_loss_clip": 1.06093049, "balance_loss_mlp": 1.02687621, "epoch": 0.27319184753201464, "flos": 26944570680960.0, "grad_norm": 1.8582444255586772, "language_loss": 0.78596437, "learning_rate": 3.4111808311104227e-06, "loss": 0.80833602, "num_input_tokens_seen": 48768825, "step": 2272, "time_per_iteration": 2.5704243183135986 }, { "auxiliary_loss_clip": 0.01179293, "auxiliary_loss_mlp": 0.01035102, "balance_loss_clip": 1.05407786, "balance_loss_mlp": 1.026209, "epoch": 0.27331209042265375, "flos": 31759832589120.0, "grad_norm": 1.7371960524616334, "language_loss": 0.69689196, "learning_rate": 3.410628729274517e-06, "loss": 0.71903586, "num_input_tokens_seen": 48790345, "step": 2273, "time_per_iteration": 2.6923253536224365 }, { "auxiliary_loss_clip": 0.01168961, "auxiliary_loss_mlp": 0.00763857, "balance_loss_clip": 1.05384421, "balance_loss_mlp": 1.00011027, "epoch": 0.27343233331329286, "flos": 25739085065280.0, "grad_norm": 2.008766958373696, "language_loss": 0.82621884, "learning_rate": 3.4100764134473546e-06, "loss": 0.84554702, "num_input_tokens_seen": 48809630, "step": 2274, "time_per_iteration": 2.611969232559204 }, { "auxiliary_loss_clip": 0.01202482, "auxiliary_loss_mlp": 0.01030664, "balance_loss_clip": 1.05951869, "balance_loss_mlp": 1.02142489, "epoch": 0.2735525762039319, "flos": 24389418408480.0, "grad_norm": 2.207274634311326, "language_loss": 0.85148841, "learning_rate": 3.4095238837127215e-06, "loss": 0.87381995, "num_input_tokens_seen": 48828770, "step": 2275, "time_per_iteration": 2.5680060386657715 }, { "auxiliary_loss_clip": 0.01158762, "auxiliary_loss_mlp": 0.01032786, "balance_loss_clip": 1.05533588, "balance_loss_mlp": 1.02434027, "epoch": 0.27367281909457103, "flos": 14465360467680.0, "grad_norm": 2.016595994704089, "language_loss": 0.79079294, "learning_rate": 3.4089711401544355e-06, "loss": 0.81270838, "num_input_tokens_seen": 48846365, "step": 2276, "time_per_iteration": 2.5926055908203125 }, { "auxiliary_loss_clip": 0.01186468, "auxiliary_loss_mlp": 0.01028569, "balance_loss_clip": 1.0555985, "balance_loss_mlp": 1.01990223, "epoch": 0.27379306198521014, "flos": 23476999996800.0, "grad_norm": 2.370714442777603, "language_loss": 0.67393291, "learning_rate": 3.4084181828563486e-06, "loss": 0.69608331, "num_input_tokens_seen": 48863085, "step": 2277, "time_per_iteration": 3.3407840728759766 }, { "auxiliary_loss_clip": 0.011364, "auxiliary_loss_mlp": 0.01026398, "balance_loss_clip": 1.04728675, "balance_loss_mlp": 1.0181607, "epoch": 0.2739133048758492, "flos": 17458156071840.0, "grad_norm": 1.817640001008341, "language_loss": 0.70595813, "learning_rate": 3.4078650119023428e-06, "loss": 0.72758615, "num_input_tokens_seen": 48881400, "step": 2278, "time_per_iteration": 3.3869571685791016 }, { "auxiliary_loss_clip": 0.0113271, "auxiliary_loss_mlp": 0.01033647, "balance_loss_clip": 1.04856145, "balance_loss_mlp": 1.02515888, "epoch": 0.2740335477664883, "flos": 19273115731200.0, "grad_norm": 3.5048838287517965, "language_loss": 0.7415418, "learning_rate": 3.4073116273763337e-06, "loss": 0.76320535, "num_input_tokens_seen": 48895845, "step": 2279, "time_per_iteration": 2.6735012531280518 }, { "auxiliary_loss_clip": 0.01179445, "auxiliary_loss_mlp": 0.01038022, "balance_loss_clip": 1.05535138, "balance_loss_mlp": 1.02887821, "epoch": 0.2741537906571274, "flos": 26104727668800.0, "grad_norm": 1.767220940755944, "language_loss": 0.81072807, "learning_rate": 3.40675802936227e-06, "loss": 0.83290279, "num_input_tokens_seen": 48916630, "step": 2280, "time_per_iteration": 3.5586695671081543 }, { "auxiliary_loss_clip": 0.01165584, "auxiliary_loss_mlp": 0.01031162, "balance_loss_clip": 1.05370152, "balance_loss_mlp": 1.02284169, "epoch": 0.27427403354776647, "flos": 34164195099840.0, "grad_norm": 5.267855725061723, "language_loss": 0.71909732, "learning_rate": 3.4062042179441318e-06, "loss": 0.74106479, "num_input_tokens_seen": 48937100, "step": 2281, "time_per_iteration": 2.6989269256591797 }, { "auxiliary_loss_clip": 0.01185703, "auxiliary_loss_mlp": 0.0103105, "balance_loss_clip": 1.05850315, "balance_loss_mlp": 1.02306914, "epoch": 0.2743942764384056, "flos": 18766991533440.0, "grad_norm": 1.96158280378127, "language_loss": 0.80189049, "learning_rate": 3.4056501932059314e-06, "loss": 0.824058, "num_input_tokens_seen": 48955175, "step": 2282, "time_per_iteration": 2.5541343688964844 }, { "auxiliary_loss_clip": 0.01099739, "auxiliary_loss_mlp": 0.0100047, "balance_loss_clip": 1.0208149, "balance_loss_mlp": 0.99893236, "epoch": 0.2745145193290447, "flos": 64904057147040.0, "grad_norm": 0.7736570952521735, "language_loss": 0.58088326, "learning_rate": 3.405095955231715e-06, "loss": 0.60188544, "num_input_tokens_seen": 49006830, "step": 2283, "time_per_iteration": 3.050610303878784 }, { "auxiliary_loss_clip": 0.0119241, "auxiliary_loss_mlp": 0.0102816, "balance_loss_clip": 1.0570755, "balance_loss_mlp": 1.01971436, "epoch": 0.27463476221968375, "flos": 16136929259040.0, "grad_norm": 5.008520022343631, "language_loss": 0.94395936, "learning_rate": 3.4045415041055585e-06, "loss": 0.96616507, "num_input_tokens_seen": 49022470, "step": 2284, "time_per_iteration": 2.5435872077941895 }, { "auxiliary_loss_clip": 0.0117761, "auxiliary_loss_mlp": 0.01027718, "balance_loss_clip": 1.05589557, "balance_loss_mlp": 1.01912284, "epoch": 0.27475500511032286, "flos": 10376714561760.0, "grad_norm": 2.1192429637190844, "language_loss": 0.78432822, "learning_rate": 3.4039868399115728e-06, "loss": 0.80638146, "num_input_tokens_seen": 49037110, "step": 2285, "time_per_iteration": 2.543254852294922 }, { "auxiliary_loss_clip": 0.01135833, "auxiliary_loss_mlp": 0.01038142, "balance_loss_clip": 1.05416834, "balance_loss_mlp": 1.02966034, "epoch": 0.27487524800096197, "flos": 17311065756960.0, "grad_norm": 2.5269851832472603, "language_loss": 0.80469507, "learning_rate": 3.4034319627339003e-06, "loss": 0.82643485, "num_input_tokens_seen": 49053975, "step": 2286, "time_per_iteration": 3.531960964202881 }, { "auxiliary_loss_clip": 0.01176237, "auxiliary_loss_mlp": 0.01028472, "balance_loss_clip": 1.05591214, "balance_loss_mlp": 1.02008009, "epoch": 0.274995490891601, "flos": 27120208590720.0, "grad_norm": 2.398308010105636, "language_loss": 0.69342381, "learning_rate": 3.402876872656715e-06, "loss": 0.71547091, "num_input_tokens_seen": 49072295, "step": 2287, "time_per_iteration": 2.6278975009918213 }, { "auxiliary_loss_clip": 0.01170831, "auxiliary_loss_mlp": 0.01032615, "balance_loss_clip": 1.05557597, "balance_loss_mlp": 1.02415729, "epoch": 0.27511573378224013, "flos": 23436096967680.0, "grad_norm": 4.321973395528485, "language_loss": 0.89381135, "learning_rate": 3.402321569764223e-06, "loss": 0.91584575, "num_input_tokens_seen": 49091600, "step": 2288, "time_per_iteration": 2.596726179122925 }, { "auxiliary_loss_clip": 0.01145923, "auxiliary_loss_mlp": 0.00763839, "balance_loss_clip": 1.05259228, "balance_loss_mlp": 1.00013065, "epoch": 0.2752359766728792, "flos": 16722022075200.0, "grad_norm": 1.880239391162674, "language_loss": 0.83645242, "learning_rate": 3.4017660541406635e-06, "loss": 0.85555005, "num_input_tokens_seen": 49107665, "step": 2289, "time_per_iteration": 2.644167900085449 }, { "auxiliary_loss_clip": 0.01182067, "auxiliary_loss_mlp": 0.01028519, "balance_loss_clip": 1.05522048, "balance_loss_mlp": 1.02019227, "epoch": 0.2753562195635183, "flos": 25297742286720.0, "grad_norm": 1.8613840569892637, "language_loss": 0.73945481, "learning_rate": 3.4012103258703092e-06, "loss": 0.76156068, "num_input_tokens_seen": 49126420, "step": 2290, "time_per_iteration": 2.631107807159424 }, { "auxiliary_loss_clip": 0.01160654, "auxiliary_loss_mlp": 0.010295, "balance_loss_clip": 1.05430079, "balance_loss_mlp": 1.02141166, "epoch": 0.2754764624541574, "flos": 27338976381120.0, "grad_norm": 2.296462724005392, "language_loss": 0.82666945, "learning_rate": 3.4006543850374616e-06, "loss": 0.84857094, "num_input_tokens_seen": 49141470, "step": 2291, "time_per_iteration": 2.6618194580078125 }, { "auxiliary_loss_clip": 0.01194499, "auxiliary_loss_mlp": 0.01029971, "balance_loss_clip": 1.05972648, "balance_loss_mlp": 1.02166843, "epoch": 0.27559670534479647, "flos": 17238382606560.0, "grad_norm": 2.214873661121701, "language_loss": 0.7460109, "learning_rate": 3.400098231726458e-06, "loss": 0.76825559, "num_input_tokens_seen": 49158570, "step": 2292, "time_per_iteration": 2.5552632808685303 }, { "auxiliary_loss_clip": 0.0116116, "auxiliary_loss_mlp": 0.0102716, "balance_loss_clip": 1.05088985, "balance_loss_mlp": 1.01844013, "epoch": 0.2757169482354356, "flos": 21939088577280.0, "grad_norm": 2.0563238388468075, "language_loss": 0.86740595, "learning_rate": 3.3995418660216657e-06, "loss": 0.88928914, "num_input_tokens_seen": 49176025, "step": 2293, "time_per_iteration": 2.6397006511688232 }, { "auxiliary_loss_clip": 0.01207503, "auxiliary_loss_mlp": 0.0103658, "balance_loss_clip": 1.05986834, "balance_loss_mlp": 1.02753258, "epoch": 0.2758371911260747, "flos": 20850673086240.0, "grad_norm": 2.2404488426173823, "language_loss": 0.80319279, "learning_rate": 3.3989852880074848e-06, "loss": 0.82563365, "num_input_tokens_seen": 49197455, "step": 2294, "time_per_iteration": 2.553840160369873 }, { "auxiliary_loss_clip": 0.01067521, "auxiliary_loss_mlp": 0.01006018, "balance_loss_clip": 1.01589561, "balance_loss_mlp": 1.00427723, "epoch": 0.27595743401671374, "flos": 69269075258880.0, "grad_norm": 0.7426010105098813, "language_loss": 0.6068328, "learning_rate": 3.398428497768348e-06, "loss": 0.62756819, "num_input_tokens_seen": 49262625, "step": 2295, "time_per_iteration": 3.2702269554138184 }, { "auxiliary_loss_clip": 0.01167998, "auxiliary_loss_mlp": 0.01029124, "balance_loss_clip": 1.05321324, "balance_loss_mlp": 1.02050507, "epoch": 0.27607767690735285, "flos": 21215022679200.0, "grad_norm": 1.7232269472628603, "language_loss": 0.71862078, "learning_rate": 3.3978714953887205e-06, "loss": 0.74059194, "num_input_tokens_seen": 49282380, "step": 2296, "time_per_iteration": 2.6420536041259766 }, { "auxiliary_loss_clip": 0.01130975, "auxiliary_loss_mlp": 0.01034638, "balance_loss_clip": 1.04956794, "balance_loss_mlp": 1.02651358, "epoch": 0.27619791979799196, "flos": 24825337726080.0, "grad_norm": 1.9559500140614676, "language_loss": 0.86473715, "learning_rate": 3.397314280953098e-06, "loss": 0.88639331, "num_input_tokens_seen": 49303205, "step": 2297, "time_per_iteration": 2.725390911102295 }, { "auxiliary_loss_clip": 0.01168374, "auxiliary_loss_mlp": 0.01029748, "balance_loss_clip": 1.05500042, "balance_loss_mlp": 1.02152228, "epoch": 0.276318162688631, "flos": 24753552499680.0, "grad_norm": 1.7856013748665762, "language_loss": 0.800924, "learning_rate": 3.3967568545460108e-06, "loss": 0.82290518, "num_input_tokens_seen": 49322745, "step": 2298, "time_per_iteration": 2.6582460403442383 }, { "auxiliary_loss_clip": 0.011839, "auxiliary_loss_mlp": 0.01038728, "balance_loss_clip": 1.05505061, "balance_loss_mlp": 1.03005528, "epoch": 0.27643840557927013, "flos": 18150010845120.0, "grad_norm": 1.7646812063037995, "language_loss": 0.80226982, "learning_rate": 3.3961992162520185e-06, "loss": 0.82449615, "num_input_tokens_seen": 49341370, "step": 2299, "time_per_iteration": 2.5415494441986084 }, { "auxiliary_loss_clip": 0.01188328, "auxiliary_loss_mlp": 0.01031832, "balance_loss_clip": 1.05648279, "balance_loss_mlp": 1.02286696, "epoch": 0.27655864846990924, "flos": 24823936964640.0, "grad_norm": 2.0376232056716903, "language_loss": 0.71562731, "learning_rate": 3.3956413661557156e-06, "loss": 0.73782897, "num_input_tokens_seen": 49361545, "step": 2300, "time_per_iteration": 2.602334499359131 }, { "auxiliary_loss_clip": 0.01168924, "auxiliary_loss_mlp": 0.01033985, "balance_loss_clip": 1.05424225, "balance_loss_mlp": 1.02558684, "epoch": 0.2766788913605483, "flos": 20266585944960.0, "grad_norm": 3.3720346774067727, "language_loss": 0.66280544, "learning_rate": 3.3950833043417273e-06, "loss": 0.68483448, "num_input_tokens_seen": 49379690, "step": 2301, "time_per_iteration": 2.6161515712738037 }, { "auxiliary_loss_clip": 0.01195176, "auxiliary_loss_mlp": 0.01029652, "balance_loss_clip": 1.06279063, "balance_loss_mlp": 1.02072275, "epoch": 0.2767991342511874, "flos": 21470275712640.0, "grad_norm": 2.809588610150859, "language_loss": 0.72959805, "learning_rate": 3.3945250308947105e-06, "loss": 0.75184637, "num_input_tokens_seen": 49395995, "step": 2302, "time_per_iteration": 2.566408395767212 }, { "auxiliary_loss_clip": 0.01088829, "auxiliary_loss_mlp": 0.01004897, "balance_loss_clip": 1.01938593, "balance_loss_mlp": 1.00339508, "epoch": 0.2769193771418265, "flos": 66002637137760.0, "grad_norm": 1.224247082906793, "language_loss": 0.68321168, "learning_rate": 3.3939665458993556e-06, "loss": 0.70414895, "num_input_tokens_seen": 49450415, "step": 2303, "time_per_iteration": 3.9219067096710205 }, { "auxiliary_loss_clip": 0.01166882, "auxiliary_loss_mlp": 0.01034949, "balance_loss_clip": 1.05575824, "balance_loss_mlp": 1.02644968, "epoch": 0.27703962003246557, "flos": 20704444778400.0, "grad_norm": 2.2648163588690458, "language_loss": 0.76746178, "learning_rate": 3.3934078494403843e-06, "loss": 0.78948009, "num_input_tokens_seen": 49469990, "step": 2304, "time_per_iteration": 2.635122299194336 }, { "auxiliary_loss_clip": 0.01109944, "auxiliary_loss_mlp": 0.00763911, "balance_loss_clip": 1.04830551, "balance_loss_mlp": 1.00012147, "epoch": 0.2771598629231047, "flos": 22929900936000.0, "grad_norm": 1.9956292613512823, "language_loss": 0.81158221, "learning_rate": 3.3928489416025495e-06, "loss": 0.83032072, "num_input_tokens_seen": 49490835, "step": 2305, "time_per_iteration": 2.73683500289917 }, { "auxiliary_loss_clip": 0.0117378, "auxiliary_loss_mlp": 0.01033493, "balance_loss_clip": 1.05633676, "balance_loss_mlp": 1.02441478, "epoch": 0.27728010581374374, "flos": 18369461057760.0, "grad_norm": 2.5883740543053357, "language_loss": 0.78892261, "learning_rate": 3.392289822470638e-06, "loss": 0.81099534, "num_input_tokens_seen": 49508815, "step": 2306, "time_per_iteration": 3.501736640930176 }, { "auxiliary_loss_clip": 0.01172757, "auxiliary_loss_mlp": 0.01032238, "balance_loss_clip": 1.05533886, "balance_loss_mlp": 1.02389884, "epoch": 0.27740034870438285, "flos": 19427640856800.0, "grad_norm": 2.3718343791983525, "language_loss": 0.76102388, "learning_rate": 3.3917304921294674e-06, "loss": 0.78307378, "num_input_tokens_seen": 49526980, "step": 2307, "time_per_iteration": 2.5912556648254395 }, { "auxiliary_loss_clip": 0.01188111, "auxiliary_loss_mlp": 0.0102547, "balance_loss_clip": 1.05576086, "balance_loss_mlp": 1.01723266, "epoch": 0.27752059159502196, "flos": 21614780006400.0, "grad_norm": 2.017812802973103, "language_loss": 0.80650985, "learning_rate": 3.3911709506638876e-06, "loss": 0.82864571, "num_input_tokens_seen": 49546290, "step": 2308, "time_per_iteration": 2.57138729095459 }, { "auxiliary_loss_clip": 0.01150765, "auxiliary_loss_mlp": 0.00763761, "balance_loss_clip": 1.05044985, "balance_loss_mlp": 1.00005817, "epoch": 0.277640834485661, "flos": 26608014426720.0, "grad_norm": 2.121574644517383, "language_loss": 0.81613904, "learning_rate": 3.390611198158781e-06, "loss": 0.83528429, "num_input_tokens_seen": 49564165, "step": 2309, "time_per_iteration": 2.6738150119781494 }, { "auxiliary_loss_clip": 0.01209344, "auxiliary_loss_mlp": 0.0103053, "balance_loss_clip": 1.06231093, "balance_loss_mlp": 1.02151823, "epoch": 0.2777610773763001, "flos": 19492817362560.0, "grad_norm": 2.4238281252424834, "language_loss": 0.89584041, "learning_rate": 3.3900512346990612e-06, "loss": 0.91823912, "num_input_tokens_seen": 49580155, "step": 2310, "time_per_iteration": 2.5200068950653076 }, { "auxiliary_loss_clip": 0.01146211, "auxiliary_loss_mlp": 0.01030792, "balance_loss_clip": 1.05085182, "balance_loss_mlp": 1.02143979, "epoch": 0.27788132026693924, "flos": 38290655176320.0, "grad_norm": 1.9416515753441228, "language_loss": 0.65583163, "learning_rate": 3.389491060369674e-06, "loss": 0.6776017, "num_input_tokens_seen": 49605830, "step": 2311, "time_per_iteration": 2.8053762912750244 }, { "auxiliary_loss_clip": 0.01140735, "auxiliary_loss_mlp": 0.01030014, "balance_loss_clip": 1.05462003, "balance_loss_mlp": 1.02182424, "epoch": 0.2780015631575783, "flos": 22382550456480.0, "grad_norm": 1.9165940357452285, "language_loss": 0.89472425, "learning_rate": 3.388930675255598e-06, "loss": 0.91643173, "num_input_tokens_seen": 49625680, "step": 2312, "time_per_iteration": 3.563308000564575 }, { "auxiliary_loss_clip": 0.01177941, "auxiliary_loss_mlp": 0.01035063, "balance_loss_clip": 1.05486107, "balance_loss_mlp": 1.02631295, "epoch": 0.2781218060482174, "flos": 12203203565280.0, "grad_norm": 2.793969060843609, "language_loss": 0.79797822, "learning_rate": 3.388370079441843e-06, "loss": 0.82010823, "num_input_tokens_seen": 49641195, "step": 2313, "time_per_iteration": 2.5540637969970703 }, { "auxiliary_loss_clip": 0.01166444, "auxiliary_loss_mlp": 0.01030986, "balance_loss_clip": 1.05974698, "balance_loss_mlp": 1.02178311, "epoch": 0.2782420489388565, "flos": 18107635220640.0, "grad_norm": 1.9172151807035915, "language_loss": 0.92844832, "learning_rate": 3.3878092730134505e-06, "loss": 0.95042264, "num_input_tokens_seen": 49659180, "step": 2314, "time_per_iteration": 2.5955896377563477 }, { "auxiliary_loss_clip": 0.01182715, "auxiliary_loss_mlp": 0.01034868, "balance_loss_clip": 1.05605125, "balance_loss_mlp": 1.02531886, "epoch": 0.27836229182949557, "flos": 18514755524640.0, "grad_norm": 1.5911213951770962, "language_loss": 0.80456275, "learning_rate": 3.3872482560554947e-06, "loss": 0.8267386, "num_input_tokens_seen": 49677955, "step": 2315, "time_per_iteration": 2.5578012466430664 }, { "auxiliary_loss_clip": 0.01087806, "auxiliary_loss_mlp": 0.00999417, "balance_loss_clip": 1.01927686, "balance_loss_mlp": 0.99792117, "epoch": 0.2784825347201347, "flos": 67079242337280.0, "grad_norm": 0.7949182194926061, "language_loss": 0.56991565, "learning_rate": 3.386687028653082e-06, "loss": 0.59078789, "num_input_tokens_seen": 49740800, "step": 2316, "time_per_iteration": 3.180739402770996 }, { "auxiliary_loss_clip": 0.01147198, "auxiliary_loss_mlp": 0.01029432, "balance_loss_clip": 1.05377352, "balance_loss_mlp": 1.02100372, "epoch": 0.2786027776107738, "flos": 22631122935360.0, "grad_norm": 1.9483131569800363, "language_loss": 0.85091329, "learning_rate": 3.386125590891349e-06, "loss": 0.87267959, "num_input_tokens_seen": 49757675, "step": 2317, "time_per_iteration": 2.644926071166992 }, { "auxiliary_loss_clip": 0.01163097, "auxiliary_loss_mlp": 0.0102814, "balance_loss_clip": 1.05348933, "balance_loss_mlp": 1.0198307, "epoch": 0.27872302050141284, "flos": 15778829217120.0, "grad_norm": 1.9855536428281249, "language_loss": 0.82883584, "learning_rate": 3.3855639428554657e-06, "loss": 0.85074824, "num_input_tokens_seen": 49775205, "step": 2318, "time_per_iteration": 2.5496323108673096 }, { "auxiliary_loss_clip": 0.01144317, "auxiliary_loss_mlp": 0.01027512, "balance_loss_clip": 1.04838228, "balance_loss_mlp": 1.01931596, "epoch": 0.27884326339205195, "flos": 22126974170400.0, "grad_norm": 1.8707073950573931, "language_loss": 0.80146313, "learning_rate": 3.385002084630635e-06, "loss": 0.82318139, "num_input_tokens_seen": 49794175, "step": 2319, "time_per_iteration": 2.5817129611968994 }, { "auxiliary_loss_clip": 0.01196099, "auxiliary_loss_mlp": 0.01033949, "balance_loss_clip": 1.06018591, "balance_loss_mlp": 1.02568769, "epoch": 0.278963506282691, "flos": 20558719308000.0, "grad_norm": 2.2673294808258873, "language_loss": 0.84790981, "learning_rate": 3.384440016302088e-06, "loss": 0.87021023, "num_input_tokens_seen": 49812850, "step": 2320, "time_per_iteration": 2.545694351196289 }, { "auxiliary_loss_clip": 0.01185144, "auxiliary_loss_mlp": 0.01029375, "balance_loss_clip": 1.05647767, "balance_loss_mlp": 1.0207262, "epoch": 0.2790837491733301, "flos": 21942931692000.0, "grad_norm": 1.8901854149244042, "language_loss": 0.61975467, "learning_rate": 3.3838777379550923e-06, "loss": 0.64189982, "num_input_tokens_seen": 49832295, "step": 2321, "time_per_iteration": 2.5477516651153564 }, { "auxiliary_loss_clip": 0.01174655, "auxiliary_loss_mlp": 0.01035574, "balance_loss_clip": 1.05592012, "balance_loss_mlp": 1.02728319, "epoch": 0.27920399206396923, "flos": 26286794714400.0, "grad_norm": 2.7468169558170956, "language_loss": 0.78915161, "learning_rate": 3.383315249674944e-06, "loss": 0.81125391, "num_input_tokens_seen": 49850860, "step": 2322, "time_per_iteration": 2.6317172050476074 }, { "auxiliary_loss_clip": 0.01163875, "auxiliary_loss_mlp": 0.01036206, "balance_loss_clip": 1.05582666, "balance_loss_mlp": 1.02834427, "epoch": 0.2793242349546083, "flos": 25400984381760.0, "grad_norm": 2.006139273143541, "language_loss": 0.85993248, "learning_rate": 3.3827525515469715e-06, "loss": 0.88193333, "num_input_tokens_seen": 49865765, "step": 2323, "time_per_iteration": 2.642104148864746 }, { "auxiliary_loss_clip": 0.01152196, "auxiliary_loss_mlp": 0.01034565, "balance_loss_clip": 1.05101609, "balance_loss_mlp": 1.0254631, "epoch": 0.2794444778452474, "flos": 20850349833600.0, "grad_norm": 2.460962243016697, "language_loss": 0.70481342, "learning_rate": 3.3821896436565367e-06, "loss": 0.72668099, "num_input_tokens_seen": 49885425, "step": 2324, "time_per_iteration": 2.655522108078003 }, { "auxiliary_loss_clip": 0.01190243, "auxiliary_loss_mlp": 0.0103434, "balance_loss_clip": 1.05881035, "balance_loss_mlp": 1.0260849, "epoch": 0.2795647207358865, "flos": 21576247496640.0, "grad_norm": 1.6841073588949012, "language_loss": 0.70061421, "learning_rate": 3.381626526089032e-06, "loss": 0.72285998, "num_input_tokens_seen": 49904990, "step": 2325, "time_per_iteration": 2.567300796508789 }, { "auxiliary_loss_clip": 0.01169851, "auxiliary_loss_mlp": 0.01032399, "balance_loss_clip": 1.05250502, "balance_loss_mlp": 1.02418542, "epoch": 0.27968496362652556, "flos": 21471748308000.0, "grad_norm": 2.5649306327279486, "language_loss": 0.79259348, "learning_rate": 3.3810631989298815e-06, "loss": 0.81461596, "num_input_tokens_seen": 49924600, "step": 2326, "time_per_iteration": 2.615076780319214 }, { "auxiliary_loss_clip": 0.01153613, "auxiliary_loss_mlp": 0.01039412, "balance_loss_clip": 1.055264, "balance_loss_mlp": 1.0301609, "epoch": 0.2798052065171647, "flos": 23258699126880.0, "grad_norm": 2.053324108868607, "language_loss": 0.84293306, "learning_rate": 3.3804996622645423e-06, "loss": 0.86486328, "num_input_tokens_seen": 49942600, "step": 2327, "time_per_iteration": 2.671381711959839 }, { "auxiliary_loss_clip": 0.01205089, "auxiliary_loss_mlp": 0.01034722, "balance_loss_clip": 1.06105876, "balance_loss_mlp": 1.02675891, "epoch": 0.2799254494078038, "flos": 21539331250080.0, "grad_norm": 2.026696875939431, "language_loss": 0.89384836, "learning_rate": 3.3799359161785015e-06, "loss": 0.91624641, "num_input_tokens_seen": 49962250, "step": 2328, "time_per_iteration": 2.5989603996276855 }, { "auxiliary_loss_clip": 0.01190418, "auxiliary_loss_mlp": 0.01028923, "balance_loss_clip": 1.05947638, "balance_loss_mlp": 1.01999426, "epoch": 0.28004569229844284, "flos": 26393915841120.0, "grad_norm": 1.5047727872644099, "language_loss": 0.85650408, "learning_rate": 3.3793719607572798e-06, "loss": 0.87869751, "num_input_tokens_seen": 49983215, "step": 2329, "time_per_iteration": 3.477705478668213 }, { "auxiliary_loss_clip": 0.01157876, "auxiliary_loss_mlp": 0.01028389, "balance_loss_clip": 1.05150795, "balance_loss_mlp": 1.02022314, "epoch": 0.28016593518908195, "flos": 33547681332000.0, "grad_norm": 1.8681278607884233, "language_loss": 0.77280468, "learning_rate": 3.378807796086428e-06, "loss": 0.79466736, "num_input_tokens_seen": 50006075, "step": 2330, "time_per_iteration": 2.7017838954925537 }, { "auxiliary_loss_clip": 0.01205205, "auxiliary_loss_mlp": 0.01033414, "balance_loss_clip": 1.06004882, "balance_loss_mlp": 1.02467632, "epoch": 0.28028617807972106, "flos": 15340826715840.0, "grad_norm": 2.2668035329215077, "language_loss": 0.76948649, "learning_rate": 3.37824342225153e-06, "loss": 0.79187274, "num_input_tokens_seen": 50022495, "step": 2331, "time_per_iteration": 2.516514301300049 }, { "auxiliary_loss_clip": 0.01151103, "auxiliary_loss_mlp": 0.01033736, "balance_loss_clip": 1.05723476, "balance_loss_mlp": 1.02518904, "epoch": 0.2804064209703601, "flos": 25520281357920.0, "grad_norm": 1.8088005318511013, "language_loss": 0.77564836, "learning_rate": 3.3776788393382006e-06, "loss": 0.79749674, "num_input_tokens_seen": 50041975, "step": 2332, "time_per_iteration": 3.5752522945404053 }, { "auxiliary_loss_clip": 0.01208191, "auxiliary_loss_mlp": 0.01030431, "balance_loss_clip": 1.062621, "balance_loss_mlp": 1.02128768, "epoch": 0.2805266638609992, "flos": 29351770631520.0, "grad_norm": 2.119625649865964, "language_loss": 0.76784861, "learning_rate": 3.3771140474320872e-06, "loss": 0.79023492, "num_input_tokens_seen": 50061925, "step": 2333, "time_per_iteration": 2.6046483516693115 }, { "auxiliary_loss_clip": 0.01170587, "auxiliary_loss_mlp": 0.01030457, "balance_loss_clip": 1.05725217, "balance_loss_mlp": 1.02190936, "epoch": 0.28064690675163834, "flos": 21463738825920.0, "grad_norm": 1.928219402924996, "language_loss": 0.79045808, "learning_rate": 3.3765490466188664e-06, "loss": 0.81246853, "num_input_tokens_seen": 50079325, "step": 2334, "time_per_iteration": 2.6375887393951416 }, { "auxiliary_loss_clip": 0.01155959, "auxiliary_loss_mlp": 0.01032891, "balance_loss_clip": 1.05036497, "balance_loss_mlp": 1.02404618, "epoch": 0.2807671496422774, "flos": 20995644300480.0, "grad_norm": 4.231249233175594, "language_loss": 0.73744369, "learning_rate": 3.3759838369842508e-06, "loss": 0.75933218, "num_input_tokens_seen": 50097400, "step": 2335, "time_per_iteration": 2.6284046173095703 }, { "auxiliary_loss_clip": 0.0115939, "auxiliary_loss_mlp": 0.01031595, "balance_loss_clip": 1.05444968, "balance_loss_mlp": 1.02292871, "epoch": 0.2808873925329165, "flos": 21506581370880.0, "grad_norm": 1.9407220228159339, "language_loss": 0.73039967, "learning_rate": 3.375418418613981e-06, "loss": 0.75230956, "num_input_tokens_seen": 50116425, "step": 2336, "time_per_iteration": 2.642003059387207 }, { "auxiliary_loss_clip": 0.01179043, "auxiliary_loss_mlp": 0.01030318, "balance_loss_clip": 1.05822361, "balance_loss_mlp": 1.02193427, "epoch": 0.28100763542355556, "flos": 16070818912320.0, "grad_norm": 2.2105563884460078, "language_loss": 0.83479786, "learning_rate": 3.374852791593831e-06, "loss": 0.85689145, "num_input_tokens_seen": 50132625, "step": 2337, "time_per_iteration": 2.5897557735443115 }, { "auxiliary_loss_clip": 0.01156576, "auxiliary_loss_mlp": 0.01033385, "balance_loss_clip": 1.05345857, "balance_loss_mlp": 1.02483797, "epoch": 0.28112787831419467, "flos": 19062608841600.0, "grad_norm": 7.374618958930375, "language_loss": 0.54098213, "learning_rate": 3.374286956009605e-06, "loss": 0.56288171, "num_input_tokens_seen": 50151190, "step": 2338, "time_per_iteration": 3.5451669692993164 }, { "auxiliary_loss_clip": 0.01193174, "auxiliary_loss_mlp": 0.01035145, "balance_loss_clip": 1.06337881, "balance_loss_mlp": 1.02713466, "epoch": 0.2812481212048338, "flos": 12823632281760.0, "grad_norm": 3.0249917505414743, "language_loss": 0.74961317, "learning_rate": 3.3737209119471405e-06, "loss": 0.77189636, "num_input_tokens_seen": 50167700, "step": 2339, "time_per_iteration": 2.5793092250823975 }, { "auxiliary_loss_clip": 0.01199148, "auxiliary_loss_mlp": 0.01035423, "balance_loss_clip": 1.0607965, "balance_loss_mlp": 1.02711725, "epoch": 0.28136836409547283, "flos": 15633067829760.0, "grad_norm": 2.8072782062872514, "language_loss": 0.63548732, "learning_rate": 3.373154659492306e-06, "loss": 0.65783298, "num_input_tokens_seen": 50185840, "step": 2340, "time_per_iteration": 2.5623202323913574 }, { "auxiliary_loss_clip": 0.01179512, "auxiliary_loss_mlp": 0.01030867, "balance_loss_clip": 1.05832362, "balance_loss_mlp": 1.02271938, "epoch": 0.28148860698611194, "flos": 19933729137600.0, "grad_norm": 1.7936574640918306, "language_loss": 0.85450727, "learning_rate": 3.3725881987310016e-06, "loss": 0.87661105, "num_input_tokens_seen": 50203375, "step": 2341, "time_per_iteration": 2.5950911045074463 }, { "auxiliary_loss_clip": 0.0117514, "auxiliary_loss_mlp": 0.01029783, "balance_loss_clip": 1.05586481, "balance_loss_mlp": 1.02155733, "epoch": 0.28160884987675106, "flos": 17457222230880.0, "grad_norm": 9.074065483970855, "language_loss": 0.87722278, "learning_rate": 3.372021529749159e-06, "loss": 0.89927197, "num_input_tokens_seen": 50222435, "step": 2342, "time_per_iteration": 2.608304738998413 }, { "auxiliary_loss_clip": 0.01136095, "auxiliary_loss_mlp": 0.01027718, "balance_loss_clip": 1.05485749, "balance_loss_mlp": 1.01924872, "epoch": 0.2817290927673901, "flos": 16834746247680.0, "grad_norm": 2.1176588469914113, "language_loss": 0.92153102, "learning_rate": 3.3714546526327405e-06, "loss": 0.94316912, "num_input_tokens_seen": 50240435, "step": 2343, "time_per_iteration": 2.660243511199951 }, { "auxiliary_loss_clip": 0.01167679, "auxiliary_loss_mlp": 0.01032879, "balance_loss_clip": 1.0546639, "balance_loss_mlp": 1.02393806, "epoch": 0.2818493356580292, "flos": 15414084537600.0, "grad_norm": 2.927192413874939, "language_loss": 0.87996459, "learning_rate": 3.3708875674677423e-06, "loss": 0.90197015, "num_input_tokens_seen": 50258410, "step": 2344, "time_per_iteration": 2.653120994567871 }, { "auxiliary_loss_clip": 0.01186635, "auxiliary_loss_mlp": 0.01031354, "balance_loss_clip": 1.06050801, "balance_loss_mlp": 1.02290177, "epoch": 0.28196957854866833, "flos": 20412455083200.0, "grad_norm": 1.9731414826842502, "language_loss": 0.83474028, "learning_rate": 3.37032027434019e-06, "loss": 0.85692012, "num_input_tokens_seen": 50277930, "step": 2345, "time_per_iteration": 2.5994176864624023 }, { "auxiliary_loss_clip": 0.01201025, "auxiliary_loss_mlp": 0.01033503, "balance_loss_clip": 1.05950475, "balance_loss_mlp": 1.02340579, "epoch": 0.2820898214393074, "flos": 19973123654400.0, "grad_norm": 2.5741589499200423, "language_loss": 0.82923919, "learning_rate": 3.369752773336141e-06, "loss": 0.85158443, "num_input_tokens_seen": 50297410, "step": 2346, "time_per_iteration": 2.5747122764587402 }, { "auxiliary_loss_clip": 0.01177808, "auxiliary_loss_mlp": 0.0103236, "balance_loss_clip": 1.05700421, "balance_loss_mlp": 1.02327025, "epoch": 0.2822100643299465, "flos": 22528311843840.0, "grad_norm": 2.215627211811728, "language_loss": 0.78415954, "learning_rate": 3.3691850645416864e-06, "loss": 0.80626124, "num_input_tokens_seen": 50317120, "step": 2347, "time_per_iteration": 2.6476516723632812 }, { "auxiliary_loss_clip": 0.01194584, "auxiliary_loss_mlp": 0.01031932, "balance_loss_clip": 1.05874586, "balance_loss_mlp": 1.02346849, "epoch": 0.2823303072205856, "flos": 11546684692320.0, "grad_norm": 2.019478262200728, "language_loss": 0.83022404, "learning_rate": 3.368617148042945e-06, "loss": 0.85248923, "num_input_tokens_seen": 50334790, "step": 2348, "time_per_iteration": 2.5741937160491943 }, { "auxiliary_loss_clip": 0.01169826, "auxiliary_loss_mlp": 0.01023875, "balance_loss_clip": 1.05293941, "balance_loss_mlp": 1.01483893, "epoch": 0.28245055011122466, "flos": 18259897577760.0, "grad_norm": 2.041955488776076, "language_loss": 0.84331107, "learning_rate": 3.368049023926071e-06, "loss": 0.86524808, "num_input_tokens_seen": 50353785, "step": 2349, "time_per_iteration": 2.600491762161255 }, { "auxiliary_loss_clip": 0.01189651, "auxiliary_loss_mlp": 0.0103154, "balance_loss_clip": 1.06066751, "balance_loss_mlp": 1.02405381, "epoch": 0.2825707930018638, "flos": 24608114364960.0, "grad_norm": 1.591420734160791, "language_loss": 0.83480233, "learning_rate": 3.3674806922772476e-06, "loss": 0.85701424, "num_input_tokens_seen": 50374670, "step": 2350, "time_per_iteration": 2.5981085300445557 }, { "auxiliary_loss_clip": 0.01166759, "auxiliary_loss_mlp": 0.01028795, "balance_loss_clip": 1.05586183, "balance_loss_mlp": 1.02054584, "epoch": 0.28269103589250283, "flos": 25226998652160.0, "grad_norm": 1.7728192284438202, "language_loss": 0.75086725, "learning_rate": 3.3669121531826904e-06, "loss": 0.7728228, "num_input_tokens_seen": 50395650, "step": 2351, "time_per_iteration": 2.702505111694336 }, { "auxiliary_loss_clip": 0.01156584, "auxiliary_loss_mlp": 0.01028986, "balance_loss_clip": 1.05769873, "balance_loss_mlp": 1.0205518, "epoch": 0.28281127878314194, "flos": 19281556216800.0, "grad_norm": 2.392129268740386, "language_loss": 0.83134878, "learning_rate": 3.366343406728647e-06, "loss": 0.85320443, "num_input_tokens_seen": 50415100, "step": 2352, "time_per_iteration": 2.6232995986938477 }, { "auxiliary_loss_clip": 0.01180479, "auxiliary_loss_mlp": 0.01032182, "balance_loss_clip": 1.05436051, "balance_loss_mlp": 1.02390313, "epoch": 0.28293152167378105, "flos": 23878409504160.0, "grad_norm": 3.1625826825152683, "language_loss": 0.68576062, "learning_rate": 3.3657744530013946e-06, "loss": 0.70788717, "num_input_tokens_seen": 50434335, "step": 2353, "time_per_iteration": 2.6049630641937256 }, { "auxiliary_loss_clip": 0.01197775, "auxiliary_loss_mlp": 0.01031445, "balance_loss_clip": 1.06107903, "balance_loss_mlp": 1.02265334, "epoch": 0.2830517645644201, "flos": 43866971063040.0, "grad_norm": 1.973158405146505, "language_loss": 0.71060711, "learning_rate": 3.3652052920872437e-06, "loss": 0.73289931, "num_input_tokens_seen": 50457200, "step": 2354, "time_per_iteration": 2.779451608657837 }, { "auxiliary_loss_clip": 0.01181824, "auxiliary_loss_mlp": 0.01032715, "balance_loss_clip": 1.06017733, "balance_loss_mlp": 1.02370918, "epoch": 0.2831720074550592, "flos": 26651754895680.0, "grad_norm": 1.928799189402782, "language_loss": 0.85533297, "learning_rate": 3.3646359240725355e-06, "loss": 0.87747836, "num_input_tokens_seen": 50476390, "step": 2355, "time_per_iteration": 3.5136396884918213 }, { "auxiliary_loss_clip": 0.01184397, "auxiliary_loss_mlp": 0.00763627, "balance_loss_clip": 1.05737686, "balance_loss_mlp": 1.00033534, "epoch": 0.2832922503456983, "flos": 31029984060480.0, "grad_norm": 2.133171496525383, "language_loss": 0.67347467, "learning_rate": 3.364066349043643e-06, "loss": 0.69295496, "num_input_tokens_seen": 50497595, "step": 2356, "time_per_iteration": 2.7033910751342773 }, { "auxiliary_loss_clip": 0.01171519, "auxiliary_loss_mlp": 0.0102355, "balance_loss_clip": 1.05496156, "balance_loss_mlp": 1.01604533, "epoch": 0.2834124932363374, "flos": 20405702694720.0, "grad_norm": 1.9760113368285273, "language_loss": 0.81909716, "learning_rate": 3.363496567086969e-06, "loss": 0.84104788, "num_input_tokens_seen": 50514690, "step": 2357, "time_per_iteration": 2.6059703826904297 }, { "auxiliary_loss_clip": 0.01205375, "auxiliary_loss_mlp": 0.01029042, "balance_loss_clip": 1.06001329, "balance_loss_mlp": 1.02134693, "epoch": 0.2835327361269765, "flos": 39384853297920.0, "grad_norm": 2.423444719326948, "language_loss": 0.75800455, "learning_rate": 3.3629265782889506e-06, "loss": 0.78034872, "num_input_tokens_seen": 50536515, "step": 2358, "time_per_iteration": 3.525283098220825 }, { "auxiliary_loss_clip": 0.01154245, "auxiliary_loss_mlp": 0.01034617, "balance_loss_clip": 1.05256987, "balance_loss_mlp": 1.02606952, "epoch": 0.2836529790176156, "flos": 30261603022080.0, "grad_norm": 1.868257975212941, "language_loss": 0.71812677, "learning_rate": 3.362356382736054e-06, "loss": 0.74001539, "num_input_tokens_seen": 50557120, "step": 2359, "time_per_iteration": 2.702363967895508 }, { "auxiliary_loss_clip": 0.01160216, "auxiliary_loss_mlp": 0.01028445, "balance_loss_clip": 1.05200064, "balance_loss_mlp": 1.02038026, "epoch": 0.28377322190825466, "flos": 12677799060480.0, "grad_norm": 2.2532224324011807, "language_loss": 0.90812111, "learning_rate": 3.361785980514777e-06, "loss": 0.9300077, "num_input_tokens_seen": 50573320, "step": 2360, "time_per_iteration": 2.6107168197631836 }, { "auxiliary_loss_clip": 0.01122099, "auxiliary_loss_mlp": 0.01035663, "balance_loss_clip": 1.04947269, "balance_loss_mlp": 1.02662075, "epoch": 0.28389346479889377, "flos": 18296670156480.0, "grad_norm": 2.2349318126151974, "language_loss": 0.76305109, "learning_rate": 3.361215371711649e-06, "loss": 0.78462875, "num_input_tokens_seen": 50592415, "step": 2361, "time_per_iteration": 2.6655068397521973 }, { "auxiliary_loss_clip": 0.0115362, "auxiliary_loss_mlp": 0.01024962, "balance_loss_clip": 1.05511069, "balance_loss_mlp": 1.01639676, "epoch": 0.2840137076895329, "flos": 20406995705280.0, "grad_norm": 1.743366341935156, "language_loss": 0.83005083, "learning_rate": 3.3606445564132326e-06, "loss": 0.85183662, "num_input_tokens_seen": 50609710, "step": 2362, "time_per_iteration": 2.6518595218658447 }, { "auxiliary_loss_clip": 0.01208709, "auxiliary_loss_mlp": 0.00762843, "balance_loss_clip": 1.06381869, "balance_loss_mlp": 1.00017536, "epoch": 0.28413395058017193, "flos": 20048033656320.0, "grad_norm": 3.630823306333969, "language_loss": 0.82233727, "learning_rate": 3.360073534706118e-06, "loss": 0.84205282, "num_input_tokens_seen": 50626865, "step": 2363, "time_per_iteration": 2.5477964878082275 }, { "auxiliary_loss_clip": 0.01180348, "auxiliary_loss_mlp": 0.01034617, "balance_loss_clip": 1.06129396, "balance_loss_mlp": 1.02612925, "epoch": 0.28425419347081105, "flos": 37663617739200.0, "grad_norm": 2.4130161548945677, "language_loss": 0.75736248, "learning_rate": 3.35950230667693e-06, "loss": 0.77951211, "num_input_tokens_seen": 50648560, "step": 2364, "time_per_iteration": 3.5724546909332275 }, { "auxiliary_loss_clip": 0.01193449, "auxiliary_loss_mlp": 0.01024533, "balance_loss_clip": 1.05989003, "balance_loss_mlp": 1.01673114, "epoch": 0.28437443636145016, "flos": 13845075419040.0, "grad_norm": 2.3018256785822273, "language_loss": 0.86145705, "learning_rate": 3.358930872412323e-06, "loss": 0.88363683, "num_input_tokens_seen": 50665725, "step": 2365, "time_per_iteration": 2.5695199966430664 }, { "auxiliary_loss_clip": 0.01189972, "auxiliary_loss_mlp": 0.01032567, "balance_loss_clip": 1.0608691, "balance_loss_mlp": 1.02486575, "epoch": 0.2844946792520892, "flos": 22747797973440.0, "grad_norm": 2.34269986896554, "language_loss": 0.81075037, "learning_rate": 3.3583592319989825e-06, "loss": 0.8329758, "num_input_tokens_seen": 50685095, "step": 2366, "time_per_iteration": 2.5706138610839844 }, { "auxiliary_loss_clip": 0.01202993, "auxiliary_loss_mlp": 0.01028973, "balance_loss_clip": 1.06348538, "balance_loss_mlp": 1.0203898, "epoch": 0.2846149221427283, "flos": 32415992292480.0, "grad_norm": 2.2544440014889253, "language_loss": 0.69442868, "learning_rate": 3.357787385523627e-06, "loss": 0.71674836, "num_input_tokens_seen": 50706500, "step": 2367, "time_per_iteration": 2.6929140090942383 }, { "auxiliary_loss_clip": 0.01141021, "auxiliary_loss_mlp": 0.01034382, "balance_loss_clip": 1.05287051, "balance_loss_mlp": 1.02624607, "epoch": 0.2847351650333674, "flos": 28475981130720.0, "grad_norm": 1.8105456124828656, "language_loss": 0.82682937, "learning_rate": 3.3572153330730048e-06, "loss": 0.8485834, "num_input_tokens_seen": 50727595, "step": 2368, "time_per_iteration": 2.736341714859009 }, { "auxiliary_loss_clip": 0.01073212, "auxiliary_loss_mlp": 0.01000664, "balance_loss_clip": 1.02595186, "balance_loss_mlp": 0.99907905, "epoch": 0.2848554079240065, "flos": 55753408619040.0, "grad_norm": 0.8409593743811221, "language_loss": 0.64767873, "learning_rate": 3.3566430747338956e-06, "loss": 0.66841751, "num_input_tokens_seen": 50782800, "step": 2369, "time_per_iteration": 3.0695292949676514 }, { "auxiliary_loss_clip": 0.01194658, "auxiliary_loss_mlp": 0.01026653, "balance_loss_clip": 1.05925632, "balance_loss_mlp": 1.01852608, "epoch": 0.2849756508146456, "flos": 11836878539520.0, "grad_norm": 2.297847138493238, "language_loss": 0.86260021, "learning_rate": 3.35607061059311e-06, "loss": 0.88481331, "num_input_tokens_seen": 50797730, "step": 2370, "time_per_iteration": 2.5394649505615234 }, { "auxiliary_loss_clip": 0.01202715, "auxiliary_loss_mlp": 0.01024198, "balance_loss_clip": 1.06146908, "balance_loss_mlp": 1.01621091, "epoch": 0.28509589370528465, "flos": 25155213425760.0, "grad_norm": 1.7947001791632362, "language_loss": 0.74894899, "learning_rate": 3.3554979407374917e-06, "loss": 0.77121812, "num_input_tokens_seen": 50819840, "step": 2371, "time_per_iteration": 2.5974180698394775 }, { "auxiliary_loss_clip": 0.01188905, "auxiliary_loss_mlp": 0.01023977, "balance_loss_clip": 1.05735576, "balance_loss_mlp": 1.01569796, "epoch": 0.28521613659592376, "flos": 19974811751520.0, "grad_norm": 1.6364417949289802, "language_loss": 0.736233, "learning_rate": 3.3549250652539134e-06, "loss": 0.75836188, "num_input_tokens_seen": 50838935, "step": 2372, "time_per_iteration": 2.5678157806396484 }, { "auxiliary_loss_clip": 0.01173495, "auxiliary_loss_mlp": 0.01025623, "balance_loss_clip": 1.05324483, "balance_loss_mlp": 1.01724839, "epoch": 0.2853363794865629, "flos": 23367975271200.0, "grad_norm": 1.7240089045987819, "language_loss": 0.81465769, "learning_rate": 3.3543519842292794e-06, "loss": 0.83664882, "num_input_tokens_seen": 50858590, "step": 2373, "time_per_iteration": 2.6362719535827637 }, { "auxiliary_loss_clip": 0.01206646, "auxiliary_loss_mlp": 0.00763356, "balance_loss_clip": 1.06311774, "balance_loss_mlp": 1.00022244, "epoch": 0.28545662237720193, "flos": 19861943911200.0, "grad_norm": 1.7143875938265989, "language_loss": 0.8362844, "learning_rate": 3.353778697750527e-06, "loss": 0.85598445, "num_input_tokens_seen": 50876995, "step": 2374, "time_per_iteration": 2.5445151329040527 }, { "auxiliary_loss_clip": 0.01167217, "auxiliary_loss_mlp": 0.01032269, "balance_loss_clip": 1.05342388, "balance_loss_mlp": 1.02463961, "epoch": 0.28557686526784104, "flos": 23879020092480.0, "grad_norm": 1.7957364189510416, "language_loss": 0.8924396, "learning_rate": 3.353205205904622e-06, "loss": 0.91443443, "num_input_tokens_seen": 50896105, "step": 2375, "time_per_iteration": 2.62347674369812 }, { "auxiliary_loss_clip": 0.01174068, "auxiliary_loss_mlp": 0.01034103, "balance_loss_clip": 1.05764568, "balance_loss_mlp": 1.02481103, "epoch": 0.28569710815848015, "flos": 44890389633120.0, "grad_norm": 2.0418033342999027, "language_loss": 0.71903646, "learning_rate": 3.3526315087785637e-06, "loss": 0.74111813, "num_input_tokens_seen": 50917220, "step": 2376, "time_per_iteration": 2.8065452575683594 }, { "auxiliary_loss_clip": 0.01126326, "auxiliary_loss_mlp": 0.01037215, "balance_loss_clip": 1.05039549, "balance_loss_mlp": 1.02879918, "epoch": 0.2858173510491192, "flos": 26829763324800.0, "grad_norm": 1.5880959470656764, "language_loss": 0.80950165, "learning_rate": 3.3520576064593805e-06, "loss": 0.83113712, "num_input_tokens_seen": 50937175, "step": 2377, "time_per_iteration": 2.6893391609191895 }, { "auxiliary_loss_clip": 0.01194272, "auxiliary_loss_mlp": 0.0103487, "balance_loss_clip": 1.05959356, "balance_loss_mlp": 1.02635241, "epoch": 0.2859375939397583, "flos": 23148920145120.0, "grad_norm": 1.524694687780088, "language_loss": 0.81844789, "learning_rate": 3.3514834990341337e-06, "loss": 0.84073925, "num_input_tokens_seen": 50957500, "step": 2378, "time_per_iteration": 2.580538511276245 }, { "auxiliary_loss_clip": 0.01182422, "auxiliary_loss_mlp": 0.01027916, "balance_loss_clip": 1.05806565, "balance_loss_mlp": 1.01998281, "epoch": 0.2860578368303974, "flos": 12129802075680.0, "grad_norm": 2.448423313476702, "language_loss": 0.93152016, "learning_rate": 3.3509091865899144e-06, "loss": 0.95362347, "num_input_tokens_seen": 50972690, "step": 2379, "time_per_iteration": 2.6029138565063477 }, { "auxiliary_loss_clip": 0.01205874, "auxiliary_loss_mlp": 0.01034008, "balance_loss_clip": 1.06050503, "balance_loss_mlp": 1.0260148, "epoch": 0.2861780797210365, "flos": 19938039172800.0, "grad_norm": 1.9613435977336753, "language_loss": 0.70821512, "learning_rate": 3.350334669213846e-06, "loss": 0.73061395, "num_input_tokens_seen": 50990095, "step": 2380, "time_per_iteration": 3.3479669094085693 }, { "auxiliary_loss_clip": 0.01188212, "auxiliary_loss_mlp": 0.01029813, "balance_loss_clip": 1.05952179, "balance_loss_mlp": 1.02161098, "epoch": 0.2862983226116756, "flos": 27563131715520.0, "grad_norm": 2.1004676587748605, "language_loss": 0.76037419, "learning_rate": 3.3497599469930816e-06, "loss": 0.78255439, "num_input_tokens_seen": 51008305, "step": 2381, "time_per_iteration": 3.36173939704895 }, { "auxiliary_loss_clip": 0.01209764, "auxiliary_loss_mlp": 0.01033002, "balance_loss_clip": 1.06307268, "balance_loss_mlp": 1.02440119, "epoch": 0.28641856550231465, "flos": 22053967767360.0, "grad_norm": 2.58263574890305, "language_loss": 0.83435822, "learning_rate": 3.349185020014807e-06, "loss": 0.85678589, "num_input_tokens_seen": 51025570, "step": 2382, "time_per_iteration": 2.540071964263916 }, { "auxiliary_loss_clip": 0.0118979, "auxiliary_loss_mlp": 0.01032798, "balance_loss_clip": 1.0569216, "balance_loss_mlp": 1.02425647, "epoch": 0.28653880839295376, "flos": 22378779175680.0, "grad_norm": 1.8234638442422932, "language_loss": 0.74236846, "learning_rate": 3.348609888366237e-06, "loss": 0.76459432, "num_input_tokens_seen": 51044585, "step": 2383, "time_per_iteration": 2.571429491043091 }, { "auxiliary_loss_clip": 0.0112639, "auxiliary_loss_mlp": 0.0103288, "balance_loss_clip": 1.04878473, "balance_loss_mlp": 1.02442765, "epoch": 0.28665905128359287, "flos": 23367975271200.0, "grad_norm": 2.0525855904649384, "language_loss": 0.62458372, "learning_rate": 3.348034552134619e-06, "loss": 0.64617646, "num_input_tokens_seen": 51063990, "step": 2384, "time_per_iteration": 3.5913188457489014 }, { "auxiliary_loss_clip": 0.01138537, "auxiliary_loss_mlp": 0.01027355, "balance_loss_clip": 1.05209422, "balance_loss_mlp": 1.01939797, "epoch": 0.2867792941742319, "flos": 20881698951360.0, "grad_norm": 1.8922062760430785, "language_loss": 0.83953846, "learning_rate": 3.3474590114072316e-06, "loss": 0.86119735, "num_input_tokens_seen": 51081990, "step": 2385, "time_per_iteration": 2.6702888011932373 }, { "auxiliary_loss_clip": 0.01157254, "auxiliary_loss_mlp": 0.01031494, "balance_loss_clip": 1.05736709, "balance_loss_mlp": 1.02247548, "epoch": 0.28689953706487104, "flos": 20664008669760.0, "grad_norm": 1.7829309775575113, "language_loss": 0.82817769, "learning_rate": 3.3468832662713836e-06, "loss": 0.85006523, "num_input_tokens_seen": 51100235, "step": 2386, "time_per_iteration": 2.6354358196258545 }, { "auxiliary_loss_clip": 0.0115852, "auxiliary_loss_mlp": 0.0102618, "balance_loss_clip": 1.05684352, "balance_loss_mlp": 1.01794815, "epoch": 0.28701977995551015, "flos": 12675536292000.0, "grad_norm": 2.1986291863338034, "language_loss": 0.83957183, "learning_rate": 3.346307316814415e-06, "loss": 0.86141884, "num_input_tokens_seen": 51115405, "step": 2387, "time_per_iteration": 2.6000359058380127 }, { "auxiliary_loss_clip": 0.01189531, "auxiliary_loss_mlp": 0.01032856, "balance_loss_clip": 1.05968678, "balance_loss_mlp": 1.02349854, "epoch": 0.2871400228461492, "flos": 21252369929280.0, "grad_norm": 2.046496494380374, "language_loss": 0.75558496, "learning_rate": 3.3457311631236965e-06, "loss": 0.77780879, "num_input_tokens_seen": 51136390, "step": 2388, "time_per_iteration": 2.5774505138397217 }, { "auxiliary_loss_clip": 0.01164038, "auxiliary_loss_mlp": 0.01034243, "balance_loss_clip": 1.05455482, "balance_loss_mlp": 1.02539802, "epoch": 0.2872602657367883, "flos": 25119266937120.0, "grad_norm": 2.0505391020923738, "language_loss": 0.84487647, "learning_rate": 3.345154805286631e-06, "loss": 0.86685932, "num_input_tokens_seen": 51156650, "step": 2389, "time_per_iteration": 3.491222381591797 }, { "auxiliary_loss_clip": 0.01183342, "auxiliary_loss_mlp": 0.01029134, "balance_loss_clip": 1.05630827, "balance_loss_mlp": 1.02055097, "epoch": 0.2873805086274274, "flos": 16646609235840.0, "grad_norm": 2.392146864613524, "language_loss": 0.76615876, "learning_rate": 3.344578243390651e-06, "loss": 0.78828353, "num_input_tokens_seen": 51172210, "step": 2390, "time_per_iteration": 2.62777042388916 }, { "auxiliary_loss_clip": 0.01176087, "auxiliary_loss_mlp": 0.01029495, "balance_loss_clip": 1.05937648, "balance_loss_mlp": 1.02053654, "epoch": 0.2875007515180665, "flos": 17420126399520.0, "grad_norm": 2.2365194553451317, "language_loss": 0.78566766, "learning_rate": 3.3440014775232206e-06, "loss": 0.80772346, "num_input_tokens_seen": 51190265, "step": 2391, "time_per_iteration": 2.591801643371582 }, { "auxiliary_loss_clip": 0.01160502, "auxiliary_loss_mlp": 0.01033151, "balance_loss_clip": 1.05244577, "balance_loss_mlp": 1.02509284, "epoch": 0.2876209944087056, "flos": 23434193368800.0, "grad_norm": 1.967870517223461, "language_loss": 0.71193379, "learning_rate": 3.343424507771834e-06, "loss": 0.73387033, "num_input_tokens_seen": 51208475, "step": 2392, "time_per_iteration": 2.6398746967315674 }, { "auxiliary_loss_clip": 0.01161808, "auxiliary_loss_mlp": 0.01029868, "balance_loss_clip": 1.05711389, "balance_loss_mlp": 1.02176738, "epoch": 0.2877412372993447, "flos": 13735511939040.0, "grad_norm": 6.762852908362851, "language_loss": 0.86303294, "learning_rate": 3.342847334224018e-06, "loss": 0.88494968, "num_input_tokens_seen": 51225875, "step": 2393, "time_per_iteration": 2.6270647048950195 }, { "auxiliary_loss_clip": 0.01088377, "auxiliary_loss_mlp": 0.01006639, "balance_loss_clip": 1.02078271, "balance_loss_mlp": 1.00511932, "epoch": 0.28786148018998375, "flos": 58079706823680.0, "grad_norm": 0.9591905661747259, "language_loss": 0.62443751, "learning_rate": 3.342269956967329e-06, "loss": 0.64538765, "num_input_tokens_seen": 51287780, "step": 2394, "time_per_iteration": 3.216602087020874 }, { "auxiliary_loss_clip": 0.01193732, "auxiliary_loss_mlp": 0.01035116, "balance_loss_clip": 1.05949736, "balance_loss_mlp": 1.02612782, "epoch": 0.28798172308062286, "flos": 23435055375840.0, "grad_norm": 2.6953635008831625, "language_loss": 0.72106224, "learning_rate": 3.341692376089355e-06, "loss": 0.74335074, "num_input_tokens_seen": 51303335, "step": 2395, "time_per_iteration": 2.5741357803344727 }, { "auxiliary_loss_clip": 0.01185825, "auxiliary_loss_mlp": 0.01029209, "balance_loss_clip": 1.0580132, "balance_loss_mlp": 1.0208528, "epoch": 0.288101965971262, "flos": 25110036278400.0, "grad_norm": 3.2640186264123927, "language_loss": 0.84581196, "learning_rate": 3.3411145916777146e-06, "loss": 0.86796224, "num_input_tokens_seen": 51317495, "step": 2396, "time_per_iteration": 2.5939688682556152 }, { "auxiliary_loss_clip": 0.01171615, "auxiliary_loss_mlp": 0.01031779, "balance_loss_clip": 1.05635118, "balance_loss_mlp": 1.02333927, "epoch": 0.28822220886190103, "flos": 16252562705280.0, "grad_norm": 2.1585607573017502, "language_loss": 0.90759516, "learning_rate": 3.3405366038200566e-06, "loss": 0.92962909, "num_input_tokens_seen": 51336430, "step": 2397, "time_per_iteration": 2.5949363708496094 }, { "auxiliary_loss_clip": 0.01183359, "auxiliary_loss_mlp": 0.01036091, "balance_loss_clip": 1.06359982, "balance_loss_mlp": 1.02744794, "epoch": 0.28834245175254014, "flos": 24535646716320.0, "grad_norm": 2.3447281141443943, "language_loss": 0.84881634, "learning_rate": 3.3399584126040617e-06, "loss": 0.87101084, "num_input_tokens_seen": 51355930, "step": 2398, "time_per_iteration": 2.6453142166137695 }, { "auxiliary_loss_clip": 0.01206932, "auxiliary_loss_mlp": 0.00763128, "balance_loss_clip": 1.06172025, "balance_loss_mlp": 1.00023961, "epoch": 0.2884626946431792, "flos": 24571449537120.0, "grad_norm": 1.7842695449607633, "language_loss": 0.9070673, "learning_rate": 3.339380018117441e-06, "loss": 0.92676789, "num_input_tokens_seen": 51376765, "step": 2399, "time_per_iteration": 2.5663950443267822 }, { "auxiliary_loss_clip": 0.0118406, "auxiliary_loss_mlp": 0.0103128, "balance_loss_clip": 1.05789173, "balance_loss_mlp": 1.02247632, "epoch": 0.2885829375338183, "flos": 16544660151360.0, "grad_norm": 2.373195301794254, "language_loss": 0.78075898, "learning_rate": 3.3388014204479366e-06, "loss": 0.80291241, "num_input_tokens_seen": 51394570, "step": 2400, "time_per_iteration": 2.5368540287017822 }, { "auxiliary_loss_clip": 0.01206901, "auxiliary_loss_mlp": 0.01033222, "balance_loss_clip": 1.06129384, "balance_loss_mlp": 1.02515697, "epoch": 0.2887031804244574, "flos": 24061230805920.0, "grad_norm": 2.687956196124405, "language_loss": 0.91165221, "learning_rate": 3.338222619683321e-06, "loss": 0.93405342, "num_input_tokens_seen": 51414535, "step": 2401, "time_per_iteration": 2.5497725009918213 }, { "auxiliary_loss_clip": 0.0117928, "auxiliary_loss_mlp": 0.01040082, "balance_loss_clip": 1.05735016, "balance_loss_mlp": 1.03154087, "epoch": 0.2888234233150965, "flos": 23330699855040.0, "grad_norm": 2.6597382900492312, "language_loss": 0.73548943, "learning_rate": 3.337643615911398e-06, "loss": 0.75768304, "num_input_tokens_seen": 51434160, "step": 2402, "time_per_iteration": 2.6037418842315674 }, { "auxiliary_loss_clip": 0.01193834, "auxiliary_loss_mlp": 0.01031107, "balance_loss_clip": 1.05950069, "balance_loss_mlp": 1.02293539, "epoch": 0.2889436662057356, "flos": 22272771474720.0, "grad_norm": 2.213155841345337, "language_loss": 0.79004234, "learning_rate": 3.3370644092200026e-06, "loss": 0.81229174, "num_input_tokens_seen": 51451435, "step": 2403, "time_per_iteration": 2.5974531173706055 }, { "auxiliary_loss_clip": 0.01146917, "auxiliary_loss_mlp": 0.01032161, "balance_loss_clip": 1.05030823, "balance_loss_mlp": 1.02379847, "epoch": 0.2890639090963747, "flos": 21616935024000.0, "grad_norm": 2.0181007088041016, "language_loss": 0.78258312, "learning_rate": 3.3364849996969985e-06, "loss": 0.80437386, "num_input_tokens_seen": 51471455, "step": 2404, "time_per_iteration": 2.6645233631134033 }, { "auxiliary_loss_clip": 0.01188892, "auxiliary_loss_mlp": 0.01032498, "balance_loss_clip": 1.05900705, "balance_loss_mlp": 1.0244633, "epoch": 0.28918415198701375, "flos": 28585544610720.0, "grad_norm": 2.210600351190991, "language_loss": 0.85219103, "learning_rate": 3.335905387430283e-06, "loss": 0.87440491, "num_input_tokens_seen": 51492890, "step": 2405, "time_per_iteration": 2.622821807861328 }, { "auxiliary_loss_clip": 0.01179087, "auxiliary_loss_mlp": 0.01031132, "balance_loss_clip": 1.05583584, "balance_loss_mlp": 1.02297211, "epoch": 0.28930439487765286, "flos": 21944691623040.0, "grad_norm": 1.7823606764852415, "language_loss": 0.82758301, "learning_rate": 3.335325572507782e-06, "loss": 0.84968519, "num_input_tokens_seen": 51513390, "step": 2406, "time_per_iteration": 2.6305298805236816 }, { "auxiliary_loss_clip": 0.01211314, "auxiliary_loss_mlp": 0.00764069, "balance_loss_clip": 1.06605065, "balance_loss_mlp": 1.00035739, "epoch": 0.28942463776829197, "flos": 19281915386400.0, "grad_norm": 1.7217627663602293, "language_loss": 0.73516119, "learning_rate": 3.3347455550174537e-06, "loss": 0.75491506, "num_input_tokens_seen": 51532730, "step": 2407, "time_per_iteration": 4.069361448287964 }, { "auxiliary_loss_clip": 0.0115667, "auxiliary_loss_mlp": 0.01029959, "balance_loss_clip": 1.05336976, "balance_loss_mlp": 1.0215075, "epoch": 0.289544880658931, "flos": 14645703499200.0, "grad_norm": 1.7636913289910754, "language_loss": 0.68104994, "learning_rate": 3.3341653350472864e-06, "loss": 0.70291615, "num_input_tokens_seen": 51549560, "step": 2408, "time_per_iteration": 2.6175804138183594 }, { "auxiliary_loss_clip": 0.01215546, "auxiliary_loss_mlp": 0.01031669, "balance_loss_clip": 1.06394863, "balance_loss_mlp": 1.02204847, "epoch": 0.28966512354957014, "flos": 28621886185920.0, "grad_norm": 2.6991696466822765, "language_loss": 0.69488668, "learning_rate": 3.333584912685298e-06, "loss": 0.71735883, "num_input_tokens_seen": 51568180, "step": 2409, "time_per_iteration": 2.613995313644409 }, { "auxiliary_loss_clip": 0.01060951, "auxiliary_loss_mlp": 0.01005129, "balance_loss_clip": 1.02074599, "balance_loss_mlp": 1.0036211, "epoch": 0.28978536644020925, "flos": 64711796073120.0, "grad_norm": 0.883074187950859, "language_loss": 0.55565673, "learning_rate": 3.3330042880195385e-06, "loss": 0.57631755, "num_input_tokens_seen": 51622530, "step": 2410, "time_per_iteration": 3.9964849948883057 }, { "auxiliary_loss_clip": 0.01171817, "auxiliary_loss_mlp": 0.01027272, "balance_loss_clip": 1.05270696, "balance_loss_mlp": 1.01922524, "epoch": 0.2899056093308483, "flos": 18624642257280.0, "grad_norm": 1.8946580293923816, "language_loss": 0.78696668, "learning_rate": 3.3324234611380888e-06, "loss": 0.80895758, "num_input_tokens_seen": 51641260, "step": 2411, "time_per_iteration": 2.6162281036376953 }, { "auxiliary_loss_clip": 0.01157498, "auxiliary_loss_mlp": 0.01037367, "balance_loss_clip": 1.05876446, "balance_loss_mlp": 1.02875447, "epoch": 0.2900258522214874, "flos": 22893738945600.0, "grad_norm": 1.7764012237325524, "language_loss": 0.81763238, "learning_rate": 3.3318424321290596e-06, "loss": 0.83958107, "num_input_tokens_seen": 51660975, "step": 2412, "time_per_iteration": 2.6618990898132324 }, { "auxiliary_loss_clip": 0.01068564, "auxiliary_loss_mlp": 0.01004288, "balance_loss_clip": 1.02497005, "balance_loss_mlp": 1.00300086, "epoch": 0.2901460951121265, "flos": 71106045626400.0, "grad_norm": 0.8338811278138784, "language_loss": 0.59967995, "learning_rate": 3.3312612010805917e-06, "loss": 0.62040854, "num_input_tokens_seen": 51720550, "step": 2413, "time_per_iteration": 3.2762982845306396 }, { "auxiliary_loss_clip": 0.01165965, "auxiliary_loss_mlp": 0.01037389, "balance_loss_clip": 1.05614889, "balance_loss_mlp": 1.02902055, "epoch": 0.2902663380027656, "flos": 32160990677760.0, "grad_norm": 1.823440660321823, "language_loss": 0.7010355, "learning_rate": 3.330679768080858e-06, "loss": 0.72306901, "num_input_tokens_seen": 51744435, "step": 2414, "time_per_iteration": 2.70393443107605 }, { "auxiliary_loss_clip": 0.01191345, "auxiliary_loss_mlp": 0.01039333, "balance_loss_clip": 1.0623374, "balance_loss_mlp": 1.03057134, "epoch": 0.2903865808934047, "flos": 29351662880640.0, "grad_norm": 2.0547839046831697, "language_loss": 0.83458853, "learning_rate": 3.3300981332180627e-06, "loss": 0.85689533, "num_input_tokens_seen": 51763640, "step": 2415, "time_per_iteration": 2.6627938747406006 }, { "auxiliary_loss_clip": 0.01169954, "auxiliary_loss_mlp": 0.01039006, "balance_loss_clip": 1.0588541, "balance_loss_mlp": 1.03068209, "epoch": 0.29050682378404374, "flos": 17089029523200.0, "grad_norm": 1.9514114230204607, "language_loss": 0.80077827, "learning_rate": 3.3295162965804373e-06, "loss": 0.82286781, "num_input_tokens_seen": 51782135, "step": 2416, "time_per_iteration": 3.5188148021698 }, { "auxiliary_loss_clip": 0.01159687, "auxiliary_loss_mlp": 0.01029079, "balance_loss_clip": 1.05747533, "balance_loss_mlp": 1.02087164, "epoch": 0.29062706667468285, "flos": 17858236651680.0, "grad_norm": 1.97381354176952, "language_loss": 0.78321546, "learning_rate": 3.328934258256247e-06, "loss": 0.80510312, "num_input_tokens_seen": 51800200, "step": 2417, "time_per_iteration": 2.7095389366149902 }, { "auxiliary_loss_clip": 0.01191677, "auxiliary_loss_mlp": 0.01030511, "balance_loss_clip": 1.0592345, "balance_loss_mlp": 1.02160585, "epoch": 0.29074730956532197, "flos": 24279819011520.0, "grad_norm": 5.220544952667399, "language_loss": 0.6688993, "learning_rate": 3.3283520183337856e-06, "loss": 0.69112122, "num_input_tokens_seen": 51819905, "step": 2418, "time_per_iteration": 2.5779225826263428 }, { "auxiliary_loss_clip": 0.01175792, "auxiliary_loss_mlp": 0.01035371, "balance_loss_clip": 1.05962038, "balance_loss_mlp": 1.02737784, "epoch": 0.290867552455961, "flos": 22340965005120.0, "grad_norm": 2.0254784148450726, "language_loss": 0.6900602, "learning_rate": 3.3277695769013797e-06, "loss": 0.71217191, "num_input_tokens_seen": 51839350, "step": 2419, "time_per_iteration": 2.6588613986968994 }, { "auxiliary_loss_clip": 0.01193811, "auxiliary_loss_mlp": 0.01032611, "balance_loss_clip": 1.06148398, "balance_loss_mlp": 1.02320552, "epoch": 0.29098779534660013, "flos": 23186159644320.0, "grad_norm": 10.381094505966, "language_loss": 0.77327657, "learning_rate": 3.327186934047385e-06, "loss": 0.79554081, "num_input_tokens_seen": 51858045, "step": 2420, "time_per_iteration": 2.5861308574676514 }, { "auxiliary_loss_clip": 0.011631, "auxiliary_loss_mlp": 0.01032854, "balance_loss_clip": 1.05093324, "balance_loss_mlp": 1.0246644, "epoch": 0.29110803823723924, "flos": 15304197804960.0, "grad_norm": 1.9316449552359007, "language_loss": 0.6548385, "learning_rate": 3.3266040898601877e-06, "loss": 0.67679805, "num_input_tokens_seen": 51875880, "step": 2421, "time_per_iteration": 2.640913963317871 }, { "auxiliary_loss_clip": 0.01138402, "auxiliary_loss_mlp": 0.01040601, "balance_loss_clip": 1.05127263, "balance_loss_mlp": 1.03219032, "epoch": 0.2912282811278783, "flos": 22595356031520.0, "grad_norm": 1.7697936689665326, "language_loss": 0.78507024, "learning_rate": 3.3260210444282045e-06, "loss": 0.80686027, "num_input_tokens_seen": 51893835, "step": 2422, "time_per_iteration": 2.6600658893585205 }, { "auxiliary_loss_clip": 0.01182011, "auxiliary_loss_mlp": 0.01028547, "balance_loss_clip": 1.05614924, "balance_loss_mlp": 1.02049434, "epoch": 0.2913485240185174, "flos": 24497904379680.0, "grad_norm": 2.0380718247922043, "language_loss": 0.72742796, "learning_rate": 3.325437797839883e-06, "loss": 0.74953353, "num_input_tokens_seen": 51912205, "step": 2423, "time_per_iteration": 2.6483495235443115 }, { "auxiliary_loss_clip": 0.01206494, "auxiliary_loss_mlp": 0.01029865, "balance_loss_clip": 1.06036115, "balance_loss_mlp": 1.02146029, "epoch": 0.2914687669091565, "flos": 17931027552960.0, "grad_norm": 2.3439067742109816, "language_loss": 0.75139427, "learning_rate": 3.3248543501837015e-06, "loss": 0.77375787, "num_input_tokens_seen": 51929410, "step": 2424, "time_per_iteration": 2.514359474182129 }, { "auxiliary_loss_clip": 0.01155533, "auxiliary_loss_mlp": 0.0103168, "balance_loss_clip": 1.06073773, "balance_loss_mlp": 1.02320468, "epoch": 0.2915890097997956, "flos": 22529317518720.0, "grad_norm": 1.8197974855912709, "language_loss": 0.77517688, "learning_rate": 3.3242707015481684e-06, "loss": 0.79704899, "num_input_tokens_seen": 51949345, "step": 2425, "time_per_iteration": 2.7476577758789062 }, { "auxiliary_loss_clip": 0.01173854, "auxiliary_loss_mlp": 0.01030522, "balance_loss_clip": 1.05426335, "balance_loss_mlp": 1.02186108, "epoch": 0.2917092526904347, "flos": 13845219086880.0, "grad_norm": 1.695955258801741, "language_loss": 0.80942225, "learning_rate": 3.323686852021823e-06, "loss": 0.83146596, "num_input_tokens_seen": 51966855, "step": 2426, "time_per_iteration": 2.5948116779327393 }, { "auxiliary_loss_clip": 0.01166575, "auxiliary_loss_mlp": 0.01032232, "balance_loss_clip": 1.05540729, "balance_loss_mlp": 1.02390492, "epoch": 0.2918294955810738, "flos": 22674863404320.0, "grad_norm": 1.9614923928982282, "language_loss": 0.79652423, "learning_rate": 3.323102801693235e-06, "loss": 0.81851232, "num_input_tokens_seen": 51985620, "step": 2427, "time_per_iteration": 2.683980703353882 }, { "auxiliary_loss_clip": 0.01182081, "auxiliary_loss_mlp": 0.01031547, "balance_loss_clip": 1.05655146, "balance_loss_mlp": 1.02329457, "epoch": 0.29194973847171285, "flos": 23438287902240.0, "grad_norm": 2.213825989358753, "language_loss": 0.80581725, "learning_rate": 3.322518550651003e-06, "loss": 0.82795352, "num_input_tokens_seen": 52004930, "step": 2428, "time_per_iteration": 2.619575023651123 }, { "auxiliary_loss_clip": 0.01182634, "auxiliary_loss_mlp": 0.01027405, "balance_loss_clip": 1.05753255, "balance_loss_mlp": 1.01860678, "epoch": 0.29206998136235196, "flos": 21909068387040.0, "grad_norm": 3.567936502212087, "language_loss": 0.80962753, "learning_rate": 3.3219340989837586e-06, "loss": 0.83172792, "num_input_tokens_seen": 52024920, "step": 2429, "time_per_iteration": 2.6509289741516113 }, { "auxiliary_loss_clip": 0.0117656, "auxiliary_loss_mlp": 0.01033171, "balance_loss_clip": 1.05720866, "balance_loss_mlp": 1.0255239, "epoch": 0.292190224252991, "flos": 23215928415840.0, "grad_norm": 1.9557625169415898, "language_loss": 0.80174524, "learning_rate": 3.3213494467801625e-06, "loss": 0.82384253, "num_input_tokens_seen": 52044095, "step": 2430, "time_per_iteration": 2.6131832599639893 }, { "auxiliary_loss_clip": 0.01114997, "auxiliary_loss_mlp": 0.01032032, "balance_loss_clip": 1.04777682, "balance_loss_mlp": 1.02297783, "epoch": 0.2923104671436301, "flos": 20740822270560.0, "grad_norm": 2.0437046199076905, "language_loss": 0.71793938, "learning_rate": 3.3207645941289063e-06, "loss": 0.73940969, "num_input_tokens_seen": 52062440, "step": 2431, "time_per_iteration": 2.7262861728668213 }, { "auxiliary_loss_clip": 0.01186979, "auxiliary_loss_mlp": 0.00763885, "balance_loss_clip": 1.05901563, "balance_loss_mlp": 1.00016522, "epoch": 0.29243071003426924, "flos": 35809119895200.0, "grad_norm": 1.806116683641446, "language_loss": 0.80370235, "learning_rate": 3.320179541118711e-06, "loss": 0.82321095, "num_input_tokens_seen": 52084940, "step": 2432, "time_per_iteration": 2.7042934894561768 }, { "auxiliary_loss_clip": 0.01086087, "auxiliary_loss_mlp": 0.00999804, "balance_loss_clip": 1.01896763, "balance_loss_mlp": 0.99830228, "epoch": 0.2925509529249083, "flos": 58081610422560.0, "grad_norm": 0.9921869019411106, "language_loss": 0.60326105, "learning_rate": 3.3195942878383293e-06, "loss": 0.62411994, "num_input_tokens_seen": 52141040, "step": 2433, "time_per_iteration": 4.602193355560303 }, { "auxiliary_loss_clip": 0.01191572, "auxiliary_loss_mlp": 0.01033118, "balance_loss_clip": 1.06135917, "balance_loss_mlp": 1.0245049, "epoch": 0.2926711958155474, "flos": 21397125641760.0, "grad_norm": 2.3097992911065868, "language_loss": 0.7815268, "learning_rate": 3.319008834376543e-06, "loss": 0.80377376, "num_input_tokens_seen": 52160730, "step": 2434, "time_per_iteration": 2.5932013988494873 }, { "auxiliary_loss_clip": 0.01165843, "auxiliary_loss_mlp": 0.01031591, "balance_loss_clip": 1.05181623, "balance_loss_mlp": 1.02356231, "epoch": 0.2927914387061865, "flos": 23185800474720.0, "grad_norm": 2.6477414030723683, "language_loss": 0.88353479, "learning_rate": 3.3184231808221654e-06, "loss": 0.90550911, "num_input_tokens_seen": 52175055, "step": 2435, "time_per_iteration": 2.6487412452697754 }, { "auxiliary_loss_clip": 0.01161538, "auxiliary_loss_mlp": 0.01029799, "balance_loss_clip": 1.05665994, "balance_loss_mlp": 1.02158523, "epoch": 0.29291168159682557, "flos": 22455556859520.0, "grad_norm": 1.8946595309832526, "language_loss": 0.62978387, "learning_rate": 3.3178373272640394e-06, "loss": 0.65169728, "num_input_tokens_seen": 52194150, "step": 2436, "time_per_iteration": 3.5758843421936035 }, { "auxiliary_loss_clip": 0.01201356, "auxiliary_loss_mlp": 0.01027734, "balance_loss_clip": 1.05911446, "balance_loss_mlp": 1.01935387, "epoch": 0.2930319244874647, "flos": 21170635704960.0, "grad_norm": 2.6595637227021927, "language_loss": 0.85312933, "learning_rate": 3.3172512737910387e-06, "loss": 0.87542021, "num_input_tokens_seen": 52211660, "step": 2437, "time_per_iteration": 2.5506908893585205 }, { "auxiliary_loss_clip": 0.01190281, "auxiliary_loss_mlp": 0.0103007, "balance_loss_clip": 1.05895126, "balance_loss_mlp": 1.02142692, "epoch": 0.2931521673781038, "flos": 31357848410400.0, "grad_norm": 2.0861950192493706, "language_loss": 0.88018554, "learning_rate": 3.3166650204920674e-06, "loss": 0.90238905, "num_input_tokens_seen": 52232830, "step": 2438, "time_per_iteration": 2.6634299755096436 }, { "auxiliary_loss_clip": 0.01192524, "auxiliary_loss_mlp": 0.01032755, "balance_loss_clip": 1.06172371, "balance_loss_mlp": 1.02431476, "epoch": 0.29327241026874284, "flos": 24200994060960.0, "grad_norm": 2.488338458530495, "language_loss": 0.81597704, "learning_rate": 3.316078567456059e-06, "loss": 0.83822978, "num_input_tokens_seen": 52250670, "step": 2439, "time_per_iteration": 2.675952434539795 }, { "auxiliary_loss_clip": 0.01135698, "auxiliary_loss_mlp": 0.01026933, "balance_loss_clip": 1.05167222, "balance_loss_mlp": 1.01889277, "epoch": 0.29339265315938196, "flos": 24242615429280.0, "grad_norm": 1.6191033449481935, "language_loss": 0.7616787, "learning_rate": 3.3154919147719786e-06, "loss": 0.78330505, "num_input_tokens_seen": 52271685, "step": 2440, "time_per_iteration": 2.7167651653289795 }, { "auxiliary_loss_clip": 0.01188664, "auxiliary_loss_mlp": 0.01029464, "balance_loss_clip": 1.0580101, "balance_loss_mlp": 1.02092886, "epoch": 0.29351289605002107, "flos": 16946644330080.0, "grad_norm": 2.2627403660693814, "language_loss": 0.86371303, "learning_rate": 3.31490506252882e-06, "loss": 0.8858943, "num_input_tokens_seen": 52291065, "step": 2441, "time_per_iteration": 3.4878382682800293 }, { "auxiliary_loss_clip": 0.0115177, "auxiliary_loss_mlp": 0.01031387, "balance_loss_clip": 1.05226827, "balance_loss_mlp": 1.02328408, "epoch": 0.2936331389406601, "flos": 19829086281120.0, "grad_norm": 4.034333080955411, "language_loss": 0.84230638, "learning_rate": 3.31431801081561e-06, "loss": 0.86413789, "num_input_tokens_seen": 52310000, "step": 2442, "time_per_iteration": 2.6407697200775146 }, { "auxiliary_loss_clip": 0.01067637, "auxiliary_loss_mlp": 0.01000425, "balance_loss_clip": 1.01605034, "balance_loss_mlp": 0.99886364, "epoch": 0.29375338183129923, "flos": 71416855808640.0, "grad_norm": 0.902537603078251, "language_loss": 0.67972708, "learning_rate": 3.313730759721402e-06, "loss": 0.70040774, "num_input_tokens_seen": 52372930, "step": 2443, "time_per_iteration": 3.267057418823242 }, { "auxiliary_loss_clip": 0.01173847, "auxiliary_loss_mlp": 0.01033404, "balance_loss_clip": 1.05996644, "balance_loss_mlp": 1.02599239, "epoch": 0.29387362472193834, "flos": 22054506521760.0, "grad_norm": 2.6519379871600646, "language_loss": 0.86487973, "learning_rate": 3.313143309335282e-06, "loss": 0.88695222, "num_input_tokens_seen": 52391420, "step": 2444, "time_per_iteration": 2.6267075538635254 }, { "auxiliary_loss_clip": 0.01159349, "auxiliary_loss_mlp": 0.01028726, "balance_loss_clip": 1.05838192, "balance_loss_mlp": 1.02097714, "epoch": 0.2939938676125774, "flos": 22966422096000.0, "grad_norm": 1.959678752376201, "language_loss": 0.84649479, "learning_rate": 3.3125556597463665e-06, "loss": 0.8683756, "num_input_tokens_seen": 52410725, "step": 2445, "time_per_iteration": 2.6302614212036133 }, { "auxiliary_loss_clip": 0.0118942, "auxiliary_loss_mlp": 0.01029384, "balance_loss_clip": 1.06259298, "balance_loss_mlp": 1.02192712, "epoch": 0.2941141105032165, "flos": 31358710417440.0, "grad_norm": 1.6663763270108987, "language_loss": 0.66317236, "learning_rate": 3.311967811043801e-06, "loss": 0.68536031, "num_input_tokens_seen": 52432645, "step": 2446, "time_per_iteration": 2.6677019596099854 }, { "auxiliary_loss_clip": 0.01190389, "auxiliary_loss_mlp": 0.01037411, "balance_loss_clip": 1.06254029, "balance_loss_mlp": 1.02939451, "epoch": 0.29423435339385556, "flos": 23222142049920.0, "grad_norm": 2.1163123745526122, "language_loss": 0.81962705, "learning_rate": 3.3113797633167617e-06, "loss": 0.84190506, "num_input_tokens_seen": 52450940, "step": 2447, "time_per_iteration": 2.5928142070770264 }, { "auxiliary_loss_clip": 0.01202923, "auxiliary_loss_mlp": 0.01030313, "balance_loss_clip": 1.06019509, "balance_loss_mlp": 1.02236462, "epoch": 0.2943545962844947, "flos": 26864057633280.0, "grad_norm": 1.98290800126595, "language_loss": 0.68772173, "learning_rate": 3.310791516654455e-06, "loss": 0.71005404, "num_input_tokens_seen": 52468000, "step": 2448, "time_per_iteration": 2.589033365249634 }, { "auxiliary_loss_clip": 0.01165587, "auxiliary_loss_mlp": 0.01029237, "balance_loss_clip": 1.0558337, "balance_loss_mlp": 1.02125025, "epoch": 0.2944748391751338, "flos": 20231681048160.0, "grad_norm": 2.4593327046968136, "language_loss": 0.79626936, "learning_rate": 3.3102030711461177e-06, "loss": 0.81821764, "num_input_tokens_seen": 52487575, "step": 2449, "time_per_iteration": 2.654857635498047 }, { "auxiliary_loss_clip": 0.01163629, "auxiliary_loss_mlp": 0.01030648, "balance_loss_clip": 1.05625296, "balance_loss_mlp": 1.02239287, "epoch": 0.29459508206577284, "flos": 15960968096640.0, "grad_norm": 1.8573929934507967, "language_loss": 0.67918181, "learning_rate": 3.3096144268810156e-06, "loss": 0.70112467, "num_input_tokens_seen": 52506335, "step": 2450, "time_per_iteration": 2.629971504211426 }, { "auxiliary_loss_clip": 0.01178071, "auxiliary_loss_mlp": 0.01023575, "balance_loss_clip": 1.05483818, "balance_loss_mlp": 1.01531351, "epoch": 0.29471532495641195, "flos": 20412886086720.0, "grad_norm": 1.980041326102886, "language_loss": 0.72786319, "learning_rate": 3.3090255839484462e-06, "loss": 0.74987966, "num_input_tokens_seen": 52524330, "step": 2451, "time_per_iteration": 2.5572733879089355 }, { "auxiliary_loss_clip": 0.01175968, "auxiliary_loss_mlp": 0.01032162, "balance_loss_clip": 1.0563519, "balance_loss_mlp": 1.02359629, "epoch": 0.29483556784705106, "flos": 20376580428480.0, "grad_norm": 2.1543970539466843, "language_loss": 0.85317254, "learning_rate": 3.3084365424377366e-06, "loss": 0.8752538, "num_input_tokens_seen": 52543095, "step": 2452, "time_per_iteration": 2.6143100261688232 }, { "auxiliary_loss_clip": 0.01052386, "auxiliary_loss_mlp": 0.01020328, "balance_loss_clip": 1.0188055, "balance_loss_mlp": 1.01837337, "epoch": 0.2949558107376901, "flos": 68555670309600.0, "grad_norm": 0.7273133876762012, "language_loss": 0.56015551, "learning_rate": 3.307847302438245e-06, "loss": 0.58088267, "num_input_tokens_seen": 52597075, "step": 2453, "time_per_iteration": 3.112044095993042 }, { "auxiliary_loss_clip": 0.01129252, "auxiliary_loss_mlp": 0.01030615, "balance_loss_clip": 1.04790735, "balance_loss_mlp": 1.02193666, "epoch": 0.2950760536283292, "flos": 16107088653600.0, "grad_norm": 2.1151134453894445, "language_loss": 0.77634686, "learning_rate": 3.3072578640393562e-06, "loss": 0.79794556, "num_input_tokens_seen": 52614410, "step": 2454, "time_per_iteration": 2.648270845413208 }, { "auxiliary_loss_clip": 0.01173941, "auxiliary_loss_mlp": 0.01028295, "balance_loss_clip": 1.05622101, "balance_loss_mlp": 1.02063322, "epoch": 0.29519629651896834, "flos": 20483629721280.0, "grad_norm": 1.8616160714792307, "language_loss": 0.7947014, "learning_rate": 3.3066682273304886e-06, "loss": 0.81672376, "num_input_tokens_seen": 52632055, "step": 2455, "time_per_iteration": 2.648501396179199 }, { "auxiliary_loss_clip": 0.01192201, "auxiliary_loss_mlp": 0.00763568, "balance_loss_clip": 1.06005239, "balance_loss_mlp": 1.0002321, "epoch": 0.2953165394096074, "flos": 18916488284640.0, "grad_norm": 1.937778975677286, "language_loss": 0.78579986, "learning_rate": 3.3060783924010904e-06, "loss": 0.80535758, "num_input_tokens_seen": 52649980, "step": 2456, "time_per_iteration": 2.578819751739502 }, { "auxiliary_loss_clip": 0.01162143, "auxiliary_loss_mlp": 0.01026938, "balance_loss_clip": 1.05578864, "balance_loss_mlp": 1.01841462, "epoch": 0.2954367823002465, "flos": 20624470485120.0, "grad_norm": 2.609866364375765, "language_loss": 0.84252167, "learning_rate": 3.3054883593406387e-06, "loss": 0.86441255, "num_input_tokens_seen": 52664730, "step": 2457, "time_per_iteration": 2.669039249420166 }, { "auxiliary_loss_clip": 0.01177734, "auxiliary_loss_mlp": 0.01026673, "balance_loss_clip": 1.05599248, "balance_loss_mlp": 1.01828647, "epoch": 0.2955570251908856, "flos": 31175529946080.0, "grad_norm": 2.5316543196600305, "language_loss": 0.65345109, "learning_rate": 3.3048981282386404e-06, "loss": 0.67549515, "num_input_tokens_seen": 52686040, "step": 2458, "time_per_iteration": 3.4892501831054688 }, { "auxiliary_loss_clip": 0.01149786, "auxiliary_loss_mlp": 0.01032597, "balance_loss_clip": 1.05506194, "balance_loss_mlp": 1.02449036, "epoch": 0.29567726808152467, "flos": 21650331408480.0, "grad_norm": 1.9174588688815264, "language_loss": 0.82671082, "learning_rate": 3.304307699184634e-06, "loss": 0.84853458, "num_input_tokens_seen": 52704630, "step": 2459, "time_per_iteration": 3.404440402984619 }, { "auxiliary_loss_clip": 0.01172396, "auxiliary_loss_mlp": 0.01028673, "balance_loss_clip": 1.0572623, "balance_loss_mlp": 1.02113938, "epoch": 0.2957975109721638, "flos": 24243872522880.0, "grad_norm": 1.6202706588574844, "language_loss": 0.78431368, "learning_rate": 3.3037170722681866e-06, "loss": 0.80632436, "num_input_tokens_seen": 52725465, "step": 2460, "time_per_iteration": 2.64125657081604 }, { "auxiliary_loss_clip": 0.0115125, "auxiliary_loss_mlp": 0.01028986, "balance_loss_clip": 1.05433571, "balance_loss_mlp": 1.02153492, "epoch": 0.29591775386280283, "flos": 13479720151200.0, "grad_norm": 1.9485435841191232, "language_loss": 0.68080175, "learning_rate": 3.3031262475788956e-06, "loss": 0.70260406, "num_input_tokens_seen": 52742405, "step": 2461, "time_per_iteration": 2.620535135269165 }, { "auxiliary_loss_clip": 0.01169734, "auxiliary_loss_mlp": 0.01030315, "balance_loss_clip": 1.05546737, "balance_loss_mlp": 1.02279329, "epoch": 0.29603799675344195, "flos": 17749786597440.0, "grad_norm": 1.9349152318790646, "language_loss": 0.73241866, "learning_rate": 3.3025352252063897e-06, "loss": 0.75441909, "num_input_tokens_seen": 52761100, "step": 2462, "time_per_iteration": 3.4863734245300293 }, { "auxiliary_loss_clip": 0.01188246, "auxiliary_loss_mlp": 0.01036943, "balance_loss_clip": 1.06042099, "balance_loss_mlp": 1.02874732, "epoch": 0.29615823964408106, "flos": 22783924046880.0, "grad_norm": 1.9724484686786075, "language_loss": 0.75231099, "learning_rate": 3.3019440052403252e-06, "loss": 0.77456284, "num_input_tokens_seen": 52780965, "step": 2463, "time_per_iteration": 2.579895496368408 }, { "auxiliary_loss_clip": 0.01173537, "auxiliary_loss_mlp": 0.01030989, "balance_loss_clip": 1.05580425, "balance_loss_mlp": 1.0228585, "epoch": 0.2962784825347201, "flos": 23514203579040.0, "grad_norm": 1.8211153793670338, "language_loss": 0.70872676, "learning_rate": 3.30135258777039e-06, "loss": 0.73077208, "num_input_tokens_seen": 52800335, "step": 2464, "time_per_iteration": 2.652266025543213 }, { "auxiliary_loss_clip": 0.01193091, "auxiliary_loss_mlp": 0.00763489, "balance_loss_clip": 1.0569247, "balance_loss_mlp": 1.00021052, "epoch": 0.2963987254253592, "flos": 16362772690560.0, "grad_norm": 1.9795419509314838, "language_loss": 0.70175743, "learning_rate": 3.3007609728863024e-06, "loss": 0.72132325, "num_input_tokens_seen": 52818425, "step": 2465, "time_per_iteration": 2.5582435131073 }, { "auxiliary_loss_clip": 0.01120948, "auxiliary_loss_mlp": 0.01029363, "balance_loss_clip": 1.05317199, "balance_loss_mlp": 1.02138197, "epoch": 0.29651896831599833, "flos": 33472268492640.0, "grad_norm": 1.7441140288261563, "language_loss": 0.72811145, "learning_rate": 3.300169160677809e-06, "loss": 0.74961454, "num_input_tokens_seen": 52842340, "step": 2466, "time_per_iteration": 2.803490161895752 }, { "auxiliary_loss_clip": 0.01169814, "auxiliary_loss_mlp": 0.01028577, "balance_loss_clip": 1.05687261, "balance_loss_mlp": 1.01913595, "epoch": 0.2966392112066374, "flos": 23805367184160.0, "grad_norm": 2.3508690566295822, "language_loss": 0.77457565, "learning_rate": 3.2995771512346878e-06, "loss": 0.79655957, "num_input_tokens_seen": 52860690, "step": 2467, "time_per_iteration": 3.5547070503234863 }, { "auxiliary_loss_clip": 0.01205807, "auxiliary_loss_mlp": 0.00763034, "balance_loss_clip": 1.060969, "balance_loss_mlp": 1.00022173, "epoch": 0.2967594540972765, "flos": 19938470176320.0, "grad_norm": 1.9837607389396252, "language_loss": 0.72923881, "learning_rate": 3.298984944646746e-06, "loss": 0.74892724, "num_input_tokens_seen": 52879370, "step": 2468, "time_per_iteration": 2.548994541168213 }, { "auxiliary_loss_clip": 0.01191621, "auxiliary_loss_mlp": 0.00762678, "balance_loss_clip": 1.06004012, "balance_loss_mlp": 1.00025392, "epoch": 0.2968796969879156, "flos": 23732827701600.0, "grad_norm": 1.8160556997278494, "language_loss": 0.81537271, "learning_rate": 3.298392541003822e-06, "loss": 0.83491564, "num_input_tokens_seen": 52898775, "step": 2469, "time_per_iteration": 2.6062989234924316 }, { "auxiliary_loss_clip": 0.01170825, "auxiliary_loss_mlp": 0.01028947, "balance_loss_clip": 1.05586004, "balance_loss_mlp": 1.02108502, "epoch": 0.29699993987855466, "flos": 22893703028640.0, "grad_norm": 1.5470460191291318, "language_loss": 0.89670581, "learning_rate": 3.2977999403957806e-06, "loss": 0.9187035, "num_input_tokens_seen": 52917535, "step": 2470, "time_per_iteration": 2.600262403488159 }, { "auxiliary_loss_clip": 0.01203039, "auxiliary_loss_mlp": 0.01022387, "balance_loss_clip": 1.06146765, "balance_loss_mlp": 1.01443028, "epoch": 0.2971201827691938, "flos": 33832559469120.0, "grad_norm": 1.9592052067228407, "language_loss": 0.67114687, "learning_rate": 3.2972071429125207e-06, "loss": 0.6934011, "num_input_tokens_seen": 52938755, "step": 2471, "time_per_iteration": 2.6667847633361816 }, { "auxiliary_loss_clip": 0.01154916, "auxiliary_loss_mlp": 0.01031927, "balance_loss_clip": 1.05471945, "balance_loss_mlp": 1.02337337, "epoch": 0.2972404256598329, "flos": 22054362853920.0, "grad_norm": 2.064170698380891, "language_loss": 0.88300693, "learning_rate": 3.2966141486439682e-06, "loss": 0.90487534, "num_input_tokens_seen": 52957945, "step": 2472, "time_per_iteration": 2.6552815437316895 }, { "auxiliary_loss_clip": 0.01132059, "auxiliary_loss_mlp": 0.01029991, "balance_loss_clip": 1.04807353, "balance_loss_mlp": 1.0216707, "epoch": 0.29736066855047194, "flos": 31978600379520.0, "grad_norm": 2.8061720142689652, "language_loss": 0.63651353, "learning_rate": 3.29602095768008e-06, "loss": 0.65813398, "num_input_tokens_seen": 52978460, "step": 2473, "time_per_iteration": 2.793267011642456 }, { "auxiliary_loss_clip": 0.0116451, "auxiliary_loss_mlp": 0.01031015, "balance_loss_clip": 1.05443883, "balance_loss_mlp": 1.02304244, "epoch": 0.29748091144111105, "flos": 33510405915840.0, "grad_norm": 2.094783840462, "language_loss": 0.6359067, "learning_rate": 3.2954275701108437e-06, "loss": 0.65786195, "num_input_tokens_seen": 52999640, "step": 2474, "time_per_iteration": 2.698695659637451 }, { "auxiliary_loss_clip": 0.01138258, "auxiliary_loss_mlp": 0.01026927, "balance_loss_clip": 1.05178237, "balance_loss_mlp": 1.01899338, "epoch": 0.29760115433175016, "flos": 41283378946560.0, "grad_norm": 3.348972116343797, "language_loss": 0.68817198, "learning_rate": 3.294833986026275e-06, "loss": 0.70982385, "num_input_tokens_seen": 53022880, "step": 2475, "time_per_iteration": 2.8366878032684326 }, { "auxiliary_loss_clip": 0.01152464, "auxiliary_loss_mlp": 0.01033232, "balance_loss_clip": 1.05447471, "balance_loss_mlp": 1.02536416, "epoch": 0.2977213972223892, "flos": 24493342925760.0, "grad_norm": 1.9956069210418677, "language_loss": 0.85394049, "learning_rate": 3.29424020551642e-06, "loss": 0.87579751, "num_input_tokens_seen": 53041515, "step": 2476, "time_per_iteration": 2.6939313411712646 }, { "auxiliary_loss_clip": 0.01207889, "auxiliary_loss_mlp": 0.0103581, "balance_loss_clip": 1.0607121, "balance_loss_mlp": 1.02716184, "epoch": 0.2978416401130283, "flos": 21285119808480.0, "grad_norm": 2.058463824838783, "language_loss": 0.72227252, "learning_rate": 3.2936462286713546e-06, "loss": 0.74470955, "num_input_tokens_seen": 53059865, "step": 2477, "time_per_iteration": 2.5557973384857178 }, { "auxiliary_loss_clip": 0.01188905, "auxiliary_loss_mlp": 0.01027342, "balance_loss_clip": 1.0591296, "balance_loss_mlp": 1.01856792, "epoch": 0.2979618830036674, "flos": 25772158197120.0, "grad_norm": 2.1254624940305793, "language_loss": 0.77269459, "learning_rate": 3.2930520555811846e-06, "loss": 0.79485708, "num_input_tokens_seen": 53079490, "step": 2478, "time_per_iteration": 2.6064717769622803 }, { "auxiliary_loss_clip": 0.01088589, "auxiliary_loss_mlp": 0.00763596, "balance_loss_clip": 1.04387498, "balance_loss_mlp": 1.0002867, "epoch": 0.2980821258943065, "flos": 23476999996800.0, "grad_norm": 1.7723244127671638, "language_loss": 0.80086511, "learning_rate": 3.292457686336046e-06, "loss": 0.81938696, "num_input_tokens_seen": 53098810, "step": 2479, "time_per_iteration": 2.7821767330169678 }, { "auxiliary_loss_clip": 0.01080562, "auxiliary_loss_mlp": 0.01005872, "balance_loss_clip": 1.01981914, "balance_loss_mlp": 1.00416756, "epoch": 0.2982023687849456, "flos": 69752326741440.0, "grad_norm": 1.088996941903825, "language_loss": 0.61315238, "learning_rate": 3.291863121026105e-06, "loss": 0.63401675, "num_input_tokens_seen": 53162590, "step": 2480, "time_per_iteration": 3.2318437099456787 }, { "auxiliary_loss_clip": 0.01191276, "auxiliary_loss_mlp": 0.01038423, "balance_loss_clip": 1.05919826, "balance_loss_mlp": 1.03014421, "epoch": 0.29832261167558466, "flos": 29825935123200.0, "grad_norm": 1.9334283129509222, "language_loss": 0.76326269, "learning_rate": 3.2912683597415547e-06, "loss": 0.78555965, "num_input_tokens_seen": 53186675, "step": 2481, "time_per_iteration": 2.6407854557037354 }, { "auxiliary_loss_clip": 0.01162553, "auxiliary_loss_mlp": 0.01029847, "balance_loss_clip": 1.05387545, "balance_loss_mlp": 1.02181852, "epoch": 0.29844285456622377, "flos": 33910163243040.0, "grad_norm": 2.353627400635171, "language_loss": 0.7806083, "learning_rate": 3.2906734025726213e-06, "loss": 0.80253232, "num_input_tokens_seen": 53205940, "step": 2482, "time_per_iteration": 2.7013938426971436 }, { "auxiliary_loss_clip": 0.01194564, "auxiliary_loss_mlp": 0.01032694, "balance_loss_clip": 1.0597949, "balance_loss_mlp": 1.02413464, "epoch": 0.2985630974568629, "flos": 23876936908800.0, "grad_norm": 1.8305612498808996, "language_loss": 0.87523919, "learning_rate": 3.290078249609559e-06, "loss": 0.89751178, "num_input_tokens_seen": 53225360, "step": 2483, "time_per_iteration": 2.6324596405029297 }, { "auxiliary_loss_clip": 0.011852, "auxiliary_loss_mlp": 0.01034569, "balance_loss_clip": 1.05869019, "balance_loss_mlp": 1.02608705, "epoch": 0.29868334034750194, "flos": 21799109820480.0, "grad_norm": 2.068086438018397, "language_loss": 0.87923479, "learning_rate": 3.2894829009426514e-06, "loss": 0.90143251, "num_input_tokens_seen": 53243195, "step": 2484, "time_per_iteration": 3.368054151535034 }, { "auxiliary_loss_clip": 0.01185539, "auxiliary_loss_mlp": 0.01029814, "balance_loss_clip": 1.05802357, "balance_loss_mlp": 1.02176476, "epoch": 0.29880358323814105, "flos": 25666653333600.0, "grad_norm": 2.206816131385158, "language_loss": 0.77686399, "learning_rate": 3.288887356662213e-06, "loss": 0.79901743, "num_input_tokens_seen": 53264530, "step": 2485, "time_per_iteration": 3.378922939300537 }, { "auxiliary_loss_clip": 0.01079073, "auxiliary_loss_mlp": 0.01003041, "balance_loss_clip": 1.01740265, "balance_loss_mlp": 1.00134873, "epoch": 0.29892382612878016, "flos": 71005856472960.0, "grad_norm": 0.7663595371204152, "language_loss": 0.59776652, "learning_rate": 3.288291616858588e-06, "loss": 0.61858761, "num_input_tokens_seen": 53319920, "step": 2486, "time_per_iteration": 3.034541368484497 }, { "auxiliary_loss_clip": 0.01136774, "auxiliary_loss_mlp": 0.01030188, "balance_loss_clip": 1.05313182, "balance_loss_mlp": 1.02206349, "epoch": 0.2990440690194192, "flos": 25481138259840.0, "grad_norm": 1.6478888133962222, "language_loss": 0.76479441, "learning_rate": 3.287695681622149e-06, "loss": 0.7864641, "num_input_tokens_seen": 53339270, "step": 2487, "time_per_iteration": 3.5714402198791504 }, { "auxiliary_loss_clip": 0.01179891, "auxiliary_loss_mlp": 0.01026038, "balance_loss_clip": 1.05605197, "balance_loss_mlp": 1.01778257, "epoch": 0.2991643119100583, "flos": 23732360781120.0, "grad_norm": 1.839579939527233, "language_loss": 0.81059402, "learning_rate": 3.2870995510432982e-06, "loss": 0.83265328, "num_input_tokens_seen": 53357750, "step": 2488, "time_per_iteration": 2.6247966289520264 }, { "auxiliary_loss_clip": 0.01182849, "auxiliary_loss_mlp": 0.01029101, "balance_loss_clip": 1.05854034, "balance_loss_mlp": 1.02152491, "epoch": 0.29928455480069743, "flos": 27417549912960.0, "grad_norm": 1.7872927929301698, "language_loss": 0.77219963, "learning_rate": 3.2865032252124697e-06, "loss": 0.79431915, "num_input_tokens_seen": 53378265, "step": 2489, "time_per_iteration": 2.6103122234344482 }, { "auxiliary_loss_clip": 0.01170291, "auxiliary_loss_mlp": 0.01032769, "balance_loss_clip": 1.05319476, "balance_loss_mlp": 1.02437663, "epoch": 0.2994047976913365, "flos": 33692939881920.0, "grad_norm": 2.069109202988335, "language_loss": 0.77586561, "learning_rate": 3.2859067042201243e-06, "loss": 0.79789621, "num_input_tokens_seen": 53400305, "step": 2490, "time_per_iteration": 2.7301437854766846 }, { "auxiliary_loss_clip": 0.0110567, "auxiliary_loss_mlp": 0.01028898, "balance_loss_clip": 1.04525232, "balance_loss_mlp": 1.02054703, "epoch": 0.2995250405819756, "flos": 16763966696160.0, "grad_norm": 2.7719393078373837, "language_loss": 0.78249937, "learning_rate": 3.2853099881567544e-06, "loss": 0.80384505, "num_input_tokens_seen": 53418705, "step": 2491, "time_per_iteration": 2.649764060974121 }, { "auxiliary_loss_clip": 0.0119819, "auxiliary_loss_mlp": 0.01026779, "balance_loss_clip": 1.05856991, "balance_loss_mlp": 1.01914382, "epoch": 0.29964528347261465, "flos": 22963979742720.0, "grad_norm": 1.8493021303485684, "language_loss": 0.79473037, "learning_rate": 3.284713077112881e-06, "loss": 0.81698006, "num_input_tokens_seen": 53438135, "step": 2492, "time_per_iteration": 2.5716934204101562 }, { "auxiliary_loss_clip": 0.01166054, "auxiliary_loss_mlp": 0.01033679, "balance_loss_clip": 1.05694604, "balance_loss_mlp": 1.02535212, "epoch": 0.29976552636325376, "flos": 16938024259680.0, "grad_norm": 2.379553379630944, "language_loss": 0.86496913, "learning_rate": 3.284115971179056e-06, "loss": 0.88696641, "num_input_tokens_seen": 53452165, "step": 2493, "time_per_iteration": 3.4528863430023193 }, { "auxiliary_loss_clip": 0.0113094, "auxiliary_loss_mlp": 0.01032703, "balance_loss_clip": 1.0511601, "balance_loss_mlp": 1.02441227, "epoch": 0.2998857692538929, "flos": 17056459228800.0, "grad_norm": 1.9999313006250408, "language_loss": 0.78641593, "learning_rate": 3.283518670445859e-06, "loss": 0.80805236, "num_input_tokens_seen": 53470075, "step": 2494, "time_per_iteration": 2.7148542404174805 }, { "auxiliary_loss_clip": 0.01060957, "auxiliary_loss_mlp": 0.00754457, "balance_loss_clip": 1.01558971, "balance_loss_mlp": 0.99978018, "epoch": 0.30000601214453193, "flos": 68831539678080.0, "grad_norm": 0.6835441932951887, "language_loss": 0.5435828, "learning_rate": 3.2829211750038995e-06, "loss": 0.56173694, "num_input_tokens_seen": 53538705, "step": 2495, "time_per_iteration": 3.2461740970611572 }, { "auxiliary_loss_clip": 0.01154981, "auxiliary_loss_mlp": 0.01031603, "balance_loss_clip": 1.05456877, "balance_loss_mlp": 1.02293062, "epoch": 0.30012625503517104, "flos": 17603270953920.0, "grad_norm": 1.807893591271622, "language_loss": 0.88735455, "learning_rate": 3.2823234849438183e-06, "loss": 0.90922034, "num_input_tokens_seen": 53556740, "step": 2496, "time_per_iteration": 2.604569435119629 }, { "auxiliary_loss_clip": 0.01175223, "auxiliary_loss_mlp": 0.0102782, "balance_loss_clip": 1.05633652, "balance_loss_mlp": 1.01998794, "epoch": 0.30024649792581015, "flos": 21252585431040.0, "grad_norm": 3.3628959874068385, "language_loss": 0.75429976, "learning_rate": 3.2817256003562836e-06, "loss": 0.77633011, "num_input_tokens_seen": 53577115, "step": 2497, "time_per_iteration": 2.6361327171325684 }, { "auxiliary_loss_clip": 0.01133217, "auxiliary_loss_mlp": 0.01031347, "balance_loss_clip": 1.052531, "balance_loss_mlp": 1.02293682, "epoch": 0.3003667408164492, "flos": 23003266508640.0, "grad_norm": 1.8069458557008802, "language_loss": 0.65886784, "learning_rate": 3.281127521331995e-06, "loss": 0.6805135, "num_input_tokens_seen": 53598295, "step": 2498, "time_per_iteration": 2.7298786640167236 }, { "auxiliary_loss_clip": 0.01092377, "auxiliary_loss_mlp": 0.01004459, "balance_loss_clip": 1.01507068, "balance_loss_mlp": 1.00284982, "epoch": 0.3004869837070883, "flos": 64232351788320.0, "grad_norm": 0.8818528986426962, "language_loss": 0.60703462, "learning_rate": 3.2805292479616798e-06, "loss": 0.62800294, "num_input_tokens_seen": 53657160, "step": 2499, "time_per_iteration": 3.0221927165985107 }, { "auxiliary_loss_clip": 0.01180381, "auxiliary_loss_mlp": 0.01037732, "balance_loss_clip": 1.05919206, "balance_loss_mlp": 1.02935719, "epoch": 0.30060722659772743, "flos": 26248657291200.0, "grad_norm": 3.7051045278530266, "language_loss": 0.91517055, "learning_rate": 3.2799307803360955e-06, "loss": 0.9373517, "num_input_tokens_seen": 53673090, "step": 2500, "time_per_iteration": 2.6401984691619873 }, { "auxiliary_loss_clip": 0.01200861, "auxiliary_loss_mlp": 0.01031883, "balance_loss_clip": 1.05913866, "balance_loss_mlp": 1.02411008, "epoch": 0.3007274694883665, "flos": 24970883611680.0, "grad_norm": 1.666541762570796, "language_loss": 0.81253034, "learning_rate": 3.27933211854603e-06, "loss": 0.83485782, "num_input_tokens_seen": 53692145, "step": 2501, "time_per_iteration": 2.605635643005371 }, { "auxiliary_loss_clip": 0.01171204, "auxiliary_loss_mlp": 0.01026023, "balance_loss_clip": 1.05447221, "balance_loss_mlp": 1.01835752, "epoch": 0.3008477123790056, "flos": 17055848640480.0, "grad_norm": 1.7089543147744803, "language_loss": 0.87137979, "learning_rate": 3.278733262682299e-06, "loss": 0.89335203, "num_input_tokens_seen": 53710000, "step": 2502, "time_per_iteration": 2.579326629638672 }, { "auxiliary_loss_clip": 0.01200995, "auxiliary_loss_mlp": 0.01027973, "balance_loss_clip": 1.05726504, "balance_loss_mlp": 1.0199858, "epoch": 0.3009679552696447, "flos": 21506401786080.0, "grad_norm": 2.454963842428214, "language_loss": 0.82285821, "learning_rate": 3.2781342128357484e-06, "loss": 0.84514785, "num_input_tokens_seen": 53729355, "step": 2503, "time_per_iteration": 2.5579028129577637 }, { "auxiliary_loss_clip": 0.01156464, "auxiliary_loss_mlp": 0.01030508, "balance_loss_clip": 1.05278563, "balance_loss_mlp": 1.0224551, "epoch": 0.30108819816028376, "flos": 21134006794080.0, "grad_norm": 2.367088294593905, "language_loss": 0.80530906, "learning_rate": 3.2775349690972547e-06, "loss": 0.82717878, "num_input_tokens_seen": 53743505, "step": 2504, "time_per_iteration": 2.6242518424987793 }, { "auxiliary_loss_clip": 0.01072176, "auxiliary_loss_mlp": 0.01003174, "balance_loss_clip": 1.01189876, "balance_loss_mlp": 1.00164783, "epoch": 0.30120844105092287, "flos": 71126446459680.0, "grad_norm": 0.7631105408092563, "language_loss": 0.51864547, "learning_rate": 3.276935531557722e-06, "loss": 0.53939897, "num_input_tokens_seen": 53808725, "step": 2505, "time_per_iteration": 3.242612600326538 }, { "auxiliary_loss_clip": 0.01150457, "auxiliary_loss_mlp": 0.01030041, "balance_loss_clip": 1.0541594, "balance_loss_mlp": 1.02136898, "epoch": 0.301328683941562, "flos": 20264574595200.0, "grad_norm": 2.130584326939956, "language_loss": 0.79107153, "learning_rate": 3.2763359003080837e-06, "loss": 0.81287652, "num_input_tokens_seen": 53825680, "step": 2506, "time_per_iteration": 2.629648447036743 }, { "auxiliary_loss_clip": 0.01068561, "auxiliary_loss_mlp": 0.01003582, "balance_loss_clip": 1.01355529, "balance_loss_mlp": 1.00203216, "epoch": 0.30144892683220104, "flos": 70648223351520.0, "grad_norm": 0.7984696229643584, "language_loss": 0.62458062, "learning_rate": 3.2757360754393047e-06, "loss": 0.64530206, "num_input_tokens_seen": 53889750, "step": 2507, "time_per_iteration": 3.2859556674957275 }, { "auxiliary_loss_clip": 0.0118799, "auxiliary_loss_mlp": 0.01032519, "balance_loss_clip": 1.05886388, "balance_loss_mlp": 1.02373338, "epoch": 0.30156916972284015, "flos": 22820552957760.0, "grad_norm": 2.11744822390398, "language_loss": 0.63887119, "learning_rate": 3.2751360570423767e-06, "loss": 0.66107631, "num_input_tokens_seen": 53908135, "step": 2508, "time_per_iteration": 2.5963571071624756 }, { "auxiliary_loss_clip": 0.01174403, "auxiliary_loss_mlp": 0.01033704, "balance_loss_clip": 1.05623817, "balance_loss_mlp": 1.0249958, "epoch": 0.3016894126134792, "flos": 29899192944960.0, "grad_norm": 2.496290316512688, "language_loss": 0.75480163, "learning_rate": 3.2745358452083236e-06, "loss": 0.77688271, "num_input_tokens_seen": 53931035, "step": 2509, "time_per_iteration": 2.6474673748016357 }, { "auxiliary_loss_clip": 0.01186579, "auxiliary_loss_mlp": 0.01030991, "balance_loss_clip": 1.05753696, "balance_loss_mlp": 1.02347493, "epoch": 0.3018096555041183, "flos": 21546335057280.0, "grad_norm": 1.3543470835100402, "language_loss": 0.82138884, "learning_rate": 3.2739354400281955e-06, "loss": 0.84356451, "num_input_tokens_seen": 53952255, "step": 2510, "time_per_iteration": 4.150083541870117 }, { "auxiliary_loss_clip": 0.01057919, "auxiliary_loss_mlp": 0.00753918, "balance_loss_clip": 1.01178503, "balance_loss_mlp": 0.99970198, "epoch": 0.3019298983947574, "flos": 59136306276480.0, "grad_norm": 0.8681829505200581, "language_loss": 0.63733453, "learning_rate": 3.2733348415930744e-06, "loss": 0.65545291, "num_input_tokens_seen": 54014125, "step": 2511, "time_per_iteration": 3.2310612201690674 }, { "auxiliary_loss_clip": 0.01150502, "auxiliary_loss_mlp": 0.01028048, "balance_loss_clip": 1.05295503, "balance_loss_mlp": 1.01997769, "epoch": 0.3020501412853965, "flos": 34423075746240.0, "grad_norm": 2.0853792578428054, "language_loss": 0.80785525, "learning_rate": 3.27273404999407e-06, "loss": 0.82964075, "num_input_tokens_seen": 54036345, "step": 2512, "time_per_iteration": 2.7431013584136963 }, { "auxiliary_loss_clip": 0.01068898, "auxiliary_loss_mlp": 0.01002965, "balance_loss_clip": 1.01309168, "balance_loss_mlp": 1.00139129, "epoch": 0.3021703841760356, "flos": 71008298826240.0, "grad_norm": 0.7997813067893021, "language_loss": 0.60444546, "learning_rate": 3.272133065322322e-06, "loss": 0.62516409, "num_input_tokens_seen": 54094615, "step": 2513, "time_per_iteration": 4.05895733833313 }, { "auxiliary_loss_clip": 0.01198289, "auxiliary_loss_mlp": 0.01030844, "balance_loss_clip": 1.05551338, "balance_loss_mlp": 1.02240348, "epoch": 0.3022906270666747, "flos": 21510532236480.0, "grad_norm": 1.5897292750588088, "language_loss": 0.79746914, "learning_rate": 3.271531887669e-06, "loss": 0.81976044, "num_input_tokens_seen": 54114675, "step": 2514, "time_per_iteration": 2.5753519535064697 }, { "auxiliary_loss_clip": 0.01146197, "auxiliary_loss_mlp": 0.01031226, "balance_loss_clip": 1.05037665, "balance_loss_mlp": 1.02185619, "epoch": 0.30241086995731375, "flos": 31132004978880.0, "grad_norm": 2.067541267609664, "language_loss": 0.63348436, "learning_rate": 3.2709305171253015e-06, "loss": 0.65525866, "num_input_tokens_seen": 54134795, "step": 2515, "time_per_iteration": 2.727205276489258 }, { "auxiliary_loss_clip": 0.01186198, "auxiliary_loss_mlp": 0.01029806, "balance_loss_clip": 1.05804002, "balance_loss_mlp": 1.02158046, "epoch": 0.30253111284795287, "flos": 23511545724000.0, "grad_norm": 1.9568291427475706, "language_loss": 0.7749818, "learning_rate": 3.2703289537824536e-06, "loss": 0.79714179, "num_input_tokens_seen": 54154595, "step": 2516, "time_per_iteration": 2.5945451259613037 }, { "auxiliary_loss_clip": 0.0114714, "auxiliary_loss_mlp": 0.01037418, "balance_loss_clip": 1.05274391, "balance_loss_mlp": 1.02864969, "epoch": 0.302651355738592, "flos": 18725370165120.0, "grad_norm": 4.580700725813826, "language_loss": 0.78256208, "learning_rate": 3.269727197731714e-06, "loss": 0.80440766, "num_input_tokens_seen": 54167360, "step": 2517, "time_per_iteration": 2.620919942855835 }, { "auxiliary_loss_clip": 0.01137055, "auxiliary_loss_mlp": 0.0102724, "balance_loss_clip": 1.05329931, "balance_loss_mlp": 1.01891935, "epoch": 0.30277159862923103, "flos": 22418892031680.0, "grad_norm": 1.6699279637742914, "language_loss": 0.77804768, "learning_rate": 3.269125249064367e-06, "loss": 0.79969066, "num_input_tokens_seen": 54187055, "step": 2518, "time_per_iteration": 2.683006763458252 }, { "auxiliary_loss_clip": 0.01203818, "auxiliary_loss_mlp": 0.01029388, "balance_loss_clip": 1.0592649, "balance_loss_mlp": 1.02114463, "epoch": 0.30289184151987014, "flos": 22273130644320.0, "grad_norm": 1.725187143831148, "language_loss": 0.83280975, "learning_rate": 3.2685231078717297e-06, "loss": 0.85514182, "num_input_tokens_seen": 54207245, "step": 2519, "time_per_iteration": 3.3937900066375732 }, { "auxiliary_loss_clip": 0.01148268, "auxiliary_loss_mlp": 0.00762922, "balance_loss_clip": 1.052966, "balance_loss_mlp": 1.00008678, "epoch": 0.30301208441050925, "flos": 25225597890720.0, "grad_norm": 2.22052399397047, "language_loss": 0.7522105, "learning_rate": 3.267920774245145e-06, "loss": 0.77132243, "num_input_tokens_seen": 54226650, "step": 2520, "time_per_iteration": 2.652005672454834 }, { "auxiliary_loss_clip": 0.01189662, "auxiliary_loss_mlp": 0.0102803, "balance_loss_clip": 1.05880332, "balance_loss_mlp": 1.01999545, "epoch": 0.3031323273011483, "flos": 23039248914240.0, "grad_norm": 1.6756471053183037, "language_loss": 0.8444761, "learning_rate": 3.2673182482759876e-06, "loss": 0.86665303, "num_input_tokens_seen": 54245765, "step": 2521, "time_per_iteration": 2.5819575786590576 }, { "auxiliary_loss_clip": 0.01186898, "auxiliary_loss_mlp": 0.0103045, "balance_loss_clip": 1.05787015, "balance_loss_mlp": 1.02249312, "epoch": 0.3032525701917874, "flos": 18876698681280.0, "grad_norm": 2.7093912245676086, "language_loss": 0.6608634, "learning_rate": 3.266715530055659e-06, "loss": 0.68303686, "num_input_tokens_seen": 54263915, "step": 2522, "time_per_iteration": 2.5556724071502686 }, { "auxiliary_loss_clip": 0.01177033, "auxiliary_loss_mlp": 0.01027611, "balance_loss_clip": 1.05389857, "balance_loss_mlp": 1.01933813, "epoch": 0.30337281308242653, "flos": 17782644227520.0, "grad_norm": 1.603728681793266, "language_loss": 0.8022815, "learning_rate": 3.2661126196755927e-06, "loss": 0.82432795, "num_input_tokens_seen": 54283025, "step": 2523, "time_per_iteration": 2.582721710205078 }, { "auxiliary_loss_clip": 0.01085379, "auxiliary_loss_mlp": 0.01001175, "balance_loss_clip": 1.01027465, "balance_loss_mlp": 0.99964947, "epoch": 0.3034930559730656, "flos": 57824310122400.0, "grad_norm": 0.7978783772293451, "language_loss": 0.56039882, "learning_rate": 3.265509517227248e-06, "loss": 0.58126432, "num_input_tokens_seen": 54339840, "step": 2524, "time_per_iteration": 3.101917028427124 }, { "auxiliary_loss_clip": 0.01171286, "auxiliary_loss_mlp": 0.01034035, "balance_loss_clip": 1.05290663, "balance_loss_mlp": 1.02553535, "epoch": 0.3036132988637047, "flos": 14755590231840.0, "grad_norm": 1.9176935079615967, "language_loss": 0.80897593, "learning_rate": 3.264906222802115e-06, "loss": 0.83102912, "num_input_tokens_seen": 54357690, "step": 2525, "time_per_iteration": 2.6128299236297607 }, { "auxiliary_loss_clip": 0.01201821, "auxiliary_loss_mlp": 0.01025292, "balance_loss_clip": 1.05739248, "balance_loss_mlp": 1.01670265, "epoch": 0.30373354175434375, "flos": 21033207052320.0, "grad_norm": 2.9693555174613873, "language_loss": 0.78266156, "learning_rate": 3.264302736491715e-06, "loss": 0.80493271, "num_input_tokens_seen": 54377810, "step": 2526, "time_per_iteration": 2.5528643131256104 }, { "auxiliary_loss_clip": 0.01181868, "auxiliary_loss_mlp": 0.01033772, "balance_loss_clip": 1.05683398, "balance_loss_mlp": 1.0257194, "epoch": 0.30385378464498286, "flos": 21143237452800.0, "grad_norm": 1.820218078794917, "language_loss": 0.87261522, "learning_rate": 3.263699058387594e-06, "loss": 0.89477164, "num_input_tokens_seen": 54395245, "step": 2527, "time_per_iteration": 2.587939500808716 }, { "auxiliary_loss_clip": 0.0115257, "auxiliary_loss_mlp": 0.01029183, "balance_loss_clip": 1.04915643, "balance_loss_mlp": 1.02089214, "epoch": 0.30397402753562197, "flos": 20629247440800.0, "grad_norm": 2.3817248877491073, "language_loss": 0.90276432, "learning_rate": 3.2630951885813315e-06, "loss": 0.92458177, "num_input_tokens_seen": 54412640, "step": 2528, "time_per_iteration": 2.6118900775909424 }, { "auxiliary_loss_clip": 0.01171658, "auxiliary_loss_mlp": 0.01023399, "balance_loss_clip": 1.05293977, "balance_loss_mlp": 1.01560271, "epoch": 0.304094270426261, "flos": 15085681433280.0, "grad_norm": 1.996917431499322, "language_loss": 0.78243303, "learning_rate": 3.262491127164533e-06, "loss": 0.80438364, "num_input_tokens_seen": 54431455, "step": 2529, "time_per_iteration": 2.5969552993774414 }, { "auxiliary_loss_clip": 0.01179417, "auxiliary_loss_mlp": 0.00763401, "balance_loss_clip": 1.05596733, "balance_loss_mlp": 1.00006366, "epoch": 0.30421451331690014, "flos": 13845219086880.0, "grad_norm": 2.2568526965345406, "language_loss": 0.79759455, "learning_rate": 3.2618868742288337e-06, "loss": 0.81702268, "num_input_tokens_seen": 54448380, "step": 2530, "time_per_iteration": 2.583705425262451 }, { "auxiliary_loss_clip": 0.01186744, "auxiliary_loss_mlp": 0.01028094, "balance_loss_clip": 1.05736625, "balance_loss_mlp": 1.01990461, "epoch": 0.30433475620753925, "flos": 17384215827840.0, "grad_norm": 1.7894377089914324, "language_loss": 0.72019339, "learning_rate": 3.261282429865899e-06, "loss": 0.74234176, "num_input_tokens_seen": 54466385, "step": 2531, "time_per_iteration": 2.5984785556793213 }, { "auxiliary_loss_clip": 0.01175065, "auxiliary_loss_mlp": 0.00762698, "balance_loss_clip": 1.05456221, "balance_loss_mlp": 1.00005889, "epoch": 0.3044549990981783, "flos": 18916955205120.0, "grad_norm": 2.625216074080123, "language_loss": 0.72508019, "learning_rate": 3.2606777941674225e-06, "loss": 0.74445784, "num_input_tokens_seen": 54485040, "step": 2532, "time_per_iteration": 2.591066598892212 }, { "auxiliary_loss_clip": 0.01134397, "auxiliary_loss_mlp": 0.01025974, "balance_loss_clip": 1.05195427, "balance_loss_mlp": 1.0183568, "epoch": 0.3045752419888174, "flos": 21068435201760.0, "grad_norm": 2.456634905197029, "language_loss": 0.84854424, "learning_rate": 3.2600729672251276e-06, "loss": 0.87014794, "num_input_tokens_seen": 54502755, "step": 2533, "time_per_iteration": 2.6543233394622803 }, { "auxiliary_loss_clip": 0.01201774, "auxiliary_loss_mlp": 0.00762765, "balance_loss_clip": 1.05885553, "balance_loss_mlp": 1.00006211, "epoch": 0.3046954848794565, "flos": 29096409847200.0, "grad_norm": 1.9458282292264188, "language_loss": 0.65460384, "learning_rate": 3.259467949130765e-06, "loss": 0.67424929, "num_input_tokens_seen": 54524165, "step": 2534, "time_per_iteration": 2.6062631607055664 }, { "auxiliary_loss_clip": 0.0117296, "auxiliary_loss_mlp": 0.01033328, "balance_loss_clip": 1.05501962, "balance_loss_mlp": 1.0251565, "epoch": 0.3048157277700956, "flos": 20295349041600.0, "grad_norm": 2.171380461113502, "language_loss": 0.82683325, "learning_rate": 3.2588627399761164e-06, "loss": 0.84889615, "num_input_tokens_seen": 54540160, "step": 2535, "time_per_iteration": 2.6010050773620605 }, { "auxiliary_loss_clip": 0.01170034, "auxiliary_loss_mlp": 0.01028627, "balance_loss_clip": 1.05389714, "balance_loss_mlp": 1.02081275, "epoch": 0.3049359706607347, "flos": 22739932159200.0, "grad_norm": 1.899923666586202, "language_loss": 0.70654929, "learning_rate": 3.2582573398529903e-06, "loss": 0.72853589, "num_input_tokens_seen": 54557515, "step": 2536, "time_per_iteration": 3.3513476848602295 }, { "auxiliary_loss_clip": 0.01157676, "auxiliary_loss_mlp": 0.01037428, "balance_loss_clip": 1.05136049, "balance_loss_mlp": 1.02911365, "epoch": 0.3050562135513738, "flos": 18434637563520.0, "grad_norm": 2.024725827763922, "language_loss": 0.74072552, "learning_rate": 3.2576517488532265e-06, "loss": 0.7626766, "num_input_tokens_seen": 54573865, "step": 2537, "time_per_iteration": 3.3788788318634033 }, { "auxiliary_loss_clip": 0.01186348, "auxiliary_loss_mlp": 0.01025167, "balance_loss_clip": 1.05522418, "balance_loss_mlp": 1.01700079, "epoch": 0.30517645644201286, "flos": 20370330877440.0, "grad_norm": 1.6777673848105499, "language_loss": 0.8748107, "learning_rate": 3.257045967068692e-06, "loss": 0.89692581, "num_input_tokens_seen": 54593120, "step": 2538, "time_per_iteration": 2.600954055786133 }, { "auxiliary_loss_clip": 0.01202168, "auxiliary_loss_mlp": 0.01040729, "balance_loss_clip": 1.05911684, "balance_loss_mlp": 1.03194332, "epoch": 0.30529669933265197, "flos": 21945122626560.0, "grad_norm": 1.6468426534801468, "language_loss": 0.82241023, "learning_rate": 3.2564399945912848e-06, "loss": 0.84483922, "num_input_tokens_seen": 54612910, "step": 2539, "time_per_iteration": 3.394562005996704 }, { "auxiliary_loss_clip": 0.01144759, "auxiliary_loss_mlp": 0.0102989, "balance_loss_clip": 1.04938817, "balance_loss_mlp": 1.02270222, "epoch": 0.305416942223291, "flos": 21835415478720.0, "grad_norm": 2.975416930012105, "language_loss": 0.82206941, "learning_rate": 3.2558338315129287e-06, "loss": 0.84381592, "num_input_tokens_seen": 54631055, "step": 2540, "time_per_iteration": 2.6685001850128174 }, { "auxiliary_loss_clip": 0.01179008, "auxiliary_loss_mlp": 0.01029546, "balance_loss_clip": 1.05514359, "balance_loss_mlp": 1.02098656, "epoch": 0.30553718511393013, "flos": 33911815423200.0, "grad_norm": 1.8218026542236112, "language_loss": 0.75822604, "learning_rate": 3.2552274779255785e-06, "loss": 0.78031152, "num_input_tokens_seen": 54651985, "step": 2541, "time_per_iteration": 2.6634621620178223 }, { "auxiliary_loss_clip": 0.01187191, "auxiliary_loss_mlp": 0.01034246, "balance_loss_clip": 1.05944324, "balance_loss_mlp": 1.02545488, "epoch": 0.30565742800456924, "flos": 22268533273440.0, "grad_norm": 1.9995756174281643, "language_loss": 0.77209693, "learning_rate": 3.2546209339212184e-06, "loss": 0.79431129, "num_input_tokens_seen": 54671005, "step": 2542, "time_per_iteration": 2.6107940673828125 }, { "auxiliary_loss_clip": 0.01173171, "auxiliary_loss_mlp": 0.01029145, "balance_loss_clip": 1.05175614, "balance_loss_mlp": 1.02086639, "epoch": 0.3057776708952083, "flos": 22565048505600.0, "grad_norm": 1.6140347502308923, "language_loss": 0.77494919, "learning_rate": 3.25401419959186e-06, "loss": 0.79697227, "num_input_tokens_seen": 54691615, "step": 2543, "time_per_iteration": 2.6466965675354004 }, { "auxiliary_loss_clip": 0.01184224, "auxiliary_loss_mlp": 0.01036123, "balance_loss_clip": 1.06007636, "balance_loss_mlp": 1.02799296, "epoch": 0.3058979137858474, "flos": 21799217571360.0, "grad_norm": 2.1557772817586796, "language_loss": 0.76200032, "learning_rate": 3.253407275029545e-06, "loss": 0.78420377, "num_input_tokens_seen": 54710520, "step": 2544, "time_per_iteration": 2.6228411197662354 }, { "auxiliary_loss_clip": 0.0116295, "auxiliary_loss_mlp": 0.01037063, "balance_loss_clip": 1.05733156, "balance_loss_mlp": 1.02844393, "epoch": 0.3060181566764865, "flos": 26979439660800.0, "grad_norm": 2.072772316554172, "language_loss": 0.80586457, "learning_rate": 3.2528001603263425e-06, "loss": 0.82786465, "num_input_tokens_seen": 54732590, "step": 2545, "time_per_iteration": 3.58266282081604 }, { "auxiliary_loss_clip": 0.01187213, "auxiliary_loss_mlp": 0.01030618, "balance_loss_clip": 1.05920708, "balance_loss_mlp": 1.02208281, "epoch": 0.3061383995671256, "flos": 19865104603680.0, "grad_norm": 1.7102663260321762, "language_loss": 0.81564867, "learning_rate": 3.2521928555743514e-06, "loss": 0.83782697, "num_input_tokens_seen": 54749935, "step": 2546, "time_per_iteration": 2.550290107727051 }, { "auxiliary_loss_clip": 0.01164731, "auxiliary_loss_mlp": 0.00763448, "balance_loss_clip": 1.05256104, "balance_loss_mlp": 1.00007474, "epoch": 0.3062586424577647, "flos": 22127512924800.0, "grad_norm": 2.2171227742233084, "language_loss": 0.67866981, "learning_rate": 3.2515853608657e-06, "loss": 0.69795156, "num_input_tokens_seen": 54767935, "step": 2547, "time_per_iteration": 2.6244900226593018 }, { "auxiliary_loss_clip": 0.01181654, "auxiliary_loss_mlp": 0.01030708, "balance_loss_clip": 1.05521798, "balance_loss_mlp": 1.02206481, "epoch": 0.3063788853484038, "flos": 20845501044000.0, "grad_norm": 2.1992183513195007, "language_loss": 0.75194836, "learning_rate": 3.250977676292545e-06, "loss": 0.77407193, "num_input_tokens_seen": 54786175, "step": 2548, "time_per_iteration": 2.5700840950012207 }, { "auxiliary_loss_clip": 0.01178016, "auxiliary_loss_mlp": 0.01031891, "balance_loss_clip": 1.05623627, "balance_loss_mlp": 1.02393365, "epoch": 0.30649912823904285, "flos": 16209720160320.0, "grad_norm": 2.1288003452316264, "language_loss": 0.79498065, "learning_rate": 3.2503698019470712e-06, "loss": 0.81707978, "num_input_tokens_seen": 54801945, "step": 2549, "time_per_iteration": 2.62515926361084 }, { "auxiliary_loss_clip": 0.01184586, "auxiliary_loss_mlp": 0.01028635, "balance_loss_clip": 1.05407465, "balance_loss_mlp": 1.02045715, "epoch": 0.30661937112968196, "flos": 18617818034880.0, "grad_norm": 2.710502916677012, "language_loss": 0.78460807, "learning_rate": 3.249761737921492e-06, "loss": 0.80674028, "num_input_tokens_seen": 54818475, "step": 2550, "time_per_iteration": 2.5573627948760986 }, { "auxiliary_loss_clip": 0.0116831, "auxiliary_loss_mlp": 0.01026855, "balance_loss_clip": 1.05617762, "balance_loss_mlp": 1.01815236, "epoch": 0.30673961402032107, "flos": 31390813791360.0, "grad_norm": 2.2754174494744723, "language_loss": 0.74373364, "learning_rate": 3.249153484308051e-06, "loss": 0.76568532, "num_input_tokens_seen": 54837090, "step": 2551, "time_per_iteration": 2.686199903488159 }, { "auxiliary_loss_clip": 0.01132662, "auxiliary_loss_mlp": 0.01034489, "balance_loss_clip": 1.05062497, "balance_loss_mlp": 1.02625155, "epoch": 0.3068598569109601, "flos": 20229813366240.0, "grad_norm": 2.1326455249969745, "language_loss": 0.77824509, "learning_rate": 3.2485450411990194e-06, "loss": 0.79991663, "num_input_tokens_seen": 54856445, "step": 2552, "time_per_iteration": 2.668086051940918 }, { "auxiliary_loss_clip": 0.01200792, "auxiliary_loss_mlp": 0.01032347, "balance_loss_clip": 1.05641139, "balance_loss_mlp": 1.02415776, "epoch": 0.30698009980159924, "flos": 29601995290560.0, "grad_norm": 1.9316244440876418, "language_loss": 0.82265753, "learning_rate": 3.2479364086866983e-06, "loss": 0.84498894, "num_input_tokens_seen": 54876700, "step": 2553, "time_per_iteration": 2.602349281311035 }, { "auxiliary_loss_clip": 0.01172977, "auxiliary_loss_mlp": 0.007635, "balance_loss_clip": 1.05701125, "balance_loss_mlp": 1.00017524, "epoch": 0.30710034269223835, "flos": 23842427098560.0, "grad_norm": 1.6983520263477754, "language_loss": 0.81347448, "learning_rate": 3.247327586863416e-06, "loss": 0.83283925, "num_input_tokens_seen": 54897580, "step": 2554, "time_per_iteration": 2.62882661819458 }, { "auxiliary_loss_clip": 0.01166246, "auxiliary_loss_mlp": 0.0102344, "balance_loss_clip": 1.05552566, "balance_loss_mlp": 1.01489234, "epoch": 0.3072205855828774, "flos": 25884990120480.0, "grad_norm": 2.0191573420545272, "language_loss": 0.76793295, "learning_rate": 3.2467185758215304e-06, "loss": 0.78982985, "num_input_tokens_seen": 54917320, "step": 2555, "time_per_iteration": 2.6613595485687256 }, { "auxiliary_loss_clip": 0.01162234, "auxiliary_loss_mlp": 0.00762866, "balance_loss_clip": 1.0556668, "balance_loss_mlp": 1.00013852, "epoch": 0.3073408284735165, "flos": 22236393982560.0, "grad_norm": 3.1559735701389195, "language_loss": 0.85638082, "learning_rate": 3.246109375653428e-06, "loss": 0.87563181, "num_input_tokens_seen": 54934085, "step": 2556, "time_per_iteration": 2.621777057647705 }, { "auxiliary_loss_clip": 0.01202644, "auxiliary_loss_mlp": 0.01029343, "balance_loss_clip": 1.05966246, "balance_loss_mlp": 1.02108145, "epoch": 0.30746107136415557, "flos": 19500288090240.0, "grad_norm": 1.6963901111012791, "language_loss": 0.78057897, "learning_rate": 3.2454999864515243e-06, "loss": 0.80289888, "num_input_tokens_seen": 54953460, "step": 2557, "time_per_iteration": 2.5539345741271973 }, { "auxiliary_loss_clip": 0.01168629, "auxiliary_loss_mlp": 0.00763749, "balance_loss_clip": 1.05603123, "balance_loss_mlp": 1.00012398, "epoch": 0.3075813142547947, "flos": 21724810406880.0, "grad_norm": 1.6945258497606404, "language_loss": 0.69467759, "learning_rate": 3.244890408308263e-06, "loss": 0.71400136, "num_input_tokens_seen": 54974165, "step": 2558, "time_per_iteration": 2.593709707260132 }, { "auxiliary_loss_clip": 0.01145733, "auxiliary_loss_mlp": 0.01026472, "balance_loss_clip": 1.05206895, "balance_loss_mlp": 1.01835346, "epoch": 0.3077015571454338, "flos": 24097967467680.0, "grad_norm": 2.112929902808662, "language_loss": 0.61700356, "learning_rate": 3.2442806413161165e-06, "loss": 0.63872564, "num_input_tokens_seen": 54993810, "step": 2559, "time_per_iteration": 2.707615375518799 }, { "auxiliary_loss_clip": 0.01149102, "auxiliary_loss_mlp": 0.01034877, "balance_loss_clip": 1.0547086, "balance_loss_mlp": 1.02661633, "epoch": 0.30782180003607285, "flos": 18405479380320.0, "grad_norm": 2.0624712217335355, "language_loss": 0.75792456, "learning_rate": 3.243670685567586e-06, "loss": 0.77976429, "num_input_tokens_seen": 55011210, "step": 2560, "time_per_iteration": 2.619453191757202 }, { "auxiliary_loss_clip": 0.01172401, "auxiliary_loss_mlp": 0.0076303, "balance_loss_clip": 1.05521214, "balance_loss_mlp": 1.0001601, "epoch": 0.30794204292671196, "flos": 23878553172000.0, "grad_norm": 2.359013109777886, "language_loss": 0.803388, "learning_rate": 3.2430605411552012e-06, "loss": 0.82274234, "num_input_tokens_seen": 55031325, "step": 2561, "time_per_iteration": 2.5944035053253174 }, { "auxiliary_loss_clip": 0.01063026, "auxiliary_loss_mlp": 0.01010221, "balance_loss_clip": 1.01377678, "balance_loss_mlp": 1.00863576, "epoch": 0.30806228581735107, "flos": 67927813170720.0, "grad_norm": 0.8925448892870776, "language_loss": 0.70591879, "learning_rate": 3.2424502081715205e-06, "loss": 0.72665119, "num_input_tokens_seen": 55094440, "step": 2562, "time_per_iteration": 3.9564945697784424 }, { "auxiliary_loss_clip": 0.01173573, "auxiliary_loss_mlp": 0.01033865, "balance_loss_clip": 1.05448484, "balance_loss_mlp": 1.02528191, "epoch": 0.3081825287079901, "flos": 23843217271680.0, "grad_norm": 2.7249217985344125, "language_loss": 0.78042805, "learning_rate": 3.241839686709132e-06, "loss": 0.80250245, "num_input_tokens_seen": 55115375, "step": 2563, "time_per_iteration": 3.429056406021118 }, { "auxiliary_loss_clip": 0.01182528, "auxiliary_loss_mlp": 0.01032909, "balance_loss_clip": 1.05260837, "balance_loss_mlp": 1.02422428, "epoch": 0.30830277159862923, "flos": 16209971579040.0, "grad_norm": 2.4592221665277094, "language_loss": 0.81991327, "learning_rate": 3.2412289768606495e-06, "loss": 0.8420676, "num_input_tokens_seen": 55131945, "step": 2564, "time_per_iteration": 2.5608699321746826 }, { "auxiliary_loss_clip": 0.01190205, "auxiliary_loss_mlp": 0.0103285, "balance_loss_clip": 1.0569433, "balance_loss_mlp": 1.02463067, "epoch": 0.30842301448926834, "flos": 29349507863040.0, "grad_norm": 1.7713944018675103, "language_loss": 0.82540274, "learning_rate": 3.240618078718718e-06, "loss": 0.84763336, "num_input_tokens_seen": 55153405, "step": 2565, "time_per_iteration": 3.5042524337768555 }, { "auxiliary_loss_clip": 0.01155582, "auxiliary_loss_mlp": 0.01033159, "balance_loss_clip": 1.05007267, "balance_loss_mlp": 1.0250411, "epoch": 0.3085432573799074, "flos": 21945194460480.0, "grad_norm": 1.8896418477249903, "language_loss": 0.74129945, "learning_rate": 3.240006992376011e-06, "loss": 0.76318687, "num_input_tokens_seen": 55173030, "step": 2566, "time_per_iteration": 2.629451274871826 }, { "auxiliary_loss_clip": 0.0117812, "auxiliary_loss_mlp": 0.0102855, "balance_loss_clip": 1.05651987, "balance_loss_mlp": 1.02060461, "epoch": 0.3086635002705465, "flos": 22054722023520.0, "grad_norm": 2.52185834426975, "language_loss": 0.7590276, "learning_rate": 3.2393957179252284e-06, "loss": 0.78109431, "num_input_tokens_seen": 55189565, "step": 2567, "time_per_iteration": 2.606243371963501 }, { "auxiliary_loss_clip": 0.0120317, "auxiliary_loss_mlp": 0.01030913, "balance_loss_clip": 1.05933976, "balance_loss_mlp": 1.02251494, "epoch": 0.3087837431611856, "flos": 32665929615840.0, "grad_norm": 1.9416769889150904, "language_loss": 0.80532265, "learning_rate": 3.2387842554591016e-06, "loss": 0.82766354, "num_input_tokens_seen": 55210380, "step": 2568, "time_per_iteration": 2.6332011222839355 }, { "auxiliary_loss_clip": 0.01202006, "auxiliary_loss_mlp": 0.01024931, "balance_loss_clip": 1.05810463, "balance_loss_mlp": 1.01646149, "epoch": 0.3089039860518247, "flos": 17599248254400.0, "grad_norm": 2.096416511246613, "language_loss": 0.87691927, "learning_rate": 3.238172605070388e-06, "loss": 0.89918864, "num_input_tokens_seen": 55225795, "step": 2569, "time_per_iteration": 2.5233469009399414 }, { "auxiliary_loss_clip": 0.01185863, "auxiliary_loss_mlp": 0.00763866, "balance_loss_clip": 1.05797577, "balance_loss_mlp": 1.00015402, "epoch": 0.3090242289424638, "flos": 14383841745120.0, "grad_norm": 5.082968858534825, "language_loss": 0.78644794, "learning_rate": 3.2375607668518745e-06, "loss": 0.80594522, "num_input_tokens_seen": 55238830, "step": 2570, "time_per_iteration": 3.427476644515991 }, { "auxiliary_loss_clip": 0.01162173, "auxiliary_loss_mlp": 0.01035321, "balance_loss_clip": 1.0526576, "balance_loss_mlp": 1.0266726, "epoch": 0.30914447183310284, "flos": 16068627977760.0, "grad_norm": 2.260633824896086, "language_loss": 0.89805448, "learning_rate": 3.236948740896377e-06, "loss": 0.92002946, "num_input_tokens_seen": 55253630, "step": 2571, "time_per_iteration": 2.567777633666992 }, { "auxiliary_loss_clip": 0.01187627, "auxiliary_loss_mlp": 0.01033042, "balance_loss_clip": 1.05729008, "balance_loss_mlp": 1.02420235, "epoch": 0.30926471472374195, "flos": 32230225800000.0, "grad_norm": 1.4784201768753034, "language_loss": 0.84046245, "learning_rate": 3.2363365272967384e-06, "loss": 0.86266917, "num_input_tokens_seen": 55276200, "step": 2572, "time_per_iteration": 2.68868088722229 }, { "auxiliary_loss_clip": 0.01185637, "auxiliary_loss_mlp": 0.01031913, "balance_loss_clip": 1.0588479, "balance_loss_mlp": 1.02315748, "epoch": 0.30938495761438106, "flos": 20370725964000.0, "grad_norm": 1.8969844671158713, "language_loss": 0.81577164, "learning_rate": 3.235724126145832e-06, "loss": 0.83794713, "num_input_tokens_seen": 55292235, "step": 2573, "time_per_iteration": 2.5475480556488037 }, { "auxiliary_loss_clip": 0.01177823, "auxiliary_loss_mlp": 0.01031239, "balance_loss_clip": 1.05329514, "balance_loss_mlp": 1.02311456, "epoch": 0.3095052005050201, "flos": 24061158972000.0, "grad_norm": 1.693340806120869, "language_loss": 0.77284443, "learning_rate": 3.235111537536558e-06, "loss": 0.79493505, "num_input_tokens_seen": 55313050, "step": 2574, "time_per_iteration": 2.625175714492798 }, { "auxiliary_loss_clip": 0.01190143, "auxiliary_loss_mlp": 0.01028177, "balance_loss_clip": 1.05804765, "balance_loss_mlp": 1.02033293, "epoch": 0.30962544339565923, "flos": 23401551240480.0, "grad_norm": 1.872391946335908, "language_loss": 0.83061618, "learning_rate": 3.2344987615618456e-06, "loss": 0.85279942, "num_input_tokens_seen": 55332885, "step": 2575, "time_per_iteration": 2.5920214653015137 }, { "auxiliary_loss_clip": 0.01155942, "auxiliary_loss_mlp": 0.01030662, "balance_loss_clip": 1.05382466, "balance_loss_mlp": 1.02237058, "epoch": 0.30974568628629834, "flos": 33799989174720.0, "grad_norm": 1.8267962940671376, "language_loss": 0.78801513, "learning_rate": 3.2338857983146533e-06, "loss": 0.80988121, "num_input_tokens_seen": 55354385, "step": 2576, "time_per_iteration": 2.7449429035186768 }, { "auxiliary_loss_clip": 0.01165868, "auxiliary_loss_mlp": 0.01026107, "balance_loss_clip": 1.05654621, "balance_loss_mlp": 1.01725531, "epoch": 0.3098659291769374, "flos": 20229597864480.0, "grad_norm": 1.9083819875891426, "language_loss": 0.76370239, "learning_rate": 3.233272647887966e-06, "loss": 0.78562224, "num_input_tokens_seen": 55373275, "step": 2577, "time_per_iteration": 2.591115713119507 }, { "auxiliary_loss_clip": 0.01205009, "auxiliary_loss_mlp": 0.01032086, "balance_loss_clip": 1.06010747, "balance_loss_mlp": 1.02365828, "epoch": 0.3099861720675765, "flos": 24748560042240.0, "grad_norm": 1.710872149216556, "language_loss": 0.90167129, "learning_rate": 3.2326593103747985e-06, "loss": 0.92404228, "num_input_tokens_seen": 55392290, "step": 2578, "time_per_iteration": 2.582329034805298 }, { "auxiliary_loss_clip": 0.01189549, "auxiliary_loss_mlp": 0.01028599, "balance_loss_clip": 1.06020284, "balance_loss_mlp": 1.01975965, "epoch": 0.3101064149582156, "flos": 11765488399680.0, "grad_norm": 2.0408397988503597, "language_loss": 0.84901559, "learning_rate": 3.2320457858681936e-06, "loss": 0.87119704, "num_input_tokens_seen": 55410680, "step": 2579, "time_per_iteration": 2.56128191947937 }, { "auxiliary_loss_clip": 0.0117254, "auxiliary_loss_mlp": 0.01030339, "balance_loss_clip": 1.05550098, "balance_loss_mlp": 1.02217865, "epoch": 0.31022665784885467, "flos": 23033250781920.0, "grad_norm": 2.394652218239499, "language_loss": 0.85239613, "learning_rate": 3.2314320744612228e-06, "loss": 0.87442493, "num_input_tokens_seen": 55425980, "step": 2580, "time_per_iteration": 2.6040399074554443 }, { "auxiliary_loss_clip": 0.01187682, "auxiliary_loss_mlp": 0.01030673, "balance_loss_clip": 1.05946016, "balance_loss_mlp": 1.02207208, "epoch": 0.3103469007394938, "flos": 16289191616160.0, "grad_norm": 1.5550810188577822, "language_loss": 0.76628685, "learning_rate": 3.2308181762469854e-06, "loss": 0.78847039, "num_input_tokens_seen": 55443925, "step": 2581, "time_per_iteration": 2.5541348457336426 }, { "auxiliary_loss_clip": 0.0120616, "auxiliary_loss_mlp": 0.01029276, "balance_loss_clip": 1.05894387, "balance_loss_mlp": 1.02028787, "epoch": 0.3104671436301329, "flos": 30515275709280.0, "grad_norm": 2.0383869698189265, "language_loss": 0.78055608, "learning_rate": 3.230204091318609e-06, "loss": 0.80291039, "num_input_tokens_seen": 55464465, "step": 2582, "time_per_iteration": 2.605419635772705 }, { "auxiliary_loss_clip": 0.01200445, "auxiliary_loss_mlp": 0.00762993, "balance_loss_clip": 1.05786407, "balance_loss_mlp": 1.00011337, "epoch": 0.31058738652077195, "flos": 20047243483200.0, "grad_norm": 2.0910997692694226, "language_loss": 0.84604341, "learning_rate": 3.2295898197692503e-06, "loss": 0.86567771, "num_input_tokens_seen": 55483425, "step": 2583, "time_per_iteration": 2.5356833934783936 }, { "auxiliary_loss_clip": 0.01203552, "auxiliary_loss_mlp": 0.01027098, "balance_loss_clip": 1.06109214, "balance_loss_mlp": 1.0191232, "epoch": 0.31070762941141106, "flos": 28074643457280.0, "grad_norm": 1.6937885150647263, "language_loss": 0.79268247, "learning_rate": 3.228975361692094e-06, "loss": 0.81498897, "num_input_tokens_seen": 55504445, "step": 2584, "time_per_iteration": 2.5995919704437256 }, { "auxiliary_loss_clip": 0.01193739, "auxiliary_loss_mlp": 0.00764125, "balance_loss_clip": 1.05769551, "balance_loss_mlp": 1.00017047, "epoch": 0.31082787230205017, "flos": 20521910812320.0, "grad_norm": 2.144408100393888, "language_loss": 0.80325902, "learning_rate": 3.228360717180352e-06, "loss": 0.82283765, "num_input_tokens_seen": 55521970, "step": 2585, "time_per_iteration": 2.578561305999756 }, { "auxiliary_loss_clip": 0.0109125, "auxiliary_loss_mlp": 0.00753775, "balance_loss_clip": 1.01547921, "balance_loss_mlp": 0.99953252, "epoch": 0.3109481151926892, "flos": 62445939723840.0, "grad_norm": 0.8096886090264103, "language_loss": 0.59449303, "learning_rate": 3.227745886327266e-06, "loss": 0.61294317, "num_input_tokens_seen": 55580665, "step": 2586, "time_per_iteration": 3.0567264556884766 }, { "auxiliary_loss_clip": 0.01090807, "auxiliary_loss_mlp": 0.01002059, "balance_loss_clip": 1.01496649, "balance_loss_mlp": 1.00062871, "epoch": 0.31106835808332833, "flos": 44746753734720.0, "grad_norm": 0.8114098282613659, "language_loss": 0.55890584, "learning_rate": 3.227130869226105e-06, "loss": 0.57983458, "num_input_tokens_seen": 55637825, "step": 2587, "time_per_iteration": 3.1268880367279053 }, { "auxiliary_loss_clip": 0.01189903, "auxiliary_loss_mlp": 0.01029883, "balance_loss_clip": 1.05745578, "balance_loss_mlp": 1.02116323, "epoch": 0.3111886009739674, "flos": 23403059752800.0, "grad_norm": 3.3500190606265567, "language_loss": 0.8264848, "learning_rate": 3.226515665970167e-06, "loss": 0.8486827, "num_input_tokens_seen": 55655365, "step": 2588, "time_per_iteration": 3.3286197185516357 }, { "auxiliary_loss_clip": 0.01183897, "auxiliary_loss_mlp": 0.01032765, "balance_loss_clip": 1.05496383, "balance_loss_mlp": 1.02368116, "epoch": 0.3113088438646065, "flos": 17530731471360.0, "grad_norm": 2.071728177555535, "language_loss": 0.86596775, "learning_rate": 3.225900276652777e-06, "loss": 0.88813436, "num_input_tokens_seen": 55672140, "step": 2589, "time_per_iteration": 3.3070473670959473 }, { "auxiliary_loss_clip": 0.01179652, "auxiliary_loss_mlp": 0.01029196, "balance_loss_clip": 1.05507302, "balance_loss_mlp": 1.02092874, "epoch": 0.3114290867552456, "flos": 28365807062400.0, "grad_norm": 1.5255207613406643, "language_loss": 0.7558074, "learning_rate": 3.2252847013672906e-06, "loss": 0.77789581, "num_input_tokens_seen": 55694800, "step": 2590, "time_per_iteration": 2.6504368782043457 }, { "auxiliary_loss_clip": 0.01148696, "auxiliary_loss_mlp": 0.01033651, "balance_loss_clip": 1.05133963, "balance_loss_mlp": 1.02525234, "epoch": 0.31154932964588467, "flos": 27379161071040.0, "grad_norm": 2.080037549421373, "language_loss": 0.75829434, "learning_rate": 3.224668940207089e-06, "loss": 0.78011787, "num_input_tokens_seen": 55713785, "step": 2591, "time_per_iteration": 3.570590019226074 }, { "auxiliary_loss_clip": 0.01132292, "auxiliary_loss_mlp": 0.01031156, "balance_loss_clip": 1.04703939, "balance_loss_mlp": 1.02232218, "epoch": 0.3116695725365238, "flos": 26541868163040.0, "grad_norm": 1.803643859469152, "language_loss": 0.86900163, "learning_rate": 3.2240529932655828e-06, "loss": 0.89063609, "num_input_tokens_seen": 55733050, "step": 2592, "time_per_iteration": 2.7174134254455566 }, { "auxiliary_loss_clip": 0.01173172, "auxiliary_loss_mlp": 0.01028904, "balance_loss_clip": 1.05831122, "balance_loss_mlp": 1.02089345, "epoch": 0.3117898154271629, "flos": 21177603595200.0, "grad_norm": 2.5040725833967645, "language_loss": 0.89052463, "learning_rate": 3.223436860636211e-06, "loss": 0.91254544, "num_input_tokens_seen": 55748685, "step": 2593, "time_per_iteration": 2.584446430206299 }, { "auxiliary_loss_clip": 0.01201687, "auxiliary_loss_mlp": 0.01031787, "balance_loss_clip": 1.05818534, "balance_loss_mlp": 1.0236032, "epoch": 0.31191005831780194, "flos": 27272435030880.0, "grad_norm": 2.148441786623884, "language_loss": 0.7419287, "learning_rate": 3.2228205424124403e-06, "loss": 0.76426351, "num_input_tokens_seen": 55771840, "step": 2594, "time_per_iteration": 2.6377439498901367 }, { "auxiliary_loss_clip": 0.01158685, "auxiliary_loss_mlp": 0.01032126, "balance_loss_clip": 1.05089533, "balance_loss_mlp": 1.02317357, "epoch": 0.31203030120844105, "flos": 12963503287680.0, "grad_norm": 2.304063983722526, "language_loss": 0.74860358, "learning_rate": 3.222204038687765e-06, "loss": 0.77051169, "num_input_tokens_seen": 55784975, "step": 2595, "time_per_iteration": 2.5809686183929443 }, { "auxiliary_loss_clip": 0.01184577, "auxiliary_loss_mlp": 0.01034161, "balance_loss_clip": 1.05829215, "balance_loss_mlp": 1.02608407, "epoch": 0.31215054409908016, "flos": 27562018289760.0, "grad_norm": 1.6262145847428455, "language_loss": 0.88353837, "learning_rate": 3.221587349555709e-06, "loss": 0.90572572, "num_input_tokens_seen": 55805235, "step": 2596, "time_per_iteration": 2.618534803390503 }, { "auxiliary_loss_clip": 0.0117369, "auxiliary_loss_mlp": 0.0103212, "balance_loss_clip": 1.05377233, "balance_loss_mlp": 1.02385235, "epoch": 0.3122707869897192, "flos": 21506329952160.0, "grad_norm": 1.6912529013131201, "language_loss": 0.69304514, "learning_rate": 3.2209704751098236e-06, "loss": 0.71510327, "num_input_tokens_seen": 55824265, "step": 2597, "time_per_iteration": 3.5185484886169434 }, { "auxiliary_loss_clip": 0.01179549, "auxiliary_loss_mlp": 0.01026539, "balance_loss_clip": 1.05743337, "balance_loss_mlp": 1.01837325, "epoch": 0.31239102988035833, "flos": 15187019929440.0, "grad_norm": 2.0240924517576078, "language_loss": 0.82725126, "learning_rate": 3.2203534154436875e-06, "loss": 0.84931213, "num_input_tokens_seen": 55838620, "step": 2598, "time_per_iteration": 2.577636241912842 }, { "auxiliary_loss_clip": 0.01126648, "auxiliary_loss_mlp": 0.01037302, "balance_loss_clip": 1.05044997, "balance_loss_mlp": 1.02874279, "epoch": 0.31251127277099744, "flos": 22053716348640.0, "grad_norm": 2.112106822659779, "language_loss": 0.75515378, "learning_rate": 3.2197361706509084e-06, "loss": 0.7767933, "num_input_tokens_seen": 55859375, "step": 2599, "time_per_iteration": 2.7175445556640625 }, { "auxiliary_loss_clip": 0.01206353, "auxiliary_loss_mlp": 0.01026634, "balance_loss_clip": 1.0599587, "balance_loss_mlp": 1.01759803, "epoch": 0.3126315156616365, "flos": 15193987819680.0, "grad_norm": 2.2617566331251036, "language_loss": 0.82925797, "learning_rate": 3.2191187408251228e-06, "loss": 0.85158789, "num_input_tokens_seen": 55876535, "step": 2600, "time_per_iteration": 2.520285129547119 }, { "auxiliary_loss_clip": 0.01193374, "auxiliary_loss_mlp": 0.01031408, "balance_loss_clip": 1.05675364, "balance_loss_mlp": 1.02210331, "epoch": 0.3127517585522756, "flos": 18145341640320.0, "grad_norm": 2.2857935837469188, "language_loss": 0.78567851, "learning_rate": 3.218501126059993e-06, "loss": 0.80792624, "num_input_tokens_seen": 55891930, "step": 2601, "time_per_iteration": 2.5555994510650635 }, { "auxiliary_loss_clip": 0.01186406, "auxiliary_loss_mlp": 0.01029064, "balance_loss_clip": 1.05291295, "balance_loss_mlp": 1.02029645, "epoch": 0.31287200144291466, "flos": 21908637383520.0, "grad_norm": 1.8910007800474127, "language_loss": 0.81458873, "learning_rate": 3.2178833264492116e-06, "loss": 0.83674347, "num_input_tokens_seen": 55910635, "step": 2602, "time_per_iteration": 2.579775810241699 }, { "auxiliary_loss_clip": 0.01195381, "auxiliary_loss_mlp": 0.01031011, "balance_loss_clip": 1.05887043, "balance_loss_mlp": 1.02246356, "epoch": 0.31299224433355377, "flos": 29897002010400.0, "grad_norm": 1.9631561702781566, "language_loss": 0.75936902, "learning_rate": 3.217265342086498e-06, "loss": 0.78163296, "num_input_tokens_seen": 55931125, "step": 2603, "time_per_iteration": 2.6516997814178467 }, { "auxiliary_loss_clip": 0.01162193, "auxiliary_loss_mlp": 0.00764059, "balance_loss_clip": 1.0544405, "balance_loss_mlp": 1.00014603, "epoch": 0.3131124872241929, "flos": 11655889002720.0, "grad_norm": 2.349726458819223, "language_loss": 0.73214579, "learning_rate": 3.216647173065599e-06, "loss": 0.75140828, "num_input_tokens_seen": 55946590, "step": 2604, "time_per_iteration": 2.610199213027954 }, { "auxiliary_loss_clip": 0.01172014, "auxiliary_loss_mlp": 0.01028151, "balance_loss_clip": 1.05870187, "balance_loss_mlp": 1.02024698, "epoch": 0.31323273011483194, "flos": 49848790990560.0, "grad_norm": 2.075686072201073, "language_loss": 0.73657095, "learning_rate": 3.216028819480292e-06, "loss": 0.75857264, "num_input_tokens_seen": 55967930, "step": 2605, "time_per_iteration": 2.861504554748535 }, { "auxiliary_loss_clip": 0.01161479, "auxiliary_loss_mlp": 0.01030242, "balance_loss_clip": 1.0540812, "balance_loss_mlp": 1.02185547, "epoch": 0.31335297300547105, "flos": 22601282329920.0, "grad_norm": 1.9881671147447162, "language_loss": 0.75341707, "learning_rate": 3.2154102814243793e-06, "loss": 0.77533424, "num_input_tokens_seen": 55987070, "step": 2606, "time_per_iteration": 2.604381799697876 }, { "auxiliary_loss_clip": 0.01165196, "auxiliary_loss_mlp": 0.01031226, "balance_loss_clip": 1.05712605, "balance_loss_mlp": 1.02270901, "epoch": 0.31347321589611016, "flos": 34710863157120.0, "grad_norm": 1.972601402942347, "language_loss": 0.66550189, "learning_rate": 3.2147915589916937e-06, "loss": 0.68746608, "num_input_tokens_seen": 56008630, "step": 2607, "time_per_iteration": 2.755615234375 }, { "auxiliary_loss_clip": 0.01167559, "auxiliary_loss_mlp": 0.01034293, "balance_loss_clip": 1.05260527, "balance_loss_mlp": 1.0258944, "epoch": 0.3135934587867492, "flos": 19755792542400.0, "grad_norm": 1.8065496863435495, "language_loss": 0.82470888, "learning_rate": 3.2141726522760938e-06, "loss": 0.84672737, "num_input_tokens_seen": 56026690, "step": 2608, "time_per_iteration": 2.58309006690979 }, { "auxiliary_loss_clip": 0.01075873, "auxiliary_loss_mlp": 0.01005603, "balance_loss_clip": 1.01490831, "balance_loss_mlp": 1.00405371, "epoch": 0.3137137016773883, "flos": 65815548106080.0, "grad_norm": 0.7077921168879837, "language_loss": 0.52635837, "learning_rate": 3.213553561371469e-06, "loss": 0.54717314, "num_input_tokens_seen": 56090425, "step": 2609, "time_per_iteration": 3.232736349105835 }, { "auxiliary_loss_clip": 0.01144406, "auxiliary_loss_mlp": 0.0103304, "balance_loss_clip": 1.05563545, "balance_loss_mlp": 1.02462959, "epoch": 0.31383394456802743, "flos": 16252742290080.0, "grad_norm": 2.8327979021160643, "language_loss": 0.95730126, "learning_rate": 3.212934286371733e-06, "loss": 0.97907573, "num_input_tokens_seen": 56107135, "step": 2610, "time_per_iteration": 2.6552233695983887 }, { "auxiliary_loss_clip": 0.01186919, "auxiliary_loss_mlp": 0.01029539, "balance_loss_clip": 1.05888963, "balance_loss_mlp": 1.02035356, "epoch": 0.3139541874586665, "flos": 38795522280480.0, "grad_norm": 2.274690527793897, "language_loss": 0.83041346, "learning_rate": 3.2123148273708304e-06, "loss": 0.85257816, "num_input_tokens_seen": 56127325, "step": 2611, "time_per_iteration": 2.7147488594055176 }, { "auxiliary_loss_clip": 0.01199337, "auxiliary_loss_mlp": 0.01037078, "balance_loss_clip": 1.05749345, "balance_loss_mlp": 1.02842927, "epoch": 0.3140744303493056, "flos": 25046332368000.0, "grad_norm": 2.373103860600147, "language_loss": 0.76409042, "learning_rate": 3.211695184462733e-06, "loss": 0.78645456, "num_input_tokens_seen": 56148500, "step": 2612, "time_per_iteration": 2.5887844562530518 }, { "auxiliary_loss_clip": 0.01052484, "auxiliary_loss_mlp": 0.01006088, "balance_loss_clip": 1.01276469, "balance_loss_mlp": 1.00446653, "epoch": 0.3141946732399447, "flos": 72504265624800.0, "grad_norm": 1.0182738279246817, "language_loss": 0.60463393, "learning_rate": 3.2110753577414383e-06, "loss": 0.62521964, "num_input_tokens_seen": 56210080, "step": 2613, "time_per_iteration": 3.186220645904541 }, { "auxiliary_loss_clip": 0.01173293, "auxiliary_loss_mlp": 0.01030084, "balance_loss_clip": 1.05327237, "balance_loss_mlp": 1.02144706, "epoch": 0.31431491613058377, "flos": 19239791180640.0, "grad_norm": 1.752502668431045, "language_loss": 0.78730482, "learning_rate": 3.2104553473009757e-06, "loss": 0.80933863, "num_input_tokens_seen": 56228200, "step": 2614, "time_per_iteration": 2.5951054096221924 }, { "auxiliary_loss_clip": 0.01139166, "auxiliary_loss_mlp": 0.01035419, "balance_loss_clip": 1.0526278, "balance_loss_mlp": 1.02699065, "epoch": 0.3144351590212229, "flos": 36210601236480.0, "grad_norm": 1.7952178995976713, "language_loss": 0.67325294, "learning_rate": 3.209835153235399e-06, "loss": 0.69499874, "num_input_tokens_seen": 56249755, "step": 2615, "time_per_iteration": 4.295687437057495 }, { "auxiliary_loss_clip": 0.01146779, "auxiliary_loss_mlp": 0.01033272, "balance_loss_clip": 1.05086768, "balance_loss_mlp": 1.02520156, "epoch": 0.314555401911862, "flos": 18551743605120.0, "grad_norm": 1.6744923533556098, "language_loss": 0.67600709, "learning_rate": 3.2092147756387916e-06, "loss": 0.69780761, "num_input_tokens_seen": 56270080, "step": 2616, "time_per_iteration": 2.6477701663970947 }, { "auxiliary_loss_clip": 0.01164538, "auxiliary_loss_mlp": 0.01030923, "balance_loss_clip": 1.05302167, "balance_loss_mlp": 1.02239335, "epoch": 0.31467564480250104, "flos": 16362880441440.0, "grad_norm": 3.288937818811615, "language_loss": 0.83495796, "learning_rate": 3.208594214605264e-06, "loss": 0.85691255, "num_input_tokens_seen": 56288625, "step": 2617, "time_per_iteration": 3.500898838043213 }, { "auxiliary_loss_clip": 0.01159726, "auxiliary_loss_mlp": 0.01027343, "balance_loss_clip": 1.05253863, "balance_loss_mlp": 1.01903975, "epoch": 0.31479588769314015, "flos": 21652378675200.0, "grad_norm": 1.9384858453373892, "language_loss": 0.77192652, "learning_rate": 3.2079734702289553e-06, "loss": 0.79379714, "num_input_tokens_seen": 56307520, "step": 2618, "time_per_iteration": 2.591216802597046 }, { "auxiliary_loss_clip": 0.01069991, "auxiliary_loss_mlp": 0.0075362, "balance_loss_clip": 1.01201344, "balance_loss_mlp": 0.99954528, "epoch": 0.3149161305837792, "flos": 66051082728480.0, "grad_norm": 0.8157051519832942, "language_loss": 0.60404932, "learning_rate": 3.207352542604031e-06, "loss": 0.62228549, "num_input_tokens_seen": 56369855, "step": 2619, "time_per_iteration": 3.2488386631011963 }, { "auxiliary_loss_clip": 0.01146975, "auxiliary_loss_mlp": 0.010254, "balance_loss_clip": 1.05106902, "balance_loss_mlp": 1.0176096, "epoch": 0.3150363734744183, "flos": 28987205536800.0, "grad_norm": 1.583316164006963, "language_loss": 0.78375787, "learning_rate": 3.2067314318246864e-06, "loss": 0.80548161, "num_input_tokens_seen": 56390570, "step": 2620, "time_per_iteration": 2.7028560638427734 }, { "auxiliary_loss_clip": 0.01159042, "auxiliary_loss_mlp": 0.01031859, "balance_loss_clip": 1.05597687, "balance_loss_mlp": 1.0237205, "epoch": 0.31515661636505743, "flos": 27636605039040.0, "grad_norm": 1.690792189377456, "language_loss": 0.776344, "learning_rate": 3.206110137985143e-06, "loss": 0.79825306, "num_input_tokens_seen": 56410775, "step": 2621, "time_per_iteration": 2.6732780933380127 }, { "auxiliary_loss_clip": 0.01147224, "auxiliary_loss_mlp": 0.01034091, "balance_loss_clip": 1.05302787, "balance_loss_mlp": 1.02548456, "epoch": 0.3152768592556965, "flos": 24605636094720.0, "grad_norm": 1.7340716418235516, "language_loss": 0.92083132, "learning_rate": 3.2054886611796505e-06, "loss": 0.94264448, "num_input_tokens_seen": 56429770, "step": 2622, "time_per_iteration": 3.539909601211548 }, { "auxiliary_loss_clip": 0.01086029, "auxiliary_loss_mlp": 0.01000411, "balance_loss_clip": 1.01254582, "balance_loss_mlp": 0.99877757, "epoch": 0.3153971021463356, "flos": 68476917192960.0, "grad_norm": 0.884482237755197, "language_loss": 0.6363045, "learning_rate": 3.204867001502487e-06, "loss": 0.65716887, "num_input_tokens_seen": 56488425, "step": 2623, "time_per_iteration": 3.1143457889556885 }, { "auxiliary_loss_clip": 0.01200481, "auxiliary_loss_mlp": 0.0103158, "balance_loss_clip": 1.05812347, "balance_loss_mlp": 1.02273464, "epoch": 0.3155173450369747, "flos": 25593718764480.0, "grad_norm": 1.7984198402851808, "language_loss": 0.80773437, "learning_rate": 3.2042451590479567e-06, "loss": 0.830055, "num_input_tokens_seen": 56508940, "step": 2624, "time_per_iteration": 2.565639019012451 }, { "auxiliary_loss_clip": 0.01196568, "auxiliary_loss_mlp": 0.01025958, "balance_loss_clip": 1.05773365, "balance_loss_mlp": 1.01788712, "epoch": 0.31563758792761376, "flos": 24309336364320.0, "grad_norm": 1.7012309076132348, "language_loss": 0.86634135, "learning_rate": 3.203623133910394e-06, "loss": 0.88856661, "num_input_tokens_seen": 56527245, "step": 2625, "time_per_iteration": 2.5698421001434326 }, { "auxiliary_loss_clip": 0.01128481, "auxiliary_loss_mlp": 0.01033509, "balance_loss_clip": 1.04947901, "balance_loss_mlp": 1.02532554, "epoch": 0.31575783081825287, "flos": 31903869962400.0, "grad_norm": 2.4556230552264524, "language_loss": 0.77106166, "learning_rate": 3.203000926184158e-06, "loss": 0.79268157, "num_input_tokens_seen": 56546170, "step": 2626, "time_per_iteration": 2.7694637775421143 }, { "auxiliary_loss_clip": 0.01201729, "auxiliary_loss_mlp": 0.01027881, "balance_loss_clip": 1.06032753, "balance_loss_mlp": 1.02023435, "epoch": 0.315878073708892, "flos": 30810964851360.0, "grad_norm": 1.68280093645401, "language_loss": 0.77469563, "learning_rate": 3.202378535963639e-06, "loss": 0.79699177, "num_input_tokens_seen": 56567085, "step": 2627, "time_per_iteration": 2.6086654663085938 }, { "auxiliary_loss_clip": 0.01165663, "auxiliary_loss_mlp": 0.00763643, "balance_loss_clip": 1.05518436, "balance_loss_mlp": 1.00008273, "epoch": 0.31599831659953104, "flos": 22200267909120.0, "grad_norm": 1.6116183312473318, "language_loss": 0.84018052, "learning_rate": 3.2017559633432516e-06, "loss": 0.85947359, "num_input_tokens_seen": 56586715, "step": 2628, "time_per_iteration": 2.621659517288208 }, { "auxiliary_loss_clip": 0.01176834, "auxiliary_loss_mlp": 0.01027751, "balance_loss_clip": 1.05475426, "balance_loss_mlp": 1.01914406, "epoch": 0.31611855949017015, "flos": 25593467345760.0, "grad_norm": 1.858237730382276, "language_loss": 0.66211152, "learning_rate": 3.2011332084174398e-06, "loss": 0.68415737, "num_input_tokens_seen": 56607585, "step": 2629, "time_per_iteration": 2.659337282180786 }, { "auxiliary_loss_clip": 0.01184541, "auxiliary_loss_mlp": 0.01031495, "balance_loss_clip": 1.05740666, "balance_loss_mlp": 1.02327514, "epoch": 0.31623880238080926, "flos": 20594091125280.0, "grad_norm": 1.5727703663158963, "language_loss": 0.89481103, "learning_rate": 3.2005102712806756e-06, "loss": 0.91697133, "num_input_tokens_seen": 56626415, "step": 2630, "time_per_iteration": 2.560108184814453 }, { "auxiliary_loss_clip": 0.01190412, "auxiliary_loss_mlp": 0.01034677, "balance_loss_clip": 1.0569582, "balance_loss_mlp": 1.02633286, "epoch": 0.3163590452714483, "flos": 12784920187200.0, "grad_norm": 1.8959048349630976, "language_loss": 0.73068416, "learning_rate": 3.1998871520274575e-06, "loss": 0.75293505, "num_input_tokens_seen": 56641750, "step": 2631, "time_per_iteration": 2.5561158657073975 }, { "auxiliary_loss_clip": 0.01171936, "auxiliary_loss_mlp": 0.01031792, "balance_loss_clip": 1.05187106, "balance_loss_mlp": 1.02317321, "epoch": 0.3164792881620874, "flos": 23041296180960.0, "grad_norm": 1.6938661698068593, "language_loss": 0.84811836, "learning_rate": 3.199263850752312e-06, "loss": 0.87015557, "num_input_tokens_seen": 56662585, "step": 2632, "time_per_iteration": 2.6141464710235596 }, { "auxiliary_loss_clip": 0.01187612, "auxiliary_loss_mlp": 0.01030158, "balance_loss_clip": 1.05848086, "balance_loss_mlp": 1.02195656, "epoch": 0.31659953105272653, "flos": 18296275069920.0, "grad_norm": 2.417187084781966, "language_loss": 0.85248065, "learning_rate": 3.198640367549795e-06, "loss": 0.87465835, "num_input_tokens_seen": 56681480, "step": 2633, "time_per_iteration": 2.5639851093292236 }, { "auxiliary_loss_clip": 0.01188381, "auxiliary_loss_mlp": 0.00762995, "balance_loss_clip": 1.05629051, "balance_loss_mlp": 1.00003219, "epoch": 0.3167197739433656, "flos": 25703425912320.0, "grad_norm": 1.6695248135266119, "language_loss": 0.85561591, "learning_rate": 3.198016702514487e-06, "loss": 0.8751297, "num_input_tokens_seen": 56701760, "step": 2634, "time_per_iteration": 2.6097848415374756 }, { "auxiliary_loss_clip": 0.01198669, "auxiliary_loss_mlp": 0.0102744, "balance_loss_clip": 1.05835319, "balance_loss_mlp": 1.01925004, "epoch": 0.3168400168340047, "flos": 23546019617280.0, "grad_norm": 1.7170912472909323, "language_loss": 0.84827161, "learning_rate": 3.1973928557409972e-06, "loss": 0.87053269, "num_input_tokens_seen": 56719800, "step": 2635, "time_per_iteration": 2.5617282390594482 }, { "auxiliary_loss_clip": 0.01197152, "auxiliary_loss_mlp": 0.01031152, "balance_loss_clip": 1.05872345, "balance_loss_mlp": 1.02303982, "epoch": 0.31696025972464376, "flos": 28366453567680.0, "grad_norm": 1.9788010143212302, "language_loss": 0.71177197, "learning_rate": 3.1967688273239636e-06, "loss": 0.73405504, "num_input_tokens_seen": 56739605, "step": 2636, "time_per_iteration": 2.580695867538452 }, { "auxiliary_loss_clip": 0.0115518, "auxiliary_loss_mlp": 0.01028926, "balance_loss_clip": 1.05368304, "balance_loss_mlp": 1.02103448, "epoch": 0.31708050261528287, "flos": 16399114265760.0, "grad_norm": 1.8915653360371056, "language_loss": 0.81856066, "learning_rate": 3.1961446173580503e-06, "loss": 0.84040177, "num_input_tokens_seen": 56756545, "step": 2637, "time_per_iteration": 2.6192362308502197 }, { "auxiliary_loss_clip": 0.0116886, "auxiliary_loss_mlp": 0.01029001, "balance_loss_clip": 1.05555964, "balance_loss_mlp": 1.02084661, "epoch": 0.317200745505922, "flos": 26212351632960.0, "grad_norm": 1.664216617511219, "language_loss": 0.77052265, "learning_rate": 3.1955202259379502e-06, "loss": 0.79250121, "num_input_tokens_seen": 56778275, "step": 2638, "time_per_iteration": 2.634406089782715 }, { "auxiliary_loss_clip": 0.01182242, "auxiliary_loss_mlp": 0.01031874, "balance_loss_clip": 1.05474877, "balance_loss_mlp": 1.02330279, "epoch": 0.31732098839656103, "flos": 31350880520160.0, "grad_norm": 1.6353607189415125, "language_loss": 0.82770079, "learning_rate": 3.194895653158381e-06, "loss": 0.84984195, "num_input_tokens_seen": 56797215, "step": 2639, "time_per_iteration": 2.656895399093628 }, { "auxiliary_loss_clip": 0.01084391, "auxiliary_loss_mlp": 0.01000297, "balance_loss_clip": 1.01172161, "balance_loss_mlp": 0.99885482, "epoch": 0.31744123128720014, "flos": 58989036376800.0, "grad_norm": 0.766409762689205, "language_loss": 0.55567932, "learning_rate": 3.194270899114093e-06, "loss": 0.57652617, "num_input_tokens_seen": 56863010, "step": 2640, "time_per_iteration": 3.9768874645233154 }, { "auxiliary_loss_clip": 0.0119363, "auxiliary_loss_mlp": 0.01030381, "balance_loss_clip": 1.0579772, "balance_loss_mlp": 1.02187586, "epoch": 0.31756147417783925, "flos": 17417576295360.0, "grad_norm": 1.742511138492741, "language_loss": 0.8211512, "learning_rate": 3.193645963899858e-06, "loss": 0.84339142, "num_input_tokens_seen": 56880625, "step": 2641, "time_per_iteration": 3.2761242389678955 }, { "auxiliary_loss_clip": 0.01169106, "auxiliary_loss_mlp": 0.01033232, "balance_loss_clip": 1.05696774, "balance_loss_mlp": 1.0248158, "epoch": 0.3176817170684783, "flos": 25481677014240.0, "grad_norm": 2.5668014473778866, "language_loss": 0.84161663, "learning_rate": 3.193020847610479e-06, "loss": 0.86364001, "num_input_tokens_seen": 56900945, "step": 2642, "time_per_iteration": 2.6418514251708984 }, { "auxiliary_loss_clip": 0.01164403, "auxiliary_loss_mlp": 0.01033402, "balance_loss_clip": 1.05517483, "balance_loss_mlp": 1.02496231, "epoch": 0.3178019599591174, "flos": 24972607625760.0, "grad_norm": 2.475853536369104, "language_loss": 0.71331644, "learning_rate": 3.192395550340787e-06, "loss": 0.73529446, "num_input_tokens_seen": 56918895, "step": 2643, "time_per_iteration": 3.519826889038086 }, { "auxiliary_loss_clip": 0.01186132, "auxiliary_loss_mlp": 0.01036106, "balance_loss_clip": 1.058465, "balance_loss_mlp": 1.02816105, "epoch": 0.31792220284975653, "flos": 12422222774400.0, "grad_norm": 1.891618685516617, "language_loss": 0.76813221, "learning_rate": 3.191770072185638e-06, "loss": 0.79035461, "num_input_tokens_seen": 56935890, "step": 2644, "time_per_iteration": 2.5547521114349365 }, { "auxiliary_loss_clip": 0.01184046, "auxiliary_loss_mlp": 0.01033038, "balance_loss_clip": 1.05897427, "balance_loss_mlp": 1.02332211, "epoch": 0.3180424457403956, "flos": 15485762013120.0, "grad_norm": 2.1781045421352334, "language_loss": 0.72806942, "learning_rate": 3.191144413239916e-06, "loss": 0.75024033, "num_input_tokens_seen": 56952460, "step": 2645, "time_per_iteration": 2.5250132083892822 }, { "auxiliary_loss_clip": 0.01171147, "auxiliary_loss_mlp": 0.01030543, "balance_loss_clip": 1.05490088, "balance_loss_mlp": 1.02181685, "epoch": 0.3181626886310347, "flos": 26174968465920.0, "grad_norm": 1.9446402523748094, "language_loss": 0.88148546, "learning_rate": 3.190518573598534e-06, "loss": 0.9035024, "num_input_tokens_seen": 56969065, "step": 2646, "time_per_iteration": 2.6322731971740723 }, { "auxiliary_loss_clip": 0.01161179, "auxiliary_loss_mlp": 0.01028664, "balance_loss_clip": 1.05232799, "balance_loss_mlp": 1.02040935, "epoch": 0.3182829315216738, "flos": 25483113692640.0, "grad_norm": 1.4744687555431457, "language_loss": 0.77395809, "learning_rate": 3.1898925533564308e-06, "loss": 0.79585648, "num_input_tokens_seen": 56990535, "step": 2647, "time_per_iteration": 2.6854491233825684 }, { "auxiliary_loss_clip": 0.01141551, "auxiliary_loss_mlp": 0.01040344, "balance_loss_clip": 1.04803848, "balance_loss_mlp": 1.03236318, "epoch": 0.31840317441231286, "flos": 18113705186880.0, "grad_norm": 1.8920267252853624, "language_loss": 0.6388824, "learning_rate": 3.1892663526085733e-06, "loss": 0.66070139, "num_input_tokens_seen": 57008910, "step": 2648, "time_per_iteration": 3.476743221282959 }, { "auxiliary_loss_clip": 0.0108583, "auxiliary_loss_mlp": 0.01002154, "balance_loss_clip": 1.01259732, "balance_loss_mlp": 1.0006876, "epoch": 0.31852341730295197, "flos": 64741959931200.0, "grad_norm": 0.7524844628664168, "language_loss": 0.56949621, "learning_rate": 3.188639971449956e-06, "loss": 0.59037602, "num_input_tokens_seen": 57074960, "step": 2649, "time_per_iteration": 3.1110198497772217 }, { "auxiliary_loss_clip": 0.01202097, "auxiliary_loss_mlp": 0.01029792, "balance_loss_clip": 1.05914664, "balance_loss_mlp": 1.02100062, "epoch": 0.318643660193591, "flos": 20668139120160.0, "grad_norm": 1.9111983145822666, "language_loss": 0.72258925, "learning_rate": 3.1880134099756e-06, "loss": 0.74490815, "num_input_tokens_seen": 57094595, "step": 2650, "time_per_iteration": 2.5537192821502686 }, { "auxiliary_loss_clip": 0.01183893, "auxiliary_loss_mlp": 0.01028688, "balance_loss_clip": 1.0562247, "balance_loss_mlp": 1.02048612, "epoch": 0.31876390308423014, "flos": 26943349504320.0, "grad_norm": 2.5810361914581823, "language_loss": 0.69649398, "learning_rate": 3.1873866682805535e-06, "loss": 0.71861982, "num_input_tokens_seen": 57115290, "step": 2651, "time_per_iteration": 2.6059820652008057 }, { "auxiliary_loss_clip": 0.01175364, "auxiliary_loss_mlp": 0.01031186, "balance_loss_clip": 1.05730462, "balance_loss_mlp": 1.02268028, "epoch": 0.31888414597486925, "flos": 18041919960480.0, "grad_norm": 1.8952733732268432, "language_loss": 0.88139457, "learning_rate": 3.186759746459894e-06, "loss": 0.90346003, "num_input_tokens_seen": 57134400, "step": 2652, "time_per_iteration": 2.593768358230591 }, { "auxiliary_loss_clip": 0.0117015, "auxiliary_loss_mlp": 0.01032027, "balance_loss_clip": 1.05428338, "balance_loss_mlp": 1.02380729, "epoch": 0.3190043888655083, "flos": 25149323044320.0, "grad_norm": 2.305292347560106, "language_loss": 0.79526228, "learning_rate": 3.1861326446087246e-06, "loss": 0.81728405, "num_input_tokens_seen": 57153140, "step": 2653, "time_per_iteration": 2.63008713722229 }, { "auxiliary_loss_clip": 0.01189469, "auxiliary_loss_mlp": 0.01028949, "balance_loss_clip": 1.05852914, "balance_loss_mlp": 1.0204252, "epoch": 0.3191246317561474, "flos": 22053895933440.0, "grad_norm": 1.9928993269151625, "language_loss": 0.71649772, "learning_rate": 3.1855053628221763e-06, "loss": 0.73868185, "num_input_tokens_seen": 57172395, "step": 2654, "time_per_iteration": 2.57148814201355 }, { "auxiliary_loss_clip": 0.01150271, "auxiliary_loss_mlp": 0.01029268, "balance_loss_clip": 1.0508647, "balance_loss_mlp": 1.02044058, "epoch": 0.3192448746467865, "flos": 14901818539680.0, "grad_norm": 2.4767191719743975, "language_loss": 0.90225208, "learning_rate": 3.184877901195407e-06, "loss": 0.92404747, "num_input_tokens_seen": 57189090, "step": 2655, "time_per_iteration": 2.6403424739837646 }, { "auxiliary_loss_clip": 0.01065874, "auxiliary_loss_mlp": 0.01003196, "balance_loss_clip": 1.01907635, "balance_loss_mlp": 1.0018723, "epoch": 0.3193651175374256, "flos": 67234844971680.0, "grad_norm": 0.800441629327811, "language_loss": 0.62860757, "learning_rate": 3.184250259823602e-06, "loss": 0.64929825, "num_input_tokens_seen": 57251620, "step": 2656, "time_per_iteration": 3.2361488342285156 }, { "auxiliary_loss_clip": 0.01161914, "auxiliary_loss_mlp": 0.01030305, "balance_loss_clip": 1.05495059, "balance_loss_mlp": 1.02162027, "epoch": 0.3194853604280647, "flos": 12233080087680.0, "grad_norm": 2.1096619032987127, "language_loss": 0.8111257, "learning_rate": 3.183622438801974e-06, "loss": 0.83304787, "num_input_tokens_seen": 57266910, "step": 2657, "time_per_iteration": 2.6116297245025635 }, { "auxiliary_loss_clip": 0.01203964, "auxiliary_loss_mlp": 0.0102941, "balance_loss_clip": 1.06301987, "balance_loss_mlp": 1.02153063, "epoch": 0.3196056033187038, "flos": 14939920045920.0, "grad_norm": 1.702360396556124, "language_loss": 0.75212228, "learning_rate": 3.1829944382257637e-06, "loss": 0.77445602, "num_input_tokens_seen": 57285040, "step": 2658, "time_per_iteration": 2.541893243789673 }, { "auxiliary_loss_clip": 0.01182317, "auxiliary_loss_mlp": 0.01037249, "balance_loss_clip": 1.05782151, "balance_loss_mlp": 1.02874315, "epoch": 0.31972584620934286, "flos": 23768881941120.0, "grad_norm": 2.3086304538454785, "language_loss": 0.81450206, "learning_rate": 3.1823662581902373e-06, "loss": 0.8366977, "num_input_tokens_seen": 57302725, "step": 2659, "time_per_iteration": 2.5991101264953613 }, { "auxiliary_loss_clip": 0.01141991, "auxiliary_loss_mlp": 0.01034974, "balance_loss_clip": 1.04672015, "balance_loss_mlp": 1.0262357, "epoch": 0.31984608909998197, "flos": 21251543839200.0, "grad_norm": 2.151315044714325, "language_loss": 0.74310124, "learning_rate": 3.1817378987906896e-06, "loss": 0.76487088, "num_input_tokens_seen": 57322230, "step": 2660, "time_per_iteration": 2.656024932861328 }, { "auxiliary_loss_clip": 0.01138684, "auxiliary_loss_mlp": 0.01033592, "balance_loss_clip": 1.05379546, "balance_loss_mlp": 1.02516437, "epoch": 0.3199663319906211, "flos": 18296239152960.0, "grad_norm": 2.0708100865471106, "language_loss": 0.79932761, "learning_rate": 3.181109360122442e-06, "loss": 0.82105041, "num_input_tokens_seen": 57339820, "step": 2661, "time_per_iteration": 2.630098342895508 }, { "auxiliary_loss_clip": 0.01152198, "auxiliary_loss_mlp": 0.0103051, "balance_loss_clip": 1.05162323, "balance_loss_mlp": 1.02195668, "epoch": 0.32008657488126013, "flos": 18733631065920.0, "grad_norm": 2.097126223318608, "language_loss": 0.78257668, "learning_rate": 3.1804806422808445e-06, "loss": 0.80440378, "num_input_tokens_seen": 57356955, "step": 2662, "time_per_iteration": 2.6348559856414795 }, { "auxiliary_loss_clip": 0.01160635, "auxiliary_loss_mlp": 0.01030513, "balance_loss_clip": 1.05371904, "balance_loss_mlp": 1.02185893, "epoch": 0.32020681777189924, "flos": 20595348218880.0, "grad_norm": 1.5825326342107184, "language_loss": 0.73076439, "learning_rate": 3.1798517453612714e-06, "loss": 0.75267589, "num_input_tokens_seen": 57376760, "step": 2663, "time_per_iteration": 2.593938112258911 }, { "auxiliary_loss_clip": 0.01183181, "auxiliary_loss_mlp": 0.0102951, "balance_loss_clip": 1.05789828, "balance_loss_mlp": 1.02093887, "epoch": 0.32032706066253835, "flos": 35261697581760.0, "grad_norm": 1.8026021670013557, "language_loss": 0.75326645, "learning_rate": 3.1792226694591265e-06, "loss": 0.77539337, "num_input_tokens_seen": 57398145, "step": 2664, "time_per_iteration": 2.6976819038391113 }, { "auxiliary_loss_clip": 0.01156324, "auxiliary_loss_mlp": 0.01035601, "balance_loss_clip": 1.05531001, "balance_loss_mlp": 1.02737522, "epoch": 0.3204473035531774, "flos": 15304233721920.0, "grad_norm": 1.9214208021972912, "language_loss": 0.80161881, "learning_rate": 3.178593414669841e-06, "loss": 0.82353806, "num_input_tokens_seen": 57416730, "step": 2665, "time_per_iteration": 2.6161022186279297 }, { "auxiliary_loss_clip": 0.0118754, "auxiliary_loss_mlp": 0.01031164, "balance_loss_clip": 1.05721903, "balance_loss_mlp": 1.02218175, "epoch": 0.3205675464438165, "flos": 24462568479360.0, "grad_norm": 2.2438978481448704, "language_loss": 0.69941652, "learning_rate": 3.1779639810888707e-06, "loss": 0.72160363, "num_input_tokens_seen": 57436325, "step": 2666, "time_per_iteration": 2.590197801589966 }, { "auxiliary_loss_clip": 0.01182876, "auxiliary_loss_mlp": 0.01032274, "balance_loss_clip": 1.0569675, "balance_loss_mlp": 1.02401257, "epoch": 0.3206877893344556, "flos": 22456239281760.0, "grad_norm": 1.670130516354444, "language_loss": 0.75596702, "learning_rate": 3.1773343688117013e-06, "loss": 0.77811855, "num_input_tokens_seen": 57457235, "step": 2667, "time_per_iteration": 3.4315836429595947 }, { "auxiliary_loss_clip": 0.01173686, "auxiliary_loss_mlp": 0.00763379, "balance_loss_clip": 1.05414772, "balance_loss_mlp": 1.00011897, "epoch": 0.3208080322250947, "flos": 20412239581440.0, "grad_norm": 2.027316556560145, "language_loss": 0.8403796, "learning_rate": 3.1767045779338445e-06, "loss": 0.85975027, "num_input_tokens_seen": 57474895, "step": 2668, "time_per_iteration": 2.59426212310791 }, { "auxiliary_loss_clip": 0.01182793, "auxiliary_loss_mlp": 0.01034335, "balance_loss_clip": 1.05401111, "balance_loss_mlp": 1.02672315, "epoch": 0.3209282751157338, "flos": 21762121740000.0, "grad_norm": 2.030980210612223, "language_loss": 0.91362607, "learning_rate": 3.176074608550839e-06, "loss": 0.93579739, "num_input_tokens_seen": 57490715, "step": 2669, "time_per_iteration": 3.457892417907715 }, { "auxiliary_loss_clip": 0.01131013, "auxiliary_loss_mlp": 0.01035278, "balance_loss_clip": 1.05209088, "balance_loss_mlp": 1.0262779, "epoch": 0.32104851800637285, "flos": 22055045276160.0, "grad_norm": 4.7198321995832355, "language_loss": 0.82331336, "learning_rate": 3.17544446075825e-06, "loss": 0.84497631, "num_input_tokens_seen": 57509880, "step": 2670, "time_per_iteration": 2.7398033142089844 }, { "auxiliary_loss_clip": 0.01172611, "auxiliary_loss_mlp": 0.01029997, "balance_loss_clip": 1.0530982, "balance_loss_mlp": 1.02222455, "epoch": 0.32116876089701196, "flos": 37012306825440.0, "grad_norm": 1.7148674370643961, "language_loss": 0.70836115, "learning_rate": 3.174814134651671e-06, "loss": 0.73038721, "num_input_tokens_seen": 57532430, "step": 2671, "time_per_iteration": 2.759014368057251 }, { "auxiliary_loss_clip": 0.01195006, "auxiliary_loss_mlp": 0.01028489, "balance_loss_clip": 1.05748546, "balance_loss_mlp": 1.02039468, "epoch": 0.3212890037876511, "flos": 21979237350240.0, "grad_norm": 1.7035826255839999, "language_loss": 0.8002333, "learning_rate": 3.1741836303267215e-06, "loss": 0.82246828, "num_input_tokens_seen": 57551965, "step": 2672, "time_per_iteration": 2.5772640705108643 }, { "auxiliary_loss_clip": 0.01197587, "auxiliary_loss_mlp": 0.0103433, "balance_loss_clip": 1.05855119, "balance_loss_mlp": 1.02551484, "epoch": 0.32140924667829013, "flos": 10342348419360.0, "grad_norm": 2.2253981997106766, "language_loss": 0.75074148, "learning_rate": 3.1735529478790496e-06, "loss": 0.77306068, "num_input_tokens_seen": 57569955, "step": 2673, "time_per_iteration": 2.537916421890259 }, { "auxiliary_loss_clip": 0.01182889, "auxiliary_loss_mlp": 0.01029437, "balance_loss_clip": 1.05561256, "balance_loss_mlp": 1.0213722, "epoch": 0.32152948956892924, "flos": 50798915821920.0, "grad_norm": 1.8625603547579788, "language_loss": 0.79691875, "learning_rate": 3.172922087404328e-06, "loss": 0.81904197, "num_input_tokens_seen": 57592215, "step": 2674, "time_per_iteration": 2.8208203315734863 }, { "auxiliary_loss_clip": 0.01088097, "auxiliary_loss_mlp": 0.01000945, "balance_loss_clip": 1.01599967, "balance_loss_mlp": 0.99940699, "epoch": 0.32164973245956835, "flos": 63863261156640.0, "grad_norm": 0.768609361978073, "language_loss": 0.55309498, "learning_rate": 3.1722910489982586e-06, "loss": 0.57398534, "num_input_tokens_seen": 57652575, "step": 2675, "time_per_iteration": 4.016841411590576 }, { "auxiliary_loss_clip": 0.01168733, "auxiliary_loss_mlp": 0.01028941, "balance_loss_clip": 1.05486727, "balance_loss_mlp": 1.02042341, "epoch": 0.3217699753502074, "flos": 23513952160320.0, "grad_norm": 1.6592396377309713, "language_loss": 0.80001855, "learning_rate": 3.1716598327565694e-06, "loss": 0.82199526, "num_input_tokens_seen": 57672215, "step": 2676, "time_per_iteration": 2.631667137145996 }, { "auxiliary_loss_clip": 0.01197453, "auxiliary_loss_mlp": 0.01030708, "balance_loss_clip": 1.05775523, "balance_loss_mlp": 1.02268577, "epoch": 0.3218902182408465, "flos": 19062537007680.0, "grad_norm": 1.676376450284969, "language_loss": 0.84270132, "learning_rate": 3.171028438775015e-06, "loss": 0.8649829, "num_input_tokens_seen": 57691410, "step": 2677, "time_per_iteration": 2.547276496887207 }, { "auxiliary_loss_clip": 0.01197474, "auxiliary_loss_mlp": 0.01029794, "balance_loss_clip": 1.05786932, "balance_loss_mlp": 1.02180123, "epoch": 0.3220104611314856, "flos": 20375574753600.0, "grad_norm": 1.7177139517570665, "language_loss": 0.83978665, "learning_rate": 3.170396867149377e-06, "loss": 0.86205935, "num_input_tokens_seen": 57709415, "step": 2678, "time_per_iteration": 2.5541303157806396 }, { "auxiliary_loss_clip": 0.01134378, "auxiliary_loss_mlp": 0.01030339, "balance_loss_clip": 1.05121827, "balance_loss_mlp": 1.02169633, "epoch": 0.3221307040221247, "flos": 20117017359840.0, "grad_norm": 1.75918749926131, "language_loss": 0.86348158, "learning_rate": 3.1697651179754653e-06, "loss": 0.88512874, "num_input_tokens_seen": 57728075, "step": 2679, "time_per_iteration": 2.6349105834960938 }, { "auxiliary_loss_clip": 0.01158639, "auxiliary_loss_mlp": 0.0103037, "balance_loss_clip": 1.05702519, "balance_loss_mlp": 1.02202582, "epoch": 0.3222509469127638, "flos": 23987793399360.0, "grad_norm": 1.6287868837744606, "language_loss": 0.72896904, "learning_rate": 3.1691331913491153e-06, "loss": 0.75085914, "num_input_tokens_seen": 57750645, "step": 2680, "time_per_iteration": 2.6826624870300293 }, { "auxiliary_loss_clip": 0.0119827, "auxiliary_loss_mlp": 0.01027717, "balance_loss_clip": 1.05719137, "balance_loss_mlp": 1.01930118, "epoch": 0.32237118980340285, "flos": 17675738602560.0, "grad_norm": 2.3043196567198407, "language_loss": 0.85063666, "learning_rate": 3.1685010873661898e-06, "loss": 0.87289649, "num_input_tokens_seen": 57769820, "step": 2681, "time_per_iteration": 2.527125835418701 }, { "auxiliary_loss_clip": 0.01181279, "auxiliary_loss_mlp": 0.01031803, "balance_loss_clip": 1.05703175, "balance_loss_mlp": 1.02359581, "epoch": 0.32249143269404196, "flos": 23147986304160.0, "grad_norm": 1.903385168463529, "language_loss": 0.79594707, "learning_rate": 3.167868806122578e-06, "loss": 0.81807786, "num_input_tokens_seen": 57788870, "step": 2682, "time_per_iteration": 2.577831745147705 }, { "auxiliary_loss_clip": 0.01177359, "auxiliary_loss_mlp": 0.01031902, "balance_loss_clip": 1.05981791, "balance_loss_mlp": 1.02387357, "epoch": 0.32261167558468107, "flos": 24422311955520.0, "grad_norm": 2.6658194164555904, "language_loss": 0.66252339, "learning_rate": 3.1672363477141968e-06, "loss": 0.68461597, "num_input_tokens_seen": 57808165, "step": 2683, "time_per_iteration": 2.6279780864715576 }, { "auxiliary_loss_clip": 0.01174808, "auxiliary_loss_mlp": 0.01031033, "balance_loss_clip": 1.05504918, "balance_loss_mlp": 1.02273619, "epoch": 0.3227319184753201, "flos": 30367179719520.0, "grad_norm": 1.8882839800909779, "language_loss": 0.84346759, "learning_rate": 3.1666037122369903e-06, "loss": 0.86552602, "num_input_tokens_seen": 57828825, "step": 2684, "time_per_iteration": 2.6720287799835205 }, { "auxiliary_loss_clip": 0.01179146, "auxiliary_loss_mlp": 0.01028497, "balance_loss_clip": 1.05216885, "balance_loss_mlp": 1.02024209, "epoch": 0.32285216136595923, "flos": 16946177409600.0, "grad_norm": 2.1794859085960385, "language_loss": 0.86603081, "learning_rate": 3.165970899786928e-06, "loss": 0.88810718, "num_input_tokens_seen": 57846740, "step": 2685, "time_per_iteration": 2.5476386547088623 }, { "auxiliary_loss_clip": 0.01160822, "auxiliary_loss_mlp": 0.01030583, "balance_loss_clip": 1.05393791, "balance_loss_mlp": 1.02248621, "epoch": 0.32297240425659834, "flos": 21981536035680.0, "grad_norm": 1.6396475617586734, "language_loss": 0.75031704, "learning_rate": 3.1653379104600067e-06, "loss": 0.7722311, "num_input_tokens_seen": 57866885, "step": 2686, "time_per_iteration": 2.633653163909912 }, { "auxiliary_loss_clip": 0.011826, "auxiliary_loss_mlp": 0.01026819, "balance_loss_clip": 1.05545688, "balance_loss_mlp": 1.01903462, "epoch": 0.3230926471472374, "flos": 22748049392160.0, "grad_norm": 1.8375259639362154, "language_loss": 0.69538999, "learning_rate": 3.164704744352251e-06, "loss": 0.71748412, "num_input_tokens_seen": 57887690, "step": 2687, "time_per_iteration": 2.602468490600586 }, { "auxiliary_loss_clip": 0.01184117, "auxiliary_loss_mlp": 0.01029292, "balance_loss_clip": 1.05764163, "balance_loss_mlp": 1.02135849, "epoch": 0.3232128900378765, "flos": 16942980800160.0, "grad_norm": 1.8147546415437348, "language_loss": 0.80843681, "learning_rate": 3.164071401559713e-06, "loss": 0.83057094, "num_input_tokens_seen": 57905090, "step": 2688, "time_per_iteration": 2.554643392562866 }, { "auxiliary_loss_clip": 0.01171707, "auxiliary_loss_mlp": 0.01030493, "balance_loss_clip": 1.05412495, "balance_loss_mlp": 1.02237535, "epoch": 0.3233331329285156, "flos": 24023739888000.0, "grad_norm": 1.7906992402705975, "language_loss": 0.71202415, "learning_rate": 3.1634378821784674e-06, "loss": 0.7340461, "num_input_tokens_seen": 57925305, "step": 2689, "time_per_iteration": 2.63692307472229 }, { "auxiliary_loss_clip": 0.01158504, "auxiliary_loss_mlp": 0.010269, "balance_loss_clip": 1.05439425, "balance_loss_mlp": 1.018466, "epoch": 0.3234533758191547, "flos": 18113848854720.0, "grad_norm": 2.425863777068181, "language_loss": 0.73980069, "learning_rate": 3.1628041863046208e-06, "loss": 0.76165473, "num_input_tokens_seen": 57942720, "step": 2690, "time_per_iteration": 2.5862162113189697 }, { "auxiliary_loss_clip": 0.01202205, "auxiliary_loss_mlp": 0.01033795, "balance_loss_clip": 1.05756307, "balance_loss_mlp": 1.02443683, "epoch": 0.3235736187097938, "flos": 16946141492640.0, "grad_norm": 3.0986812432489232, "language_loss": 0.91309786, "learning_rate": 3.162170314034304e-06, "loss": 0.93545783, "num_input_tokens_seen": 57960135, "step": 2691, "time_per_iteration": 2.525428056716919 }, { "auxiliary_loss_clip": 0.01202614, "auxiliary_loss_mlp": 0.01040863, "balance_loss_clip": 1.05862403, "balance_loss_mlp": 1.0319283, "epoch": 0.3236938616004329, "flos": 22127153755200.0, "grad_norm": 2.0408395259899845, "language_loss": 0.80832398, "learning_rate": 3.1615362654636738e-06, "loss": 0.83075881, "num_input_tokens_seen": 57980875, "step": 2692, "time_per_iteration": 2.560103178024292 }, { "auxiliary_loss_clip": 0.01150751, "auxiliary_loss_mlp": 0.01033879, "balance_loss_clip": 1.05740178, "balance_loss_mlp": 1.02648175, "epoch": 0.32381410449107195, "flos": 17164621947360.0, "grad_norm": 1.7768341932702416, "language_loss": 0.87200177, "learning_rate": 3.1609020406889163e-06, "loss": 0.89384806, "num_input_tokens_seen": 57998310, "step": 2693, "time_per_iteration": 4.151353120803833 }, { "auxiliary_loss_clip": 0.01170669, "auxiliary_loss_mlp": 0.01029046, "balance_loss_clip": 1.05323839, "balance_loss_mlp": 1.01990879, "epoch": 0.32393434738171106, "flos": 16578128369760.0, "grad_norm": 1.6545337245248444, "language_loss": 0.85248005, "learning_rate": 3.1602676398062416e-06, "loss": 0.87447715, "num_input_tokens_seen": 58017220, "step": 2694, "time_per_iteration": 2.5915801525115967 }, { "auxiliary_loss_clip": 0.01184418, "auxiliary_loss_mlp": 0.01027163, "balance_loss_clip": 1.05710196, "balance_loss_mlp": 1.01885378, "epoch": 0.3240545902723502, "flos": 25483616530080.0, "grad_norm": 3.126875984163331, "language_loss": 0.6143738, "learning_rate": 3.1596330629118886e-06, "loss": 0.63648963, "num_input_tokens_seen": 58037190, "step": 2695, "time_per_iteration": 3.485455274581909 }, { "auxiliary_loss_clip": 0.01132818, "auxiliary_loss_mlp": 0.01034987, "balance_loss_clip": 1.05015779, "balance_loss_mlp": 1.02614188, "epoch": 0.32417483316298923, "flos": 35845856556960.0, "grad_norm": 2.620239385107778, "language_loss": 0.72920203, "learning_rate": 3.1589983101021223e-06, "loss": 0.75088012, "num_input_tokens_seen": 58055820, "step": 2696, "time_per_iteration": 2.780578374862671 }, { "auxiliary_loss_clip": 0.01170692, "auxiliary_loss_mlp": 0.01034193, "balance_loss_clip": 1.05484688, "balance_loss_mlp": 1.02617049, "epoch": 0.32429507605362834, "flos": 30080505734400.0, "grad_norm": 2.1834062569449664, "language_loss": 0.84843403, "learning_rate": 3.1583633814732337e-06, "loss": 0.87048292, "num_input_tokens_seen": 58075340, "step": 2697, "time_per_iteration": 2.647461414337158 }, { "auxiliary_loss_clip": 0.01198953, "auxiliary_loss_mlp": 0.01029647, "balance_loss_clip": 1.05740499, "balance_loss_mlp": 1.02101588, "epoch": 0.3244153189442674, "flos": 18223268666880.0, "grad_norm": 2.6149228506149944, "language_loss": 0.71677476, "learning_rate": 3.157728277121541e-06, "loss": 0.73906076, "num_input_tokens_seen": 58093515, "step": 2698, "time_per_iteration": 2.543006181716919 }, { "auxiliary_loss_clip": 0.01198865, "auxiliary_loss_mlp": 0.01029835, "balance_loss_clip": 1.05628586, "balance_loss_mlp": 1.02121568, "epoch": 0.3245355618349065, "flos": 17710320246720.0, "grad_norm": 7.852280094779135, "language_loss": 0.78362316, "learning_rate": 3.1570929971433897e-06, "loss": 0.80591023, "num_input_tokens_seen": 58109300, "step": 2699, "time_per_iteration": 2.509138584136963 }, { "auxiliary_loss_clip": 0.01186394, "auxiliary_loss_mlp": 0.01039384, "balance_loss_clip": 1.05901384, "balance_loss_mlp": 1.03105164, "epoch": 0.3246558047255456, "flos": 23440802089440.0, "grad_norm": 3.8626479299219576, "language_loss": 0.83581889, "learning_rate": 3.1564575416351504e-06, "loss": 0.85807669, "num_input_tokens_seen": 58128000, "step": 2700, "time_per_iteration": 2.5704216957092285 }, { "auxiliary_loss_clip": 0.01200706, "auxiliary_loss_mlp": 0.01032916, "balance_loss_clip": 1.0585773, "balance_loss_mlp": 1.02345073, "epoch": 0.32477604761618467, "flos": 21760864646400.0, "grad_norm": 1.6576975137903558, "language_loss": 0.74281406, "learning_rate": 3.155821910693221e-06, "loss": 0.76515031, "num_input_tokens_seen": 58147415, "step": 2701, "time_per_iteration": 3.452413558959961 }, { "auxiliary_loss_clip": 0.01168605, "auxiliary_loss_mlp": 0.01033478, "balance_loss_clip": 1.05444562, "balance_loss_mlp": 1.02472782, "epoch": 0.3248962905068238, "flos": 19828332024960.0, "grad_norm": 1.587768778446732, "language_loss": 0.85787594, "learning_rate": 3.1551861044140275e-06, "loss": 0.87989682, "num_input_tokens_seen": 58167050, "step": 2702, "time_per_iteration": 2.599151372909546 }, { "auxiliary_loss_clip": 0.01133342, "auxiliary_loss_mlp": 0.01027872, "balance_loss_clip": 1.0500493, "balance_loss_mlp": 1.01980138, "epoch": 0.3250165333974629, "flos": 23948219297760.0, "grad_norm": 1.7742049778800357, "language_loss": 0.77524984, "learning_rate": 3.15455012289402e-06, "loss": 0.79686201, "num_input_tokens_seen": 58186695, "step": 2703, "time_per_iteration": 2.6858818531036377 }, { "auxiliary_loss_clip": 0.01189561, "auxiliary_loss_mlp": 0.01035463, "balance_loss_clip": 1.0605849, "balance_loss_mlp": 1.02708268, "epoch": 0.32513677628810195, "flos": 23989337828640.0, "grad_norm": 1.6512379278159146, "language_loss": 0.83980012, "learning_rate": 3.153913966229677e-06, "loss": 0.86205035, "num_input_tokens_seen": 58205815, "step": 2704, "time_per_iteration": 2.594895601272583 }, { "auxiliary_loss_clip": 0.01079239, "auxiliary_loss_mlp": 0.0100822, "balance_loss_clip": 1.01608014, "balance_loss_mlp": 1.00663424, "epoch": 0.32525701917874106, "flos": 70655801830080.0, "grad_norm": 0.6433283450015257, "language_loss": 0.50240737, "learning_rate": 3.1532776345175027e-06, "loss": 0.52328193, "num_input_tokens_seen": 58270960, "step": 2705, "time_per_iteration": 3.1391775608062744 }, { "auxiliary_loss_clip": 0.01198526, "auxiliary_loss_mlp": 0.01028079, "balance_loss_clip": 1.05795312, "balance_loss_mlp": 1.01975846, "epoch": 0.32537726206938017, "flos": 19682642471520.0, "grad_norm": 1.792037719324409, "language_loss": 0.7893889, "learning_rate": 3.1526411278540285e-06, "loss": 0.81165498, "num_input_tokens_seen": 58289390, "step": 2706, "time_per_iteration": 2.5306761264801025 }, { "auxiliary_loss_clip": 0.01179082, "auxiliary_loss_mlp": 0.0103301, "balance_loss_clip": 1.05565906, "balance_loss_mlp": 1.02523124, "epoch": 0.3254975049600192, "flos": 28761002935680.0, "grad_norm": 2.2342981550017296, "language_loss": 0.81200415, "learning_rate": 3.1520044463358116e-06, "loss": 0.83412504, "num_input_tokens_seen": 58306120, "step": 2707, "time_per_iteration": 2.6796445846557617 }, { "auxiliary_loss_clip": 0.01182974, "auxiliary_loss_mlp": 0.01030048, "balance_loss_clip": 1.0564568, "balance_loss_mlp": 1.02169752, "epoch": 0.32561774785065833, "flos": 18877381103520.0, "grad_norm": 1.5167057762213736, "language_loss": 0.80277658, "learning_rate": 3.151367590059436e-06, "loss": 0.82490677, "num_input_tokens_seen": 58324545, "step": 2708, "time_per_iteration": 2.5629241466522217 }, { "auxiliary_loss_clip": 0.01202079, "auxiliary_loss_mlp": 0.00763096, "balance_loss_clip": 1.06028199, "balance_loss_mlp": 1.00013661, "epoch": 0.32573799074129745, "flos": 23112111649440.0, "grad_norm": 1.9961560564229701, "language_loss": 0.87115204, "learning_rate": 3.1507305591215117e-06, "loss": 0.89080381, "num_input_tokens_seen": 58342455, "step": 2709, "time_per_iteration": 2.5695242881774902 }, { "auxiliary_loss_clip": 0.01077831, "auxiliary_loss_mlp": 0.01003832, "balance_loss_clip": 1.01642752, "balance_loss_mlp": 1.00213897, "epoch": 0.3258582336319365, "flos": 71237662119840.0, "grad_norm": 0.669772569701645, "language_loss": 0.55686998, "learning_rate": 3.150093353618677e-06, "loss": 0.57768667, "num_input_tokens_seen": 58407185, "step": 2710, "time_per_iteration": 3.2043068408966064 }, { "auxiliary_loss_clip": 0.01192402, "auxiliary_loss_mlp": 0.01028142, "balance_loss_clip": 1.0576272, "balance_loss_mlp": 1.0200237, "epoch": 0.3259784765225756, "flos": 22456023780000.0, "grad_norm": 2.2980864774133343, "language_loss": 0.88176203, "learning_rate": 3.149455973647596e-06, "loss": 0.9039675, "num_input_tokens_seen": 58425245, "step": 2711, "time_per_iteration": 2.5792856216430664 }, { "auxiliary_loss_clip": 0.01146269, "auxiliary_loss_mlp": 0.01036515, "balance_loss_clip": 1.04936564, "balance_loss_mlp": 1.02733636, "epoch": 0.32609871941321467, "flos": 20484814980960.0, "grad_norm": 1.934780642657707, "language_loss": 0.76930451, "learning_rate": 3.1488184193049563e-06, "loss": 0.79113233, "num_input_tokens_seen": 58444780, "step": 2712, "time_per_iteration": 2.617403984069824 }, { "auxiliary_loss_clip": 0.0120026, "auxiliary_loss_mlp": 0.01027965, "balance_loss_clip": 1.05974627, "balance_loss_mlp": 1.01982903, "epoch": 0.3262189623038538, "flos": 22416808848000.0, "grad_norm": 1.956749670530742, "language_loss": 0.72025168, "learning_rate": 3.1481806906874767e-06, "loss": 0.74253392, "num_input_tokens_seen": 58466090, "step": 2713, "time_per_iteration": 2.5713980197906494 }, { "auxiliary_loss_clip": 0.01198064, "auxiliary_loss_mlp": 0.01024892, "balance_loss_clip": 1.05913806, "balance_loss_mlp": 1.01666069, "epoch": 0.3263392051944929, "flos": 20923499904480.0, "grad_norm": 1.5632925397879804, "language_loss": 0.87562567, "learning_rate": 3.147542787891899e-06, "loss": 0.89785522, "num_input_tokens_seen": 58485435, "step": 2714, "time_per_iteration": 2.535954475402832 }, { "auxiliary_loss_clip": 0.01171049, "auxiliary_loss_mlp": 0.01033605, "balance_loss_clip": 1.05604494, "balance_loss_mlp": 1.02552307, "epoch": 0.32645944808513194, "flos": 24025176566400.0, "grad_norm": 2.021497579033049, "language_loss": 0.75037545, "learning_rate": 3.1469047110149926e-06, "loss": 0.77242196, "num_input_tokens_seen": 58504175, "step": 2715, "time_per_iteration": 2.6208207607269287 }, { "auxiliary_loss_clip": 0.01134686, "auxiliary_loss_mlp": 0.01025941, "balance_loss_clip": 1.05176473, "balance_loss_mlp": 1.01695228, "epoch": 0.32657969097577105, "flos": 21032416879200.0, "grad_norm": 2.0121165135036896, "language_loss": 0.85025394, "learning_rate": 3.146266460153554e-06, "loss": 0.87186015, "num_input_tokens_seen": 58523885, "step": 2716, "time_per_iteration": 2.666466236114502 }, { "auxiliary_loss_clip": 0.01169847, "auxiliary_loss_mlp": 0.00764384, "balance_loss_clip": 1.05598688, "balance_loss_mlp": 1.00020325, "epoch": 0.32669993386641016, "flos": 22710271138560.0, "grad_norm": 1.8693060243137862, "language_loss": 0.80090356, "learning_rate": 3.145628035404404e-06, "loss": 0.82024592, "num_input_tokens_seen": 58543085, "step": 2717, "time_per_iteration": 2.6205992698669434 }, { "auxiliary_loss_clip": 0.01075383, "auxiliary_loss_mlp": 0.01002025, "balance_loss_clip": 1.0156641, "balance_loss_mlp": 1.00043988, "epoch": 0.3268201767570492, "flos": 72105729474240.0, "grad_norm": 0.8740239984700394, "language_loss": 0.57558119, "learning_rate": 3.1449894368643922e-06, "loss": 0.5963552, "num_input_tokens_seen": 58605400, "step": 2718, "time_per_iteration": 3.2121353149414062 }, { "auxiliary_loss_clip": 0.01157243, "auxiliary_loss_mlp": 0.01035267, "balance_loss_clip": 1.05721331, "balance_loss_mlp": 1.02735758, "epoch": 0.32694041964768833, "flos": 24535179795840.0, "grad_norm": 1.6523551935286203, "language_loss": 0.71049011, "learning_rate": 3.1443506646303934e-06, "loss": 0.7324152, "num_input_tokens_seen": 58626700, "step": 2719, "time_per_iteration": 3.57454776763916 }, { "auxiliary_loss_clip": 0.01187001, "auxiliary_loss_mlp": 0.01027133, "balance_loss_clip": 1.05574822, "balance_loss_mlp": 1.01829386, "epoch": 0.32706066253832744, "flos": 33183008486400.0, "grad_norm": 2.075494423891663, "language_loss": 0.67118454, "learning_rate": 3.1437117187993086e-06, "loss": 0.69332582, "num_input_tokens_seen": 58649020, "step": 2720, "time_per_iteration": 2.6446993350982666 }, { "auxiliary_loss_clip": 0.01150344, "auxiliary_loss_mlp": 0.01024457, "balance_loss_clip": 1.0524106, "balance_loss_mlp": 1.01607656, "epoch": 0.3271809054289665, "flos": 24061625892480.0, "grad_norm": 1.589009500952305, "language_loss": 0.79976809, "learning_rate": 3.143072599468065e-06, "loss": 0.8215161, "num_input_tokens_seen": 58668845, "step": 2721, "time_per_iteration": 3.488759756088257 }, { "auxiliary_loss_clip": 0.01170913, "auxiliary_loss_mlp": 0.01032966, "balance_loss_clip": 1.05739427, "balance_loss_mlp": 1.02420437, "epoch": 0.3273011483196056, "flos": 38253774846720.0, "grad_norm": 1.4902047830073195, "language_loss": 0.7542752, "learning_rate": 3.1424333067336174e-06, "loss": 0.77631402, "num_input_tokens_seen": 58691610, "step": 2722, "time_per_iteration": 2.681272506713867 }, { "auxiliary_loss_clip": 0.01188303, "auxiliary_loss_mlp": 0.0102889, "balance_loss_clip": 1.05567026, "balance_loss_mlp": 1.02005029, "epoch": 0.3274213912102447, "flos": 29054393392320.0, "grad_norm": 1.6507252402111467, "language_loss": 0.77828747, "learning_rate": 3.141793840692945e-06, "loss": 0.80045938, "num_input_tokens_seen": 58712360, "step": 2723, "time_per_iteration": 2.59989333152771 }, { "auxiliary_loss_clip": 0.01162079, "auxiliary_loss_mlp": 0.01034003, "balance_loss_clip": 1.05406678, "balance_loss_mlp": 1.02513337, "epoch": 0.32754163410088377, "flos": 29133254259840.0, "grad_norm": 2.0771328865777376, "language_loss": 0.61540592, "learning_rate": 3.1411542014430553e-06, "loss": 0.63736677, "num_input_tokens_seen": 58733440, "step": 2724, "time_per_iteration": 2.592841863632202 }, { "auxiliary_loss_clip": 0.01155268, "auxiliary_loss_mlp": 0.01032494, "balance_loss_clip": 1.05155158, "balance_loss_mlp": 1.02417874, "epoch": 0.3276618769915229, "flos": 20631079205760.0, "grad_norm": 1.6684325606938704, "language_loss": 0.81750304, "learning_rate": 3.1405143890809804e-06, "loss": 0.83938068, "num_input_tokens_seen": 58752735, "step": 2725, "time_per_iteration": 2.612438440322876 }, { "auxiliary_loss_clip": 0.01167318, "auxiliary_loss_mlp": 0.01028341, "balance_loss_clip": 1.05490398, "balance_loss_mlp": 1.02024698, "epoch": 0.327782119882162, "flos": 18657428053440.0, "grad_norm": 1.642269106821336, "language_loss": 0.6970157, "learning_rate": 3.1398744037037796e-06, "loss": 0.71897233, "num_input_tokens_seen": 58772070, "step": 2726, "time_per_iteration": 2.5743296146392822 }, { "auxiliary_loss_clip": 0.01170764, "auxiliary_loss_mlp": 0.01028642, "balance_loss_clip": 1.05690384, "balance_loss_mlp": 1.0204289, "epoch": 0.32790236277280105, "flos": 21795805460160.0, "grad_norm": 2.162980002925623, "language_loss": 0.84307075, "learning_rate": 3.139234245408538e-06, "loss": 0.86506486, "num_input_tokens_seen": 58790950, "step": 2727, "time_per_iteration": 3.313966751098633 }, { "auxiliary_loss_clip": 0.01154257, "auxiliary_loss_mlp": 0.00762638, "balance_loss_clip": 1.05294979, "balance_loss_mlp": 1.00018203, "epoch": 0.32802260566344016, "flos": 23331418194240.0, "grad_norm": 1.3750358674830225, "language_loss": 0.7599113, "learning_rate": 3.1385939142923666e-06, "loss": 0.77908027, "num_input_tokens_seen": 58813340, "step": 2728, "time_per_iteration": 2.624997138977051 }, { "auxiliary_loss_clip": 0.01169928, "auxiliary_loss_mlp": 0.0103539, "balance_loss_clip": 1.05270779, "balance_loss_mlp": 1.02641964, "epoch": 0.3281428485540792, "flos": 24206992193280.0, "grad_norm": 2.4379979392164848, "language_loss": 0.78219777, "learning_rate": 3.137953410452405e-06, "loss": 0.80425096, "num_input_tokens_seen": 58833610, "step": 2729, "time_per_iteration": 2.5734403133392334 }, { "auxiliary_loss_clip": 0.01168849, "auxiliary_loss_mlp": 0.01032257, "balance_loss_clip": 1.05500436, "balance_loss_mlp": 1.02450895, "epoch": 0.3282630914447183, "flos": 34128966950400.0, "grad_norm": 3.4051942928220593, "language_loss": 0.74426401, "learning_rate": 3.1373127339858146e-06, "loss": 0.76627511, "num_input_tokens_seen": 58856210, "step": 2730, "time_per_iteration": 2.6568260192871094 }, { "auxiliary_loss_clip": 0.0115105, "auxiliary_loss_mlp": 0.01033123, "balance_loss_clip": 1.05233788, "balance_loss_mlp": 1.02566028, "epoch": 0.32838333433535744, "flos": 27600730384320.0, "grad_norm": 1.8128944999383658, "language_loss": 0.74402058, "learning_rate": 3.136671884989787e-06, "loss": 0.76586223, "num_input_tokens_seen": 58876120, "step": 2731, "time_per_iteration": 2.644049644470215 }, { "auxiliary_loss_clip": 0.01129364, "auxiliary_loss_mlp": 0.01031571, "balance_loss_clip": 1.05184448, "balance_loss_mlp": 1.02275586, "epoch": 0.3285035772259965, "flos": 12349503707040.0, "grad_norm": 2.248003663062854, "language_loss": 0.86944795, "learning_rate": 3.1360308635615383e-06, "loss": 0.89105731, "num_input_tokens_seen": 58894660, "step": 2732, "time_per_iteration": 2.6346631050109863 }, { "auxiliary_loss_clip": 0.01179603, "auxiliary_loss_mlp": 0.01031115, "balance_loss_clip": 1.05828702, "balance_loss_mlp": 1.02254343, "epoch": 0.3286238201166356, "flos": 24316088752800.0, "grad_norm": 9.318128869913316, "language_loss": 0.78479552, "learning_rate": 3.135389669798311e-06, "loss": 0.80690265, "num_input_tokens_seen": 58912720, "step": 2733, "time_per_iteration": 2.5948891639709473 }, { "auxiliary_loss_clip": 0.0118378, "auxiliary_loss_mlp": 0.0076348, "balance_loss_clip": 1.05570424, "balance_loss_mlp": 1.00014639, "epoch": 0.3287440630072747, "flos": 21392815606560.0, "grad_norm": 2.6426341308218286, "language_loss": 0.79475754, "learning_rate": 3.134748303797373e-06, "loss": 0.81423014, "num_input_tokens_seen": 58930090, "step": 2734, "time_per_iteration": 2.5671398639678955 }, { "auxiliary_loss_clip": 0.01139403, "auxiliary_loss_mlp": 0.01028064, "balance_loss_clip": 1.05134153, "balance_loss_mlp": 1.01897478, "epoch": 0.32886430589791377, "flos": 23732540365920.0, "grad_norm": 2.026868219440972, "language_loss": 0.81063807, "learning_rate": 3.1341067656560203e-06, "loss": 0.83231282, "num_input_tokens_seen": 58947935, "step": 2735, "time_per_iteration": 2.6397271156311035 }, { "auxiliary_loss_clip": 0.01179451, "auxiliary_loss_mlp": 0.01027076, "balance_loss_clip": 1.05507267, "balance_loss_mlp": 1.0188092, "epoch": 0.3289845487885529, "flos": 22418712446880.0, "grad_norm": 1.8306735112891743, "language_loss": 0.86222005, "learning_rate": 3.133465055471572e-06, "loss": 0.88428533, "num_input_tokens_seen": 58967720, "step": 2736, "time_per_iteration": 2.589859962463379 }, { "auxiliary_loss_clip": 0.01149713, "auxiliary_loss_mlp": 0.01028805, "balance_loss_clip": 1.05246925, "balance_loss_mlp": 1.02016211, "epoch": 0.329104791679192, "flos": 19682606554560.0, "grad_norm": 2.14023217411318, "language_loss": 0.65996957, "learning_rate": 3.1328231733413767e-06, "loss": 0.68175471, "num_input_tokens_seen": 58984360, "step": 2737, "time_per_iteration": 2.561321258544922 }, { "auxiliary_loss_clip": 0.01178668, "auxiliary_loss_mlp": 0.01033777, "balance_loss_clip": 1.05417562, "balance_loss_mlp": 1.025105, "epoch": 0.32922503456983104, "flos": 15997237837920.0, "grad_norm": 2.061546546644953, "language_loss": 0.9091022, "learning_rate": 3.1321811193628067e-06, "loss": 0.93122667, "num_input_tokens_seen": 59002505, "step": 2738, "time_per_iteration": 2.528158664703369 }, { "auxiliary_loss_clip": 0.01183941, "auxiliary_loss_mlp": 0.00763637, "balance_loss_clip": 1.05709147, "balance_loss_mlp": 1.00015497, "epoch": 0.32934527746047015, "flos": 26834073360000.0, "grad_norm": 1.7211900394744402, "language_loss": 0.70267349, "learning_rate": 3.131538893633261e-06, "loss": 0.72214925, "num_input_tokens_seen": 59022065, "step": 2739, "time_per_iteration": 2.5489261150360107 }, { "auxiliary_loss_clip": 0.0120001, "auxiliary_loss_mlp": 0.01030952, "balance_loss_clip": 1.05920315, "balance_loss_mlp": 1.02293551, "epoch": 0.32946552035110926, "flos": 23403778092000.0, "grad_norm": 2.0825337359402534, "language_loss": 0.78207034, "learning_rate": 3.130896496250165e-06, "loss": 0.80437994, "num_input_tokens_seen": 59041890, "step": 2740, "time_per_iteration": 2.502260684967041 }, { "auxiliary_loss_clip": 0.01200195, "auxiliary_loss_mlp": 0.01026526, "balance_loss_clip": 1.05812049, "balance_loss_mlp": 1.01821101, "epoch": 0.3295857632417483, "flos": 14172472848480.0, "grad_norm": 2.9809754591214985, "language_loss": 0.86865604, "learning_rate": 3.1302539273109693e-06, "loss": 0.89092326, "num_input_tokens_seen": 59058715, "step": 2741, "time_per_iteration": 2.4526467323303223 }, { "auxiliary_loss_clip": 0.01163238, "auxiliary_loss_mlp": 0.01031861, "balance_loss_clip": 1.05488837, "balance_loss_mlp": 1.02346301, "epoch": 0.32970600613238743, "flos": 22196712130080.0, "grad_norm": 1.7274166363490968, "language_loss": 0.80558407, "learning_rate": 3.1296111869131513e-06, "loss": 0.82753509, "num_input_tokens_seen": 59076140, "step": 2742, "time_per_iteration": 2.5656967163085938 }, { "auxiliary_loss_clip": 0.0119803, "auxiliary_loss_mlp": 0.01026728, "balance_loss_clip": 1.05727077, "balance_loss_mlp": 1.01844335, "epoch": 0.32982624902302654, "flos": 22053788182560.0, "grad_norm": 1.9681702831278622, "language_loss": 0.85735351, "learning_rate": 3.1289682751542153e-06, "loss": 0.87960112, "num_input_tokens_seen": 59095700, "step": 2743, "time_per_iteration": 2.4891462326049805 }, { "auxiliary_loss_clip": 0.01184238, "auxiliary_loss_mlp": 0.01027682, "balance_loss_clip": 1.05638289, "balance_loss_mlp": 1.01951587, "epoch": 0.3299464919136656, "flos": 18661630337760.0, "grad_norm": 1.9339425683986864, "language_loss": 0.70921624, "learning_rate": 3.1283251921316883e-06, "loss": 0.7313354, "num_input_tokens_seen": 59113445, "step": 2744, "time_per_iteration": 2.530313730239868 }, { "auxiliary_loss_clip": 0.01139949, "auxiliary_loss_mlp": 0.01029885, "balance_loss_clip": 1.05505037, "balance_loss_mlp": 1.0209142, "epoch": 0.3300667348043047, "flos": 13407360253440.0, "grad_norm": 2.031866108686088, "language_loss": 0.80744934, "learning_rate": 3.1276819379431277e-06, "loss": 0.82914764, "num_input_tokens_seen": 59131535, "step": 2745, "time_per_iteration": 3.459505796432495 }, { "auxiliary_loss_clip": 0.01177477, "auxiliary_loss_mlp": 0.00763843, "balance_loss_clip": 1.05659842, "balance_loss_mlp": 1.00007749, "epoch": 0.33018697769494376, "flos": 15742559475840.0, "grad_norm": 1.8175331252604243, "language_loss": 0.75022686, "learning_rate": 3.1270385126861134e-06, "loss": 0.76964003, "num_input_tokens_seen": 59149520, "step": 2746, "time_per_iteration": 2.5372025966644287 }, { "auxiliary_loss_clip": 0.0119906, "auxiliary_loss_mlp": 0.01028006, "balance_loss_clip": 1.05719268, "balance_loss_mlp": 1.01970911, "epoch": 0.3303072205855829, "flos": 18258604567200.0, "grad_norm": 2.0397658208301297, "language_loss": 0.8202492, "learning_rate": 3.1263949164582533e-06, "loss": 0.84251982, "num_input_tokens_seen": 59169170, "step": 2747, "time_per_iteration": 3.1685826778411865 }, { "auxiliary_loss_clip": 0.01196999, "auxiliary_loss_mlp": 0.01028044, "balance_loss_clip": 1.0549134, "balance_loss_mlp": 1.01921058, "epoch": 0.330427463476222, "flos": 17749427427840.0, "grad_norm": 2.2776418546693282, "language_loss": 0.78106701, "learning_rate": 3.1257511493571797e-06, "loss": 0.80331749, "num_input_tokens_seen": 59187675, "step": 2748, "time_per_iteration": 2.4930593967437744 }, { "auxiliary_loss_clip": 0.01154249, "auxiliary_loss_mlp": 0.01027566, "balance_loss_clip": 1.05415392, "balance_loss_mlp": 1.01960933, "epoch": 0.33054770636686104, "flos": 27162584215200.0, "grad_norm": 2.100098209663406, "language_loss": 0.78151143, "learning_rate": 3.125107211480552e-06, "loss": 0.80332959, "num_input_tokens_seen": 59207610, "step": 2749, "time_per_iteration": 2.648866653442383 }, { "auxiliary_loss_clip": 0.01122382, "auxiliary_loss_mlp": 0.01030304, "balance_loss_clip": 1.04923737, "balance_loss_mlp": 1.02220416, "epoch": 0.33066794925750015, "flos": 20117196944640.0, "grad_norm": 1.6628811544091668, "language_loss": 0.79857188, "learning_rate": 3.124463102926054e-06, "loss": 0.82009876, "num_input_tokens_seen": 59226945, "step": 2750, "time_per_iteration": 2.6479227542877197 }, { "auxiliary_loss_clip": 0.01071611, "auxiliary_loss_mlp": 0.0100114, "balance_loss_clip": 1.01533675, "balance_loss_mlp": 0.99938786, "epoch": 0.33078819214813926, "flos": 70642620305760.0, "grad_norm": 0.7598086122300297, "language_loss": 0.61616576, "learning_rate": 3.1238188237913984e-06, "loss": 0.63689327, "num_input_tokens_seen": 59291485, "step": 2751, "time_per_iteration": 3.1918013095855713 }, { "auxiliary_loss_clip": 0.01204484, "auxiliary_loss_mlp": 0.01034831, "balance_loss_clip": 1.05976963, "balance_loss_mlp": 1.02593207, "epoch": 0.3309084350387783, "flos": 21141944442240.0, "grad_norm": 1.9073671055477006, "language_loss": 0.75774193, "learning_rate": 3.1231743741743202e-06, "loss": 0.7801351, "num_input_tokens_seen": 59310990, "step": 2752, "time_per_iteration": 2.4718375205993652 }, { "auxiliary_loss_clip": 0.01180659, "auxiliary_loss_mlp": 0.01031007, "balance_loss_clip": 1.05399776, "balance_loss_mlp": 1.02229834, "epoch": 0.3310286779294174, "flos": 14209353178080.0, "grad_norm": 2.2406304824142977, "language_loss": 0.83211964, "learning_rate": 3.122529754172582e-06, "loss": 0.85423636, "num_input_tokens_seen": 59327875, "step": 2753, "time_per_iteration": 3.2682697772979736 }, { "auxiliary_loss_clip": 0.01182245, "auxiliary_loss_mlp": 0.01038818, "balance_loss_clip": 1.05534601, "balance_loss_mlp": 1.0300144, "epoch": 0.33114892082005654, "flos": 20778133603680.0, "grad_norm": 2.0477538632193353, "language_loss": 0.72648549, "learning_rate": 3.1218849638839736e-06, "loss": 0.74869621, "num_input_tokens_seen": 59347135, "step": 2754, "time_per_iteration": 2.548611879348755 }, { "auxiliary_loss_clip": 0.01142515, "auxiliary_loss_mlp": 0.01035814, "balance_loss_clip": 1.04742205, "balance_loss_mlp": 1.02692688, "epoch": 0.3312691637106956, "flos": 17090753537280.0, "grad_norm": 1.6842045832038597, "language_loss": 0.78276473, "learning_rate": 3.121240003406307e-06, "loss": 0.80454803, "num_input_tokens_seen": 59365985, "step": 2755, "time_per_iteration": 2.5733449459075928 }, { "auxiliary_loss_clip": 0.01161174, "auxiliary_loss_mlp": 0.01035628, "balance_loss_clip": 1.05652177, "balance_loss_mlp": 1.02643681, "epoch": 0.3313894066013347, "flos": 29456233903200.0, "grad_norm": 2.032049072945021, "language_loss": 0.7274363, "learning_rate": 3.120594872837425e-06, "loss": 0.74940437, "num_input_tokens_seen": 59384655, "step": 2756, "time_per_iteration": 2.6164462566375732 }, { "auxiliary_loss_clip": 0.0107428, "auxiliary_loss_mlp": 0.00754497, "balance_loss_clip": 1.01533341, "balance_loss_mlp": 0.99975675, "epoch": 0.3315096494919738, "flos": 61419252710400.0, "grad_norm": 0.8307159380860872, "language_loss": 0.62416416, "learning_rate": 3.1199495722751906e-06, "loss": 0.64245194, "num_input_tokens_seen": 59444185, "step": 2757, "time_per_iteration": 3.1381025314331055 }, { "auxiliary_loss_clip": 0.01144043, "auxiliary_loss_mlp": 0.01029229, "balance_loss_clip": 1.05079317, "balance_loss_mlp": 1.02087831, "epoch": 0.33162989238261287, "flos": 21653061097440.0, "grad_norm": 1.5995304891625777, "language_loss": 0.83637547, "learning_rate": 3.1193041018174972e-06, "loss": 0.85810816, "num_input_tokens_seen": 59464900, "step": 2758, "time_per_iteration": 2.6254963874816895 }, { "auxiliary_loss_clip": 0.01191323, "auxiliary_loss_mlp": 0.01031951, "balance_loss_clip": 1.05998015, "balance_loss_mlp": 1.02298677, "epoch": 0.331750135273252, "flos": 22674791570400.0, "grad_norm": 1.9095377956427895, "language_loss": 0.94225013, "learning_rate": 3.118658461562261e-06, "loss": 0.9644829, "num_input_tokens_seen": 59481000, "step": 2759, "time_per_iteration": 2.5493669509887695 }, { "auxiliary_loss_clip": 0.01168965, "auxiliary_loss_mlp": 0.01031041, "balance_loss_clip": 1.05582547, "balance_loss_mlp": 1.02227938, "epoch": 0.33187037816389103, "flos": 22746900049440.0, "grad_norm": 2.9989775692863128, "language_loss": 0.85007012, "learning_rate": 3.118012651607426e-06, "loss": 0.87207019, "num_input_tokens_seen": 59502605, "step": 2760, "time_per_iteration": 2.593226671218872 }, { "auxiliary_loss_clip": 0.01200611, "auxiliary_loss_mlp": 0.01030601, "balance_loss_clip": 1.05940413, "balance_loss_mlp": 1.02185678, "epoch": 0.33199062105453014, "flos": 19203772858080.0, "grad_norm": 2.069519710162404, "language_loss": 0.84038579, "learning_rate": 3.1173666720509603e-06, "loss": 0.86269796, "num_input_tokens_seen": 59519540, "step": 2761, "time_per_iteration": 2.470557451248169 }, { "auxiliary_loss_clip": 0.01177002, "auxiliary_loss_mlp": 0.01032555, "balance_loss_clip": 1.05640364, "balance_loss_mlp": 1.02351284, "epoch": 0.33211086394516925, "flos": 31577011287360.0, "grad_norm": 1.643618575501026, "language_loss": 0.68263328, "learning_rate": 3.116720522990859e-06, "loss": 0.70472896, "num_input_tokens_seen": 59540415, "step": 2762, "time_per_iteration": 2.643327474594116 }, { "auxiliary_loss_clip": 0.01126853, "auxiliary_loss_mlp": 0.01030487, "balance_loss_clip": 1.05092597, "balance_loss_mlp": 1.02195811, "epoch": 0.3322311068358083, "flos": 17932500148320.0, "grad_norm": 2.52334386154505, "language_loss": 0.62203079, "learning_rate": 3.116074204525142e-06, "loss": 0.64360416, "num_input_tokens_seen": 59558590, "step": 2763, "time_per_iteration": 2.625077247619629 }, { "auxiliary_loss_clip": 0.01175259, "auxiliary_loss_mlp": 0.01033556, "balance_loss_clip": 1.05452538, "balance_loss_mlp": 1.02493775, "epoch": 0.3323513497264474, "flos": 32269835818560.0, "grad_norm": 1.5390407330532483, "language_loss": 0.83608395, "learning_rate": 3.1154277167518553e-06, "loss": 0.85817212, "num_input_tokens_seen": 59580205, "step": 2764, "time_per_iteration": 2.617722988128662 }, { "auxiliary_loss_clip": 0.01059831, "auxiliary_loss_mlp": 0.01002701, "balance_loss_clip": 1.01617432, "balance_loss_mlp": 1.0010922, "epoch": 0.33247159261708653, "flos": 52668684762240.0, "grad_norm": 0.7845219102058917, "language_loss": 0.59525615, "learning_rate": 3.114781059769072e-06, "loss": 0.61588144, "num_input_tokens_seen": 59631530, "step": 2765, "time_per_iteration": 3.0003435611724854 }, { "auxiliary_loss_clip": 0.01168621, "auxiliary_loss_mlp": 0.01031534, "balance_loss_clip": 1.05394566, "balance_loss_mlp": 1.02305818, "epoch": 0.3325918355077256, "flos": 27125236965120.0, "grad_norm": 2.3615966261381187, "language_loss": 0.67609882, "learning_rate": 3.1141342336748874e-06, "loss": 0.69810039, "num_input_tokens_seen": 59651090, "step": 2766, "time_per_iteration": 2.5988595485687256 }, { "auxiliary_loss_clip": 0.01184784, "auxiliary_loss_mlp": 0.01030483, "balance_loss_clip": 1.05896282, "balance_loss_mlp": 1.02231121, "epoch": 0.3327120783983647, "flos": 23664418669440.0, "grad_norm": 1.4786522008148522, "language_loss": 0.82250071, "learning_rate": 3.1134872385674253e-06, "loss": 0.84465343, "num_input_tokens_seen": 59675245, "step": 2767, "time_per_iteration": 2.5858306884765625 }, { "auxiliary_loss_clip": 0.0117011, "auxiliary_loss_mlp": 0.01033937, "balance_loss_clip": 1.05116415, "balance_loss_mlp": 1.02562785, "epoch": 0.3328323212890038, "flos": 19171382148480.0, "grad_norm": 2.893162944808083, "language_loss": 0.85748881, "learning_rate": 3.1128400745448353e-06, "loss": 0.87952924, "num_input_tokens_seen": 59694625, "step": 2768, "time_per_iteration": 2.599024772644043 }, { "auxiliary_loss_clip": 0.01188353, "auxiliary_loss_mlp": 0.01031354, "balance_loss_clip": 1.05872846, "balance_loss_mlp": 1.0230453, "epoch": 0.33295256417964286, "flos": 37706352533280.0, "grad_norm": 2.5320500980432175, "language_loss": 0.63267958, "learning_rate": 3.11219274170529e-06, "loss": 0.65487665, "num_input_tokens_seen": 59716435, "step": 2769, "time_per_iteration": 2.6540067195892334 }, { "auxiliary_loss_clip": 0.01163317, "auxiliary_loss_mlp": 0.01030858, "balance_loss_clip": 1.05284357, "balance_loss_mlp": 1.02297258, "epoch": 0.333072807070282, "flos": 26505993508320.0, "grad_norm": 1.774899070606751, "language_loss": 0.81766284, "learning_rate": 3.1115452401469903e-06, "loss": 0.83960456, "num_input_tokens_seen": 59736835, "step": 2770, "time_per_iteration": 2.6054422855377197 }, { "auxiliary_loss_clip": 0.01132586, "auxiliary_loss_mlp": 0.01034009, "balance_loss_clip": 1.04778051, "balance_loss_mlp": 1.02585542, "epoch": 0.3331930499609211, "flos": 21430917112800.0, "grad_norm": 1.7891015363899279, "language_loss": 0.86399263, "learning_rate": 3.1108975699681613e-06, "loss": 0.88565862, "num_input_tokens_seen": 59754230, "step": 2771, "time_per_iteration": 4.122169256210327 }, { "auxiliary_loss_clip": 0.01150637, "auxiliary_loss_mlp": 0.01034348, "balance_loss_clip": 1.05002666, "balance_loss_mlp": 1.02654552, "epoch": 0.33331329285156014, "flos": 20659950053280.0, "grad_norm": 1.8411630776819836, "language_loss": 0.71749395, "learning_rate": 3.1102497312670542e-06, "loss": 0.73934376, "num_input_tokens_seen": 59772235, "step": 2772, "time_per_iteration": 2.5702712535858154 }, { "auxiliary_loss_clip": 0.01156545, "auxiliary_loss_mlp": 0.01034759, "balance_loss_clip": 1.05115974, "balance_loss_mlp": 1.02665532, "epoch": 0.33343353574219925, "flos": 28001601137280.0, "grad_norm": 1.69771374411847, "language_loss": 0.8034153, "learning_rate": 3.109601724141946e-06, "loss": 0.82532835, "num_input_tokens_seen": 59791230, "step": 2773, "time_per_iteration": 3.384336233139038 }, { "auxiliary_loss_clip": 0.01164549, "auxiliary_loss_mlp": 0.01033008, "balance_loss_clip": 1.05351055, "balance_loss_mlp": 1.02472878, "epoch": 0.33355377863283836, "flos": 23764966992480.0, "grad_norm": 1.7956802976135886, "language_loss": 0.6812681, "learning_rate": 3.108953548691138e-06, "loss": 0.70324367, "num_input_tokens_seen": 59811315, "step": 2774, "time_per_iteration": 2.564302444458008 }, { "auxiliary_loss_clip": 0.01199017, "auxiliary_loss_mlp": 0.01036296, "balance_loss_clip": 1.0580194, "balance_loss_mlp": 1.02787352, "epoch": 0.3336740215234774, "flos": 37779682188960.0, "grad_norm": 2.3172865233922595, "language_loss": 0.72693884, "learning_rate": 3.108305205012959e-06, "loss": 0.74929196, "num_input_tokens_seen": 59832010, "step": 2775, "time_per_iteration": 2.611818790435791 }, { "auxiliary_loss_clip": 0.01166635, "auxiliary_loss_mlp": 0.01026564, "balance_loss_clip": 1.05321753, "balance_loss_mlp": 1.01895893, "epoch": 0.3337942644141165, "flos": 25519060181280.0, "grad_norm": 1.9522539049787289, "language_loss": 0.87356645, "learning_rate": 3.107656693205761e-06, "loss": 0.89549839, "num_input_tokens_seen": 59851450, "step": 2776, "time_per_iteration": 2.5592455863952637 }, { "auxiliary_loss_clip": 0.01203963, "auxiliary_loss_mlp": 0.01030627, "balance_loss_clip": 1.05918407, "balance_loss_mlp": 1.02155471, "epoch": 0.3339145073047556, "flos": 25989848478720.0, "grad_norm": 3.1709970732107076, "language_loss": 0.70389843, "learning_rate": 3.107008013367924e-06, "loss": 0.72624439, "num_input_tokens_seen": 59870245, "step": 2777, "time_per_iteration": 2.4982669353485107 }, { "auxiliary_loss_clip": 0.01152823, "auxiliary_loss_mlp": 0.01031933, "balance_loss_clip": 1.05256343, "balance_loss_mlp": 1.02333856, "epoch": 0.3340347501953947, "flos": 19062572924640.0, "grad_norm": 5.912820468656039, "language_loss": 0.86386371, "learning_rate": 3.1063591655978507e-06, "loss": 0.88571125, "num_input_tokens_seen": 59886195, "step": 2778, "time_per_iteration": 2.546560049057007 }, { "auxiliary_loss_clip": 0.01127464, "auxiliary_loss_mlp": 0.01029976, "balance_loss_clip": 1.04653096, "balance_loss_mlp": 1.02148247, "epoch": 0.3341549930860338, "flos": 18109718404320.0, "grad_norm": 1.9340382412581338, "language_loss": 0.79549551, "learning_rate": 3.105710149993972e-06, "loss": 0.81706989, "num_input_tokens_seen": 59905525, "step": 2779, "time_per_iteration": 3.344407320022583 }, { "auxiliary_loss_clip": 0.01204519, "auxiliary_loss_mlp": 0.01033946, "balance_loss_clip": 1.0621047, "balance_loss_mlp": 1.02544618, "epoch": 0.33427523597667286, "flos": 22674971155200.0, "grad_norm": 1.8474427467109678, "language_loss": 0.85184014, "learning_rate": 3.1050609666547427e-06, "loss": 0.87422478, "num_input_tokens_seen": 59925085, "step": 2780, "time_per_iteration": 2.5161821842193604 }, { "auxiliary_loss_clip": 0.0116138, "auxiliary_loss_mlp": 0.01027428, "balance_loss_clip": 1.05418134, "balance_loss_mlp": 1.01929784, "epoch": 0.33439547886731197, "flos": 22638342244320.0, "grad_norm": 1.8149975501751332, "language_loss": 0.7694993, "learning_rate": 3.104411615678644e-06, "loss": 0.79138732, "num_input_tokens_seen": 59943935, "step": 2781, "time_per_iteration": 2.582747220993042 }, { "auxiliary_loss_clip": 0.0116377, "auxiliary_loss_mlp": 0.01035878, "balance_loss_clip": 1.05346143, "balance_loss_mlp": 1.02721107, "epoch": 0.3345157217579511, "flos": 24096387121440.0, "grad_norm": 2.6458463877666882, "language_loss": 0.73682684, "learning_rate": 3.1037620971641803e-06, "loss": 0.75882328, "num_input_tokens_seen": 59963725, "step": 2782, "time_per_iteration": 2.587859630584717 }, { "auxiliary_loss_clip": 0.01198271, "auxiliary_loss_mlp": 0.01031735, "balance_loss_clip": 1.05781245, "balance_loss_mlp": 1.02319407, "epoch": 0.33463596464859013, "flos": 18989494687680.0, "grad_norm": 2.2829748621592527, "language_loss": 0.64895535, "learning_rate": 3.1031124112098844e-06, "loss": 0.67125541, "num_input_tokens_seen": 59981935, "step": 2783, "time_per_iteration": 2.4894375801086426 }, { "auxiliary_loss_clip": 0.01173804, "auxiliary_loss_mlp": 0.01036699, "balance_loss_clip": 1.05667245, "balance_loss_mlp": 1.02869415, "epoch": 0.33475620753922924, "flos": 20375610670560.0, "grad_norm": 1.9289798281570327, "language_loss": 0.71834004, "learning_rate": 3.1024625579143127e-06, "loss": 0.74044514, "num_input_tokens_seen": 59999455, "step": 2784, "time_per_iteration": 2.5770628452301025 }, { "auxiliary_loss_clip": 0.01195177, "auxiliary_loss_mlp": 0.01037006, "balance_loss_clip": 1.05535579, "balance_loss_mlp": 1.02849782, "epoch": 0.33487645042986836, "flos": 18182581139520.0, "grad_norm": 1.8280955750891996, "language_loss": 0.73059916, "learning_rate": 3.101812537376048e-06, "loss": 0.75292099, "num_input_tokens_seen": 60018475, "step": 2785, "time_per_iteration": 2.4886929988861084 }, { "auxiliary_loss_clip": 0.01160468, "auxiliary_loss_mlp": 0.00763415, "balance_loss_clip": 1.05414367, "balance_loss_mlp": 1.00005424, "epoch": 0.3349966933205074, "flos": 25848828130080.0, "grad_norm": 2.720650286138413, "language_loss": 0.84151995, "learning_rate": 3.1011623496936973e-06, "loss": 0.86075878, "num_input_tokens_seen": 60036770, "step": 2786, "time_per_iteration": 2.6106956005096436 }, { "auxiliary_loss_clip": 0.01195401, "auxiliary_loss_mlp": 0.01027176, "balance_loss_clip": 1.05686307, "balance_loss_mlp": 1.01915944, "epoch": 0.3351169362111465, "flos": 28111451952960.0, "grad_norm": 1.9939581106515005, "language_loss": 0.69800538, "learning_rate": 3.100511994965893e-06, "loss": 0.72023118, "num_input_tokens_seen": 60056725, "step": 2787, "time_per_iteration": 2.53555965423584 }, { "auxiliary_loss_clip": 0.01176299, "auxiliary_loss_mlp": 0.01030266, "balance_loss_clip": 1.05561101, "balance_loss_mlp": 1.0216887, "epoch": 0.33523717910178563, "flos": 22673318975040.0, "grad_norm": 1.8247981293696345, "language_loss": 0.84571671, "learning_rate": 3.0998614732912947e-06, "loss": 0.86778241, "num_input_tokens_seen": 60076100, "step": 2788, "time_per_iteration": 2.53610897064209 }, { "auxiliary_loss_clip": 0.0118237, "auxiliary_loss_mlp": 0.01031206, "balance_loss_clip": 1.05745721, "balance_loss_mlp": 1.02226496, "epoch": 0.3353574219924247, "flos": 15669804491520.0, "grad_norm": 7.226182538555187, "language_loss": 0.67734206, "learning_rate": 3.0992107847685855e-06, "loss": 0.69947779, "num_input_tokens_seen": 60093815, "step": 2789, "time_per_iteration": 2.473815679550171 }, { "auxiliary_loss_clip": 0.01169364, "auxiliary_loss_mlp": 0.01035294, "balance_loss_clip": 1.05857611, "balance_loss_mlp": 1.02545309, "epoch": 0.3354776648830638, "flos": 24790648331040.0, "grad_norm": 1.9190640843048086, "language_loss": 0.79370749, "learning_rate": 3.0985599294964736e-06, "loss": 0.81575406, "num_input_tokens_seen": 60113370, "step": 2790, "time_per_iteration": 2.5946342945098877 }, { "auxiliary_loss_clip": 0.0116201, "auxiliary_loss_mlp": 0.01040029, "balance_loss_clip": 1.05109787, "balance_loss_mlp": 1.03185701, "epoch": 0.33559790777370285, "flos": 28694856672000.0, "grad_norm": 1.9440538790057027, "language_loss": 0.70005113, "learning_rate": 3.097908907573695e-06, "loss": 0.72207153, "num_input_tokens_seen": 60131350, "step": 2791, "time_per_iteration": 2.6193413734436035 }, { "auxiliary_loss_clip": 0.0112577, "auxiliary_loss_mlp": 0.01029449, "balance_loss_clip": 1.05255306, "balance_loss_mlp": 1.02127099, "epoch": 0.33571815066434196, "flos": 22235783394240.0, "grad_norm": 1.8984752324305574, "language_loss": 0.8926813, "learning_rate": 3.0972577190990067e-06, "loss": 0.91423345, "num_input_tokens_seen": 60149830, "step": 2792, "time_per_iteration": 2.6483449935913086 }, { "auxiliary_loss_clip": 0.01161397, "auxiliary_loss_mlp": 0.01024698, "balance_loss_clip": 1.05572367, "balance_loss_mlp": 1.01664543, "epoch": 0.3358383935549811, "flos": 23842283430720.0, "grad_norm": 1.816477317967998, "language_loss": 0.80206013, "learning_rate": 3.096606364171196e-06, "loss": 0.82392108, "num_input_tokens_seen": 60169620, "step": 2793, "time_per_iteration": 2.587965250015259 }, { "auxiliary_loss_clip": 0.01139497, "auxiliary_loss_mlp": 0.01032309, "balance_loss_clip": 1.0521574, "balance_loss_mlp": 1.02394092, "epoch": 0.33595863644562013, "flos": 22267312096800.0, "grad_norm": 1.822380996355253, "language_loss": 0.84946644, "learning_rate": 3.0959548428890703e-06, "loss": 0.87118459, "num_input_tokens_seen": 60188490, "step": 2794, "time_per_iteration": 2.5847461223602295 }, { "auxiliary_loss_clip": 0.01178272, "auxiliary_loss_mlp": 0.01026529, "balance_loss_clip": 1.05576754, "balance_loss_mlp": 1.01872635, "epoch": 0.33607887933625924, "flos": 20119782965760.0, "grad_norm": 1.5450767645512793, "language_loss": 0.84187365, "learning_rate": 3.095303155351468e-06, "loss": 0.86392164, "num_input_tokens_seen": 60208695, "step": 2795, "time_per_iteration": 2.542125701904297 }, { "auxiliary_loss_clip": 0.01129673, "auxiliary_loss_mlp": 0.01028263, "balance_loss_clip": 1.04914343, "balance_loss_mlp": 1.02013338, "epoch": 0.33619912222689835, "flos": 19318113293760.0, "grad_norm": 2.0538883682971734, "language_loss": 0.78764892, "learning_rate": 3.0946513016572464e-06, "loss": 0.8092283, "num_input_tokens_seen": 60227600, "step": 2796, "time_per_iteration": 3.466799020767212 }, { "auxiliary_loss_clip": 0.01185171, "auxiliary_loss_mlp": 0.01028395, "balance_loss_clip": 1.05612814, "balance_loss_mlp": 1.01997328, "epoch": 0.3363193651175374, "flos": 16800667440960.0, "grad_norm": 1.896214841059284, "language_loss": 0.76929289, "learning_rate": 3.0939992819052938e-06, "loss": 0.79142857, "num_input_tokens_seen": 60245110, "step": 2797, "time_per_iteration": 2.4807755947113037 }, { "auxiliary_loss_clip": 0.01167968, "auxiliary_loss_mlp": 0.0103105, "balance_loss_clip": 1.05420685, "balance_loss_mlp": 1.02295566, "epoch": 0.3364396080081765, "flos": 23550293735520.0, "grad_norm": 1.977287014013697, "language_loss": 0.81244051, "learning_rate": 3.0933470961945193e-06, "loss": 0.83443069, "num_input_tokens_seen": 60263405, "step": 2798, "time_per_iteration": 2.5662448406219482 }, { "auxiliary_loss_clip": 0.01161831, "auxiliary_loss_mlp": 0.01026778, "balance_loss_clip": 1.05447721, "balance_loss_mlp": 1.01925552, "epoch": 0.3365598508988156, "flos": 28037906795520.0, "grad_norm": 1.7186392269517132, "language_loss": 0.68390465, "learning_rate": 3.0926947446238597e-06, "loss": 0.70579076, "num_input_tokens_seen": 60282975, "step": 2799, "time_per_iteration": 3.315581798553467 }, { "auxiliary_loss_clip": 0.01186242, "auxiliary_loss_mlp": 0.01030936, "balance_loss_clip": 1.05266428, "balance_loss_mlp": 1.02257943, "epoch": 0.3366800937894547, "flos": 16982770403520.0, "grad_norm": 2.75891135384362, "language_loss": 0.82237911, "learning_rate": 3.092042227292276e-06, "loss": 0.84455091, "num_input_tokens_seen": 60299810, "step": 2800, "time_per_iteration": 2.475341558456421 }, { "auxiliary_loss_clip": 0.01193574, "auxiliary_loss_mlp": 0.01029863, "balance_loss_clip": 1.05752754, "balance_loss_mlp": 1.02218604, "epoch": 0.3368003366800938, "flos": 23915325750720.0, "grad_norm": 1.793511372729944, "language_loss": 0.88112736, "learning_rate": 3.0913895442987557e-06, "loss": 0.90336174, "num_input_tokens_seen": 60320775, "step": 2801, "time_per_iteration": 2.5248193740844727 }, { "auxiliary_loss_clip": 0.01151197, "auxiliary_loss_mlp": 0.00763203, "balance_loss_clip": 1.05280972, "balance_loss_mlp": 1.00008297, "epoch": 0.3369205795707329, "flos": 24791223002400.0, "grad_norm": 1.6771462663818504, "language_loss": 0.86067963, "learning_rate": 3.090736695742308e-06, "loss": 0.87982357, "num_input_tokens_seen": 60341905, "step": 2802, "time_per_iteration": 2.5946204662323 }, { "auxiliary_loss_clip": 0.01129684, "auxiliary_loss_mlp": 0.01034361, "balance_loss_clip": 1.04659927, "balance_loss_mlp": 1.02598703, "epoch": 0.33704082246137196, "flos": 17931099386880.0, "grad_norm": 2.463992034121681, "language_loss": 0.52340806, "learning_rate": 3.0900836817219713e-06, "loss": 0.54504853, "num_input_tokens_seen": 60358335, "step": 2803, "time_per_iteration": 2.580260753631592 }, { "auxiliary_loss_clip": 0.011945, "auxiliary_loss_mlp": 0.01031132, "balance_loss_clip": 1.05528998, "balance_loss_mlp": 1.02278161, "epoch": 0.33716106535201107, "flos": 21286520569920.0, "grad_norm": 1.8932892892094262, "language_loss": 0.83647752, "learning_rate": 3.089430502336807e-06, "loss": 0.85873383, "num_input_tokens_seen": 60378305, "step": 2804, "time_per_iteration": 2.4953935146331787 }, { "auxiliary_loss_clip": 0.01184329, "auxiliary_loss_mlp": 0.01032355, "balance_loss_clip": 1.0547297, "balance_loss_mlp": 1.02337289, "epoch": 0.3372813082426502, "flos": 18402965193120.0, "grad_norm": 2.673059934001429, "language_loss": 0.90818006, "learning_rate": 3.088777157685902e-06, "loss": 0.93034691, "num_input_tokens_seen": 60393895, "step": 2805, "time_per_iteration": 3.207437515258789 }, { "auxiliary_loss_clip": 0.0116434, "auxiliary_loss_mlp": 0.01027537, "balance_loss_clip": 1.05481029, "balance_loss_mlp": 1.01980007, "epoch": 0.33740155113328923, "flos": 17201394526080.0, "grad_norm": 3.1762480607327923, "language_loss": 0.85298741, "learning_rate": 3.088123647868367e-06, "loss": 0.87490618, "num_input_tokens_seen": 60410445, "step": 2806, "time_per_iteration": 2.5216567516326904 }, { "auxiliary_loss_clip": 0.01182544, "auxiliary_loss_mlp": 0.01028444, "balance_loss_clip": 1.05261159, "balance_loss_mlp": 1.02113986, "epoch": 0.33752179402392835, "flos": 29058954846240.0, "grad_norm": 2.3802219973277823, "language_loss": 0.81058365, "learning_rate": 3.0874699729833405e-06, "loss": 0.83269358, "num_input_tokens_seen": 60431815, "step": 2807, "time_per_iteration": 2.5935604572296143 }, { "auxiliary_loss_clip": 0.01163302, "auxiliary_loss_mlp": 0.010279, "balance_loss_clip": 1.05355024, "balance_loss_mlp": 1.01951408, "epoch": 0.3376420369145674, "flos": 25080734427360.0, "grad_norm": 1.645315352338714, "language_loss": 0.79602659, "learning_rate": 3.086816133129983e-06, "loss": 0.81793857, "num_input_tokens_seen": 60452075, "step": 2808, "time_per_iteration": 2.5603904724121094 }, { "auxiliary_loss_clip": 0.01195596, "auxiliary_loss_mlp": 0.01027981, "balance_loss_clip": 1.05778885, "balance_loss_mlp": 1.02016735, "epoch": 0.3377622798052065, "flos": 27490628149920.0, "grad_norm": 1.9372373456549032, "language_loss": 0.76108348, "learning_rate": 3.0861621284074826e-06, "loss": 0.78331929, "num_input_tokens_seen": 60472600, "step": 2809, "time_per_iteration": 2.5216190814971924 }, { "auxiliary_loss_clip": 0.01172531, "auxiliary_loss_mlp": 0.01031414, "balance_loss_clip": 1.0540688, "balance_loss_mlp": 1.02271199, "epoch": 0.3378825226958456, "flos": 21975214650720.0, "grad_norm": 1.8978019950931233, "language_loss": 0.72826171, "learning_rate": 3.085507958915051e-06, "loss": 0.75030112, "num_input_tokens_seen": 60491030, "step": 2810, "time_per_iteration": 2.53971266746521 }, { "auxiliary_loss_clip": 0.01160489, "auxiliary_loss_mlp": 0.01030655, "balance_loss_clip": 1.05113935, "balance_loss_mlp": 1.02191091, "epoch": 0.3380027655864847, "flos": 42523194787680.0, "grad_norm": 2.0574551207300567, "language_loss": 0.71323299, "learning_rate": 3.084853624751925e-06, "loss": 0.73514438, "num_input_tokens_seen": 60512615, "step": 2811, "time_per_iteration": 2.732311248779297 }, { "auxiliary_loss_clip": 0.01155306, "auxiliary_loss_mlp": 0.01023685, "balance_loss_clip": 1.05262077, "balance_loss_mlp": 1.01526904, "epoch": 0.3381230084771238, "flos": 26725084551360.0, "grad_norm": 2.0759215614178714, "language_loss": 0.8605727, "learning_rate": 3.0841991260173668e-06, "loss": 0.8823626, "num_input_tokens_seen": 60532520, "step": 2812, "time_per_iteration": 2.59912371635437 }, { "auxiliary_loss_clip": 0.01195962, "auxiliary_loss_mlp": 0.01033484, "balance_loss_clip": 1.05635059, "balance_loss_mlp": 1.02557421, "epoch": 0.3382432513677629, "flos": 22710378889440.0, "grad_norm": 2.0175778011140784, "language_loss": 0.80270541, "learning_rate": 3.0835444628106634e-06, "loss": 0.82499993, "num_input_tokens_seen": 60551500, "step": 2813, "time_per_iteration": 2.4840242862701416 }, { "auxiliary_loss_clip": 0.0119474, "auxiliary_loss_mlp": 0.00763441, "balance_loss_clip": 1.05621719, "balance_loss_mlp": 1.00013638, "epoch": 0.33836349425840195, "flos": 22122412716480.0, "grad_norm": 1.8539434875019074, "language_loss": 0.82607484, "learning_rate": 3.082889635231126e-06, "loss": 0.84565663, "num_input_tokens_seen": 60570160, "step": 2814, "time_per_iteration": 2.464776039123535 }, { "auxiliary_loss_clip": 0.01170206, "auxiliary_loss_mlp": 0.01036768, "balance_loss_clip": 1.05277526, "balance_loss_mlp": 1.02784562, "epoch": 0.33848373714904106, "flos": 27308094183840.0, "grad_norm": 2.962048557541726, "language_loss": 0.76553243, "learning_rate": 3.0822346433780925e-06, "loss": 0.78760219, "num_input_tokens_seen": 60590885, "step": 2815, "time_per_iteration": 2.5943753719329834 }, { "auxiliary_loss_clip": 0.01182547, "auxiliary_loss_mlp": 0.01026501, "balance_loss_clip": 1.0529784, "balance_loss_mlp": 1.01706517, "epoch": 0.3386039800396802, "flos": 25848720379200.0, "grad_norm": 1.8860857997600773, "language_loss": 0.87321019, "learning_rate": 3.0815794873509237e-06, "loss": 0.89530069, "num_input_tokens_seen": 60609170, "step": 2816, "time_per_iteration": 2.533353805541992 }, { "auxiliary_loss_clip": 0.01194038, "auxiliary_loss_mlp": 0.01026578, "balance_loss_clip": 1.05548573, "balance_loss_mlp": 1.01849508, "epoch": 0.33872422293031923, "flos": 18880649546880.0, "grad_norm": 1.9312918982114942, "language_loss": 0.72620535, "learning_rate": 3.0809241672490066e-06, "loss": 0.74841154, "num_input_tokens_seen": 60627340, "step": 2817, "time_per_iteration": 2.4691355228424072 }, { "auxiliary_loss_clip": 0.01167203, "auxiliary_loss_mlp": 0.01028618, "balance_loss_clip": 1.05477071, "balance_loss_mlp": 1.02121544, "epoch": 0.33884446582095834, "flos": 23146980629280.0, "grad_norm": 3.0998779187657752, "language_loss": 0.84872228, "learning_rate": 3.080268683171753e-06, "loss": 0.87068051, "num_input_tokens_seen": 60647630, "step": 2818, "time_per_iteration": 2.5847904682159424 }, { "auxiliary_loss_clip": 0.0118056, "auxiliary_loss_mlp": 0.01033531, "balance_loss_clip": 1.05207372, "balance_loss_mlp": 1.02593184, "epoch": 0.33896470871159745, "flos": 15997345588800.0, "grad_norm": 2.901547586967142, "language_loss": 0.88859087, "learning_rate": 3.0796130352185985e-06, "loss": 0.91073179, "num_input_tokens_seen": 60664485, "step": 2819, "time_per_iteration": 2.504368782043457 }, { "auxiliary_loss_clip": 0.01152072, "auxiliary_loss_mlp": 0.00763957, "balance_loss_clip": 1.04643023, "balance_loss_mlp": 1.00023377, "epoch": 0.3390849516022365, "flos": 34495758896640.0, "grad_norm": 1.6475916029833193, "language_loss": 0.66726565, "learning_rate": 3.0789572234890057e-06, "loss": 0.68642592, "num_input_tokens_seen": 60686125, "step": 2820, "time_per_iteration": 2.656726598739624 }, { "auxiliary_loss_clip": 0.01165273, "auxiliary_loss_mlp": 0.01027076, "balance_loss_clip": 1.05424261, "balance_loss_mlp": 1.01837361, "epoch": 0.3392051944928756, "flos": 16180310558400.0, "grad_norm": 1.6040779707017785, "language_loss": 0.77162981, "learning_rate": 3.0783012480824596e-06, "loss": 0.79355329, "num_input_tokens_seen": 60705270, "step": 2821, "time_per_iteration": 2.53363299369812 }, { "auxiliary_loss_clip": 0.01193329, "auxiliary_loss_mlp": 0.01039125, "balance_loss_clip": 1.05366349, "balance_loss_mlp": 1.03057814, "epoch": 0.33932543738351467, "flos": 17086659003840.0, "grad_norm": 2.009957208399566, "language_loss": 0.74187374, "learning_rate": 3.077645109098471e-06, "loss": 0.7641983, "num_input_tokens_seen": 60721540, "step": 2822, "time_per_iteration": 3.290968179702759 }, { "auxiliary_loss_clip": 0.01139469, "auxiliary_loss_mlp": 0.01031987, "balance_loss_clip": 1.04846144, "balance_loss_mlp": 1.02374351, "epoch": 0.3394456802741538, "flos": 22126974170400.0, "grad_norm": 1.9036921267808904, "language_loss": 0.72109473, "learning_rate": 3.076988806636577e-06, "loss": 0.7428093, "num_input_tokens_seen": 60739300, "step": 2823, "time_per_iteration": 2.558926582336426 }, { "auxiliary_loss_clip": 0.01169425, "auxiliary_loss_mlp": 0.00763687, "balance_loss_clip": 1.05453062, "balance_loss_mlp": 1.00010908, "epoch": 0.3395659231647929, "flos": 25226890901280.0, "grad_norm": 1.8267794864162963, "language_loss": 0.89017087, "learning_rate": 3.0763323407963377e-06, "loss": 0.90950197, "num_input_tokens_seen": 60758910, "step": 2824, "time_per_iteration": 2.5884323120117188 }, { "auxiliary_loss_clip": 0.01183044, "auxiliary_loss_mlp": 0.01024281, "balance_loss_clip": 1.05560946, "balance_loss_mlp": 1.01636529, "epoch": 0.33968616605543195, "flos": 29096481681120.0, "grad_norm": 2.1965867839828412, "language_loss": 0.80447638, "learning_rate": 3.075675711677337e-06, "loss": 0.82654965, "num_input_tokens_seen": 60779005, "step": 2825, "time_per_iteration": 3.3337643146514893 }, { "auxiliary_loss_clip": 0.01162088, "auxiliary_loss_mlp": 0.01026961, "balance_loss_clip": 1.05524516, "balance_loss_mlp": 1.01904535, "epoch": 0.33980640894607106, "flos": 21433970054400.0, "grad_norm": 1.932578401949671, "language_loss": 0.78075683, "learning_rate": 3.0750189193791865e-06, "loss": 0.80264735, "num_input_tokens_seen": 60798590, "step": 2826, "time_per_iteration": 2.5421977043151855 }, { "auxiliary_loss_clip": 0.01177918, "auxiliary_loss_mlp": 0.01031032, "balance_loss_clip": 1.05295277, "balance_loss_mlp": 1.02148974, "epoch": 0.33992665183671017, "flos": 32490040287360.0, "grad_norm": 1.7860073514586536, "language_loss": 0.70417118, "learning_rate": 3.0743619640015203e-06, "loss": 0.72626072, "num_input_tokens_seen": 60818840, "step": 2827, "time_per_iteration": 2.596404552459717 }, { "auxiliary_loss_clip": 0.01171978, "auxiliary_loss_mlp": 0.01033123, "balance_loss_clip": 1.05175138, "balance_loss_mlp": 1.02477217, "epoch": 0.3400468947273492, "flos": 17055417636960.0, "grad_norm": 1.990471430874163, "language_loss": 0.92651749, "learning_rate": 3.073704845643999e-06, "loss": 0.94856846, "num_input_tokens_seen": 60835965, "step": 2828, "time_per_iteration": 2.5404396057128906 }, { "auxiliary_loss_clip": 0.01181339, "auxiliary_loss_mlp": 0.01030628, "balance_loss_clip": 1.0515765, "balance_loss_mlp": 1.02218163, "epoch": 0.34016713761798834, "flos": 16872991421760.0, "grad_norm": 2.891113860484898, "language_loss": 0.78025788, "learning_rate": 3.0730475644063063e-06, "loss": 0.80237752, "num_input_tokens_seen": 60851065, "step": 2829, "time_per_iteration": 2.4899871349334717 }, { "auxiliary_loss_clip": 0.01159111, "auxiliary_loss_mlp": 0.00762965, "balance_loss_clip": 1.04987752, "balance_loss_mlp": 1.00009167, "epoch": 0.34028738050862745, "flos": 21907164788160.0, "grad_norm": 1.9158341629076412, "language_loss": 0.65080631, "learning_rate": 3.072390120388151e-06, "loss": 0.67002714, "num_input_tokens_seen": 60869390, "step": 2830, "time_per_iteration": 2.562991142272949 }, { "auxiliary_loss_clip": 0.01182092, "auxiliary_loss_mlp": 0.0103145, "balance_loss_clip": 1.05498481, "balance_loss_mlp": 1.02274752, "epoch": 0.3404076233992665, "flos": 22746038042400.0, "grad_norm": 3.5754627955946603, "language_loss": 0.70951378, "learning_rate": 3.071732513689267e-06, "loss": 0.73164928, "num_input_tokens_seen": 60887925, "step": 2831, "time_per_iteration": 3.2826743125915527 }, { "auxiliary_loss_clip": 0.01184487, "auxiliary_loss_mlp": 0.01038189, "balance_loss_clip": 1.05869091, "balance_loss_mlp": 1.02952874, "epoch": 0.3405278662899056, "flos": 17052364695360.0, "grad_norm": 2.2494338676609975, "language_loss": 0.67535841, "learning_rate": 3.0710747444094134e-06, "loss": 0.69758523, "num_input_tokens_seen": 60905955, "step": 2832, "time_per_iteration": 2.477678060531616 }, { "auxiliary_loss_clip": 0.01172974, "auxiliary_loss_mlp": 0.01030476, "balance_loss_clip": 1.05601811, "balance_loss_mlp": 1.02166665, "epoch": 0.3406481091805447, "flos": 42813137216160.0, "grad_norm": 1.7935724407684492, "language_loss": 0.65037018, "learning_rate": 3.070416812648372e-06, "loss": 0.67240465, "num_input_tokens_seen": 60929405, "step": 2833, "time_per_iteration": 2.7568955421447754 }, { "auxiliary_loss_clip": 0.01148841, "auxiliary_loss_mlp": 0.01025095, "balance_loss_clip": 1.04787159, "balance_loss_mlp": 1.01653624, "epoch": 0.3407683520711838, "flos": 26761462043520.0, "grad_norm": 1.6914723459333971, "language_loss": 0.64963794, "learning_rate": 3.069758718505951e-06, "loss": 0.6713773, "num_input_tokens_seen": 60951145, "step": 2834, "time_per_iteration": 2.6139678955078125 }, { "auxiliary_loss_clip": 0.01195977, "auxiliary_loss_mlp": 0.01026186, "balance_loss_clip": 1.05703855, "balance_loss_mlp": 1.01760888, "epoch": 0.3408885949618229, "flos": 28767647573280.0, "grad_norm": 1.6518083878920249, "language_loss": 0.79852509, "learning_rate": 3.0691004620819836e-06, "loss": 0.82074672, "num_input_tokens_seen": 60971275, "step": 2835, "time_per_iteration": 2.558464765548706 }, { "auxiliary_loss_clip": 0.01050259, "auxiliary_loss_mlp": 0.01012844, "balance_loss_clip": 1.02226496, "balance_loss_mlp": 1.0106982, "epoch": 0.341008837852462, "flos": 63576263918880.0, "grad_norm": 0.794144030978552, "language_loss": 0.60222089, "learning_rate": 3.0684420434763254e-06, "loss": 0.62285185, "num_input_tokens_seen": 61037460, "step": 2836, "time_per_iteration": 3.1913936138153076 }, { "auxiliary_loss_clip": 0.01143073, "auxiliary_loss_mlp": 0.01029445, "balance_loss_clip": 1.05126214, "balance_loss_mlp": 1.0216074, "epoch": 0.34112908074310105, "flos": 20812176493440.0, "grad_norm": 2.3579112908206965, "language_loss": 0.76745433, "learning_rate": 3.06778346278886e-06, "loss": 0.78917956, "num_input_tokens_seen": 61056295, "step": 2837, "time_per_iteration": 2.5783073902130127 }, { "auxiliary_loss_clip": 0.01198918, "auxiliary_loss_mlp": 0.01031027, "balance_loss_clip": 1.05926704, "balance_loss_mlp": 1.02239025, "epoch": 0.34124932363374016, "flos": 24976450740480.0, "grad_norm": 1.7848450003982472, "language_loss": 0.79077232, "learning_rate": 3.0671247201194906e-06, "loss": 0.81307173, "num_input_tokens_seen": 61078430, "step": 2838, "time_per_iteration": 2.532442808151245 }, { "auxiliary_loss_clip": 0.01157464, "auxiliary_loss_mlp": 0.01032579, "balance_loss_clip": 1.05280101, "balance_loss_mlp": 1.02328086, "epoch": 0.3413695665243792, "flos": 28402974727680.0, "grad_norm": 1.6574376631321657, "language_loss": 0.75578672, "learning_rate": 3.066465815568151e-06, "loss": 0.77768719, "num_input_tokens_seen": 61099260, "step": 2839, "time_per_iteration": 2.636019229888916 }, { "auxiliary_loss_clip": 0.01182854, "auxiliary_loss_mlp": 0.01030877, "balance_loss_clip": 1.05301118, "balance_loss_mlp": 1.02220464, "epoch": 0.34148980941501833, "flos": 25302016404960.0, "grad_norm": 3.182998840282505, "language_loss": 0.69051099, "learning_rate": 3.0658067492347947e-06, "loss": 0.71264827, "num_input_tokens_seen": 61121900, "step": 2840, "time_per_iteration": 2.5530498027801514 }, { "auxiliary_loss_clip": 0.0110528, "auxiliary_loss_mlp": 0.01036792, "balance_loss_clip": 1.04869962, "balance_loss_mlp": 1.02819645, "epoch": 0.34161005230565744, "flos": 17530084966080.0, "grad_norm": 2.008167859008023, "language_loss": 0.67063624, "learning_rate": 3.065147521219402e-06, "loss": 0.69205701, "num_input_tokens_seen": 61141155, "step": 2841, "time_per_iteration": 2.665451765060425 }, { "auxiliary_loss_clip": 0.01159849, "auxiliary_loss_mlp": 0.01037838, "balance_loss_clip": 1.05541003, "balance_loss_mlp": 1.03001761, "epoch": 0.3417302951962965, "flos": 43650106871520.0, "grad_norm": 1.4718554727069615, "language_loss": 0.74495929, "learning_rate": 3.064488131621977e-06, "loss": 0.76693612, "num_input_tokens_seen": 61164480, "step": 2842, "time_per_iteration": 2.7480342388153076 }, { "auxiliary_loss_clip": 0.01174318, "auxiliary_loss_mlp": 0.01031216, "balance_loss_clip": 1.05296397, "balance_loss_mlp": 1.02248383, "epoch": 0.3418505380869356, "flos": 30882211323360.0, "grad_norm": 1.6244661126677786, "language_loss": 0.73910201, "learning_rate": 3.063828580542549e-06, "loss": 0.76115733, "num_input_tokens_seen": 61185675, "step": 2843, "time_per_iteration": 2.6070117950439453 }, { "auxiliary_loss_clip": 0.01166971, "auxiliary_loss_mlp": 0.01024713, "balance_loss_clip": 1.05295658, "balance_loss_mlp": 1.01656508, "epoch": 0.3419707809775747, "flos": 19463874681120.0, "grad_norm": 1.910811294941346, "language_loss": 0.73479658, "learning_rate": 3.0631688680811706e-06, "loss": 0.75671339, "num_input_tokens_seen": 61205300, "step": 2844, "time_per_iteration": 2.543966293334961 }, { "auxiliary_loss_clip": 0.01197928, "auxiliary_loss_mlp": 0.0102895, "balance_loss_clip": 1.05781126, "balance_loss_mlp": 1.02005708, "epoch": 0.3420910238682138, "flos": 28727822052960.0, "grad_norm": 2.405785193833368, "language_loss": 0.75467348, "learning_rate": 3.062508994337921e-06, "loss": 0.77694231, "num_input_tokens_seen": 61224905, "step": 2845, "time_per_iteration": 2.55295991897583 }, { "auxiliary_loss_clip": 0.01181976, "auxiliary_loss_mlp": 0.01035015, "balance_loss_clip": 1.05434704, "balance_loss_mlp": 1.02535319, "epoch": 0.3422112667588529, "flos": 21397269309600.0, "grad_norm": 2.1146766406007504, "language_loss": 0.79682755, "learning_rate": 3.0618489594129013e-06, "loss": 0.81899738, "num_input_tokens_seen": 61243045, "step": 2846, "time_per_iteration": 2.520143508911133 }, { "auxiliary_loss_clip": 0.01157099, "auxiliary_loss_mlp": 0.01031413, "balance_loss_clip": 1.0540545, "balance_loss_mlp": 1.02228808, "epoch": 0.342331509649492, "flos": 13881452911200.0, "grad_norm": 1.8750732399900083, "language_loss": 0.71186286, "learning_rate": 3.061188763406239e-06, "loss": 0.73374796, "num_input_tokens_seen": 61259190, "step": 2847, "time_per_iteration": 2.5846917629241943 }, { "auxiliary_loss_clip": 0.01162466, "auxiliary_loss_mlp": 0.01026228, "balance_loss_clip": 1.05308509, "balance_loss_mlp": 1.01799023, "epoch": 0.34245175254013105, "flos": 28621455182400.0, "grad_norm": 2.0853160572154867, "language_loss": 0.82259691, "learning_rate": 3.060528406418085e-06, "loss": 0.84448379, "num_input_tokens_seen": 61279040, "step": 2848, "time_per_iteration": 3.4478886127471924 }, { "auxiliary_loss_clip": 0.0116189, "auxiliary_loss_mlp": 0.01031816, "balance_loss_clip": 1.05353248, "balance_loss_mlp": 1.02335191, "epoch": 0.34257199543077016, "flos": 34127063351520.0, "grad_norm": 1.5986217032220265, "language_loss": 0.62000263, "learning_rate": 3.0598678885486145e-06, "loss": 0.64193976, "num_input_tokens_seen": 61301580, "step": 2849, "time_per_iteration": 2.6505141258239746 }, { "auxiliary_loss_clip": 0.01151261, "auxiliary_loss_mlp": 0.00763964, "balance_loss_clip": 1.04833233, "balance_loss_mlp": 1.00007272, "epoch": 0.34269223832140927, "flos": 19974021578400.0, "grad_norm": 1.5971595031904122, "language_loss": 0.74736971, "learning_rate": 3.0592072098980282e-06, "loss": 0.76652193, "num_input_tokens_seen": 61321240, "step": 2850, "time_per_iteration": 2.5711727142333984 }, { "auxiliary_loss_clip": 0.01159274, "auxiliary_loss_mlp": 0.01033635, "balance_loss_clip": 1.05031705, "balance_loss_mlp": 1.0251174, "epoch": 0.3428124812120483, "flos": 27235662452160.0, "grad_norm": 1.9430936379346917, "language_loss": 0.7245369, "learning_rate": 3.0585463705665514e-06, "loss": 0.74646604, "num_input_tokens_seen": 61341615, "step": 2851, "time_per_iteration": 3.372849225997925 }, { "auxiliary_loss_clip": 0.01150995, "auxiliary_loss_mlp": 0.01031099, "balance_loss_clip": 1.05027819, "balance_loss_mlp": 1.02242029, "epoch": 0.34293272410268744, "flos": 24570874865760.0, "grad_norm": 2.324040210482987, "language_loss": 0.70864117, "learning_rate": 3.0578853706544304e-06, "loss": 0.73046213, "num_input_tokens_seen": 61359005, "step": 2852, "time_per_iteration": 2.604518413543701 }, { "auxiliary_loss_clip": 0.01156128, "auxiliary_loss_mlp": 0.00764212, "balance_loss_clip": 1.05285501, "balance_loss_mlp": 1.00011921, "epoch": 0.34305296699332655, "flos": 21506868706560.0, "grad_norm": 1.8931666863167609, "language_loss": 0.65631145, "learning_rate": 3.0572242102619404e-06, "loss": 0.67551488, "num_input_tokens_seen": 61376160, "step": 2853, "time_per_iteration": 2.5773966312408447 }, { "auxiliary_loss_clip": 0.01169378, "auxiliary_loss_mlp": 0.01026394, "balance_loss_clip": 1.05715656, "balance_loss_mlp": 1.01829433, "epoch": 0.3431732098839656, "flos": 24056669352000.0, "grad_norm": 1.9267467938931164, "language_loss": 0.80298018, "learning_rate": 3.0565628894893784e-06, "loss": 0.82493782, "num_input_tokens_seen": 61396795, "step": 2854, "time_per_iteration": 2.5864861011505127 }, { "auxiliary_loss_clip": 0.0117423, "auxiliary_loss_mlp": 0.01035383, "balance_loss_clip": 1.05370545, "balance_loss_mlp": 1.02702045, "epoch": 0.3432934527746047, "flos": 16800882942720.0, "grad_norm": 1.7814969512969125, "language_loss": 0.74512255, "learning_rate": 3.0559014084370655e-06, "loss": 0.76721871, "num_input_tokens_seen": 61415320, "step": 2855, "time_per_iteration": 2.4807498455047607 }, { "auxiliary_loss_clip": 0.01176069, "auxiliary_loss_mlp": 0.01034824, "balance_loss_clip": 1.05572295, "balance_loss_mlp": 1.02614546, "epoch": 0.34341369566524377, "flos": 23439724580640.0, "grad_norm": 1.6707038415253173, "language_loss": 0.78592014, "learning_rate": 3.055239767205349e-06, "loss": 0.80802906, "num_input_tokens_seen": 61437070, "step": 2856, "time_per_iteration": 3.3168463706970215 }, { "auxiliary_loss_clip": 0.01180918, "auxiliary_loss_mlp": 0.01031645, "balance_loss_clip": 1.05820131, "balance_loss_mlp": 1.02386642, "epoch": 0.3435339385558829, "flos": 17267468955840.0, "grad_norm": 1.7110657651304575, "language_loss": 0.78391266, "learning_rate": 3.054577965894599e-06, "loss": 0.80603826, "num_input_tokens_seen": 61453215, "step": 2857, "time_per_iteration": 2.4999232292175293 }, { "auxiliary_loss_clip": 0.01175445, "auxiliary_loss_mlp": 0.01035111, "balance_loss_clip": 1.05758965, "balance_loss_mlp": 1.02623606, "epoch": 0.343654181446522, "flos": 22199369985120.0, "grad_norm": 1.4591687152434507, "language_loss": 0.70248204, "learning_rate": 3.0539160046052094e-06, "loss": 0.72458762, "num_input_tokens_seen": 61472915, "step": 2858, "time_per_iteration": 2.5276877880096436 }, { "auxiliary_loss_clip": 0.01160571, "auxiliary_loss_mlp": 0.01034386, "balance_loss_clip": 1.05167878, "balance_loss_mlp": 1.02432466, "epoch": 0.34377442433716104, "flos": 19901805348480.0, "grad_norm": 2.281516230950584, "language_loss": 0.70400155, "learning_rate": 3.0532538834376003e-06, "loss": 0.72595108, "num_input_tokens_seen": 61492475, "step": 2859, "time_per_iteration": 2.5725152492523193 }, { "auxiliary_loss_clip": 0.01186739, "auxiliary_loss_mlp": 0.01034032, "balance_loss_clip": 1.05611873, "balance_loss_mlp": 1.025383, "epoch": 0.34389466722780015, "flos": 22197681888000.0, "grad_norm": 1.859367187223045, "language_loss": 0.77953196, "learning_rate": 3.0525916024922143e-06, "loss": 0.80173969, "num_input_tokens_seen": 61511660, "step": 2860, "time_per_iteration": 2.5189144611358643 }, { "auxiliary_loss_clip": 0.01161873, "auxiliary_loss_mlp": 0.01037062, "balance_loss_clip": 1.05096626, "balance_loss_mlp": 1.02862787, "epoch": 0.34401491011843927, "flos": 18624570423360.0, "grad_norm": 2.7586649326979473, "language_loss": 0.84219515, "learning_rate": 3.0519291618695193e-06, "loss": 0.8641845, "num_input_tokens_seen": 61529060, "step": 2861, "time_per_iteration": 2.5233285427093506 }, { "auxiliary_loss_clip": 0.01141808, "auxiliary_loss_mlp": 0.01036186, "balance_loss_clip": 1.04833031, "balance_loss_mlp": 1.02768004, "epoch": 0.3441351530090783, "flos": 17858200734720.0, "grad_norm": 1.6026179137777352, "language_loss": 0.7565515, "learning_rate": 3.0512665616700065e-06, "loss": 0.7783314, "num_input_tokens_seen": 61548125, "step": 2862, "time_per_iteration": 2.5348150730133057 }, { "auxiliary_loss_clip": 0.01126885, "auxiliary_loss_mlp": 0.01037561, "balance_loss_clip": 1.04499936, "balance_loss_mlp": 1.02978802, "epoch": 0.34425539589971743, "flos": 23112758154720.0, "grad_norm": 1.7433002246726976, "language_loss": 0.89289868, "learning_rate": 3.0506038019941933e-06, "loss": 0.91454315, "num_input_tokens_seen": 61568135, "step": 2863, "time_per_iteration": 2.648420572280884 }, { "auxiliary_loss_clip": 0.0115426, "auxiliary_loss_mlp": 0.01030103, "balance_loss_clip": 1.05394781, "balance_loss_mlp": 1.02154374, "epoch": 0.34437563879035654, "flos": 21907703542560.0, "grad_norm": 2.590336329546786, "language_loss": 0.68066663, "learning_rate": 3.049940882942617e-06, "loss": 0.7025103, "num_input_tokens_seen": 61586920, "step": 2864, "time_per_iteration": 2.599477767944336 }, { "auxiliary_loss_clip": 0.01194819, "auxiliary_loss_mlp": 0.01042454, "balance_loss_clip": 1.05488825, "balance_loss_mlp": 1.03373396, "epoch": 0.3444958816809956, "flos": 23076919416960.0, "grad_norm": 2.233516889583789, "language_loss": 0.80401099, "learning_rate": 3.0492778046158448e-06, "loss": 0.82638377, "num_input_tokens_seen": 61608340, "step": 2865, "time_per_iteration": 2.5213170051574707 }, { "auxiliary_loss_clip": 0.01181314, "auxiliary_loss_mlp": 0.01034563, "balance_loss_clip": 1.05842543, "balance_loss_mlp": 1.0265491, "epoch": 0.3446161245716347, "flos": 21908637383520.0, "grad_norm": 1.9288895815507037, "language_loss": 0.76632071, "learning_rate": 3.0486145671144633e-06, "loss": 0.78847945, "num_input_tokens_seen": 61628130, "step": 2866, "time_per_iteration": 2.547534227371216 }, { "auxiliary_loss_clip": 0.01106776, "auxiliary_loss_mlp": 0.01031215, "balance_loss_clip": 1.04706347, "balance_loss_mlp": 1.0232935, "epoch": 0.3447363674622738, "flos": 25112837801280.0, "grad_norm": 2.4530069823758924, "language_loss": 0.76994658, "learning_rate": 3.047951170539086e-06, "loss": 0.79132652, "num_input_tokens_seen": 61647755, "step": 2867, "time_per_iteration": 2.689758062362671 }, { "auxiliary_loss_clip": 0.01148915, "auxiliary_loss_mlp": 0.01037113, "balance_loss_clip": 1.05290008, "balance_loss_mlp": 1.02914667, "epoch": 0.3448566103529129, "flos": 11984687193600.0, "grad_norm": 1.8304048357501932, "language_loss": 0.84331751, "learning_rate": 3.047287614990349e-06, "loss": 0.86517775, "num_input_tokens_seen": 61665675, "step": 2868, "time_per_iteration": 2.552417278289795 }, { "auxiliary_loss_clip": 0.01158526, "auxiliary_loss_mlp": 0.01035948, "balance_loss_clip": 1.05194783, "balance_loss_mlp": 1.02750754, "epoch": 0.344976853243552, "flos": 40187887814400.0, "grad_norm": 2.235908270085406, "language_loss": 0.62062216, "learning_rate": 3.046623900568914e-06, "loss": 0.64256692, "num_input_tokens_seen": 61688240, "step": 2869, "time_per_iteration": 2.7168400287628174 }, { "auxiliary_loss_clip": 0.01162914, "auxiliary_loss_mlp": 0.01034333, "balance_loss_clip": 1.05113101, "balance_loss_mlp": 1.02569628, "epoch": 0.34509709613419104, "flos": 28723655685600.0, "grad_norm": 2.398847916631419, "language_loss": 0.70054746, "learning_rate": 3.045960027375465e-06, "loss": 0.72251993, "num_input_tokens_seen": 61706075, "step": 2870, "time_per_iteration": 2.5879294872283936 }, { "auxiliary_loss_clip": 0.01185045, "auxiliary_loss_mlp": 0.0103526, "balance_loss_clip": 1.0534867, "balance_loss_mlp": 1.02605677, "epoch": 0.34521733902483015, "flos": 29967601977120.0, "grad_norm": 8.305670829861505, "language_loss": 0.82582211, "learning_rate": 3.045295995510711e-06, "loss": 0.84802514, "num_input_tokens_seen": 61723045, "step": 2871, "time_per_iteration": 2.5869624614715576 }, { "auxiliary_loss_clip": 0.01165647, "auxiliary_loss_mlp": 0.01030156, "balance_loss_clip": 1.05534887, "balance_loss_mlp": 1.02264535, "epoch": 0.34533758191546926, "flos": 27923063522400.0, "grad_norm": 1.7405712670719273, "language_loss": 0.73449516, "learning_rate": 3.0446318050753865e-06, "loss": 0.75645316, "num_input_tokens_seen": 61743525, "step": 2872, "time_per_iteration": 2.600170373916626 }, { "auxiliary_loss_clip": 0.01172122, "auxiliary_loss_mlp": 0.01027907, "balance_loss_clip": 1.0521028, "balance_loss_mlp": 1.01975358, "epoch": 0.3454578248061083, "flos": 27125883470400.0, "grad_norm": 2.5691346810451097, "language_loss": 0.77311724, "learning_rate": 3.0439674561702474e-06, "loss": 0.7951175, "num_input_tokens_seen": 61763025, "step": 2873, "time_per_iteration": 2.5555460453033447 }, { "auxiliary_loss_clip": 0.01179782, "auxiliary_loss_mlp": 0.01037565, "balance_loss_clip": 1.05750322, "balance_loss_mlp": 1.02891684, "epoch": 0.3455780676967474, "flos": 19024902421920.0, "grad_norm": 2.2655773047353605, "language_loss": 0.87729752, "learning_rate": 3.043302948896076e-06, "loss": 0.89947104, "num_input_tokens_seen": 61781630, "step": 2874, "time_per_iteration": 3.3938775062561035 }, { "auxiliary_loss_clip": 0.01126729, "auxiliary_loss_mlp": 0.01028849, "balance_loss_clip": 1.04841554, "balance_loss_mlp": 1.02067733, "epoch": 0.34569831058738654, "flos": 34496010315360.0, "grad_norm": 1.7921460989129234, "language_loss": 0.60363412, "learning_rate": 3.0426382833536756e-06, "loss": 0.6251899, "num_input_tokens_seen": 61804985, "step": 2875, "time_per_iteration": 2.705984592437744 }, { "auxiliary_loss_clip": 0.01146304, "auxiliary_loss_mlp": 0.01028218, "balance_loss_clip": 1.04873693, "balance_loss_mlp": 1.02048099, "epoch": 0.3458185534780256, "flos": 31138685533440.0, "grad_norm": 2.92638831408904, "language_loss": 0.77411485, "learning_rate": 3.041973459643877e-06, "loss": 0.79586005, "num_input_tokens_seen": 61824440, "step": 2876, "time_per_iteration": 2.672592878341675 }, { "auxiliary_loss_clip": 0.01131859, "auxiliary_loss_mlp": 0.01030281, "balance_loss_clip": 1.04821134, "balance_loss_mlp": 1.02187729, "epoch": 0.3459387963686647, "flos": 32452513452480.0, "grad_norm": 1.8971428262969454, "language_loss": 0.66573101, "learning_rate": 3.0413084778675334e-06, "loss": 0.68735242, "num_input_tokens_seen": 61845690, "step": 2877, "time_per_iteration": 3.420126438140869 }, { "auxiliary_loss_clip": 0.01160105, "auxiliary_loss_mlp": 0.00762628, "balance_loss_clip": 1.05173552, "balance_loss_mlp": 1.00005293, "epoch": 0.3460590392593038, "flos": 24675661390080.0, "grad_norm": 1.732976669588977, "language_loss": 0.83583939, "learning_rate": 3.0406433381255214e-06, "loss": 0.85506678, "num_input_tokens_seen": 61863725, "step": 2878, "time_per_iteration": 2.607114791870117 }, { "auxiliary_loss_clip": 0.01182136, "auxiliary_loss_mlp": 0.01028664, "balance_loss_clip": 1.05962563, "balance_loss_mlp": 1.02068913, "epoch": 0.34617928214994287, "flos": 18807319891200.0, "grad_norm": 2.3272474921219266, "language_loss": 0.8238405, "learning_rate": 3.0399780405187425e-06, "loss": 0.84594846, "num_input_tokens_seen": 61882720, "step": 2879, "time_per_iteration": 2.5073676109313965 }, { "auxiliary_loss_clip": 0.01176265, "auxiliary_loss_mlp": 0.01023915, "balance_loss_clip": 1.05331302, "balance_loss_mlp": 1.01568389, "epoch": 0.346299525040582, "flos": 24857656601760.0, "grad_norm": 1.842205808945897, "language_loss": 0.78643352, "learning_rate": 3.0393125851481216e-06, "loss": 0.80843532, "num_input_tokens_seen": 61902595, "step": 2880, "time_per_iteration": 2.5769567489624023 }, { "auxiliary_loss_clip": 0.01143923, "auxiliary_loss_mlp": 0.01029442, "balance_loss_clip": 1.04979169, "balance_loss_mlp": 1.02144313, "epoch": 0.3464197679312211, "flos": 16434917086560.0, "grad_norm": 2.419675681997491, "language_loss": 0.86467946, "learning_rate": 3.038646972114608e-06, "loss": 0.8864131, "num_input_tokens_seen": 61918920, "step": 2881, "time_per_iteration": 2.540743112564087 }, { "auxiliary_loss_clip": 0.01149713, "auxiliary_loss_mlp": 0.01029384, "balance_loss_clip": 1.05423224, "balance_loss_mlp": 1.02120662, "epoch": 0.34654001082186014, "flos": 22382478622560.0, "grad_norm": 1.6190052872905243, "language_loss": 0.67231107, "learning_rate": 3.037981201519174e-06, "loss": 0.69410205, "num_input_tokens_seen": 61939520, "step": 2882, "time_per_iteration": 3.3452932834625244 }, { "auxiliary_loss_clip": 0.01182818, "auxiliary_loss_mlp": 0.01028834, "balance_loss_clip": 1.06044316, "balance_loss_mlp": 1.02084672, "epoch": 0.34666025371249926, "flos": 19573905081600.0, "grad_norm": 1.8707326882677155, "language_loss": 0.71617258, "learning_rate": 3.0373152734628175e-06, "loss": 0.73828906, "num_input_tokens_seen": 61957800, "step": 2883, "time_per_iteration": 2.51983904838562 }, { "auxiliary_loss_clip": 0.01173644, "auxiliary_loss_mlp": 0.01036104, "balance_loss_clip": 1.05344772, "balance_loss_mlp": 1.02822983, "epoch": 0.34678049660313837, "flos": 15267640728000.0, "grad_norm": 2.026566418551512, "language_loss": 0.76630265, "learning_rate": 3.0366491880465584e-06, "loss": 0.78840017, "num_input_tokens_seen": 61975820, "step": 2884, "time_per_iteration": 2.506344795227051 }, { "auxiliary_loss_clip": 0.01202827, "auxiliary_loss_mlp": 0.01029198, "balance_loss_clip": 1.06276011, "balance_loss_mlp": 1.02148509, "epoch": 0.3469007394937774, "flos": 21181554460800.0, "grad_norm": 2.017705581221021, "language_loss": 0.82145232, "learning_rate": 3.035982945371443e-06, "loss": 0.84377265, "num_input_tokens_seen": 61997515, "step": 2885, "time_per_iteration": 2.54636287689209 }, { "auxiliary_loss_clip": 0.01177942, "auxiliary_loss_mlp": 0.01039957, "balance_loss_clip": 1.05806196, "balance_loss_mlp": 1.0311774, "epoch": 0.34702098238441653, "flos": 22375474815360.0, "grad_norm": 2.7691061872083154, "language_loss": 0.85532546, "learning_rate": 3.035316545538537e-06, "loss": 0.87750447, "num_input_tokens_seen": 62016310, "step": 2886, "time_per_iteration": 2.5628578662872314 }, { "auxiliary_loss_clip": 0.0116775, "auxiliary_loss_mlp": 0.01035767, "balance_loss_clip": 1.0588994, "balance_loss_mlp": 1.0278281, "epoch": 0.3471412252750556, "flos": 22929434015520.0, "grad_norm": 2.2342814597297513, "language_loss": 0.79775393, "learning_rate": 3.034649988648935e-06, "loss": 0.81978917, "num_input_tokens_seen": 62036075, "step": 2887, "time_per_iteration": 2.561793327331543 }, { "auxiliary_loss_clip": 0.01168207, "auxiliary_loss_mlp": 0.01029217, "balance_loss_clip": 1.05482507, "balance_loss_mlp": 1.02138531, "epoch": 0.3472614681656947, "flos": 21324262906560.0, "grad_norm": 2.1763275762561665, "language_loss": 0.80194336, "learning_rate": 3.033983274803752e-06, "loss": 0.82391763, "num_input_tokens_seen": 62055865, "step": 2888, "time_per_iteration": 2.547379732131958 }, { "auxiliary_loss_clip": 0.01167241, "auxiliary_loss_mlp": 0.01034031, "balance_loss_clip": 1.05634224, "balance_loss_mlp": 1.02620459, "epoch": 0.3473817110563338, "flos": 23475742903200.0, "grad_norm": 2.7248176204097097, "language_loss": 0.72551906, "learning_rate": 3.0333164041041283e-06, "loss": 0.74753177, "num_input_tokens_seen": 62072180, "step": 2889, "time_per_iteration": 2.564565420150757 }, { "auxiliary_loss_clip": 0.01127691, "auxiliary_loss_mlp": 0.01031896, "balance_loss_clip": 1.05041194, "balance_loss_mlp": 1.02404642, "epoch": 0.34750195394697286, "flos": 22346029296480.0, "grad_norm": 2.0985554673763005, "language_loss": 0.71810681, "learning_rate": 3.032649376651228e-06, "loss": 0.7397027, "num_input_tokens_seen": 62091600, "step": 2890, "time_per_iteration": 2.6438329219818115 }, { "auxiliary_loss_clip": 0.01156339, "auxiliary_loss_mlp": 0.01027684, "balance_loss_clip": 1.05542254, "balance_loss_mlp": 1.01888597, "epoch": 0.347622196837612, "flos": 29095009085760.0, "grad_norm": 1.5800562976722803, "language_loss": 0.75680572, "learning_rate": 3.031982192546238e-06, "loss": 0.77864593, "num_input_tokens_seen": 62114695, "step": 2891, "time_per_iteration": 2.6486594676971436 }, { "auxiliary_loss_clip": 0.01184961, "auxiliary_loss_mlp": 0.01027015, "balance_loss_clip": 1.05666769, "balance_loss_mlp": 1.01880169, "epoch": 0.3477424397282511, "flos": 22455736444320.0, "grad_norm": 1.968561210885047, "language_loss": 0.94502765, "learning_rate": 3.0313148518903696e-06, "loss": 0.96714747, "num_input_tokens_seen": 62134520, "step": 2892, "time_per_iteration": 2.557927131652832 }, { "auxiliary_loss_clip": 0.0117471, "auxiliary_loss_mlp": 0.01032471, "balance_loss_clip": 1.06014633, "balance_loss_mlp": 1.02402461, "epoch": 0.34786268261889014, "flos": 15778793300160.0, "grad_norm": 6.66468402322876, "language_loss": 0.81031859, "learning_rate": 3.030647354784859e-06, "loss": 0.83239037, "num_input_tokens_seen": 62151560, "step": 2893, "time_per_iteration": 2.5356996059417725 }, { "auxiliary_loss_clip": 0.0115473, "auxiliary_loss_mlp": 0.0101976, "balance_loss_clip": 1.05631423, "balance_loss_mlp": 1.01265812, "epoch": 0.34798292550952925, "flos": 20777630766240.0, "grad_norm": 2.0765734906307416, "language_loss": 0.76939297, "learning_rate": 3.029979701330964e-06, "loss": 0.79113781, "num_input_tokens_seen": 62170985, "step": 2894, "time_per_iteration": 2.580944061279297 }, { "auxiliary_loss_clip": 0.01175938, "auxiliary_loss_mlp": 0.01034455, "balance_loss_clip": 1.05784965, "balance_loss_mlp": 1.02640235, "epoch": 0.34810316840016836, "flos": 19937823671040.0, "grad_norm": 2.144915647546548, "language_loss": 0.80054951, "learning_rate": 3.029311891629966e-06, "loss": 0.82265341, "num_input_tokens_seen": 62189440, "step": 2895, "time_per_iteration": 2.5518484115600586 }, { "auxiliary_loss_clip": 0.011666, "auxiliary_loss_mlp": 0.01029394, "balance_loss_clip": 1.05602527, "balance_loss_mlp": 1.02122855, "epoch": 0.3482234112908074, "flos": 23623300138560.0, "grad_norm": 1.9335153015443696, "language_loss": 0.74417961, "learning_rate": 3.0286439257831744e-06, "loss": 0.76613957, "num_input_tokens_seen": 62208910, "step": 2896, "time_per_iteration": 2.5634818077087402 }, { "auxiliary_loss_clip": 0.01202872, "auxiliary_loss_mlp": 0.01029426, "balance_loss_clip": 1.06041479, "balance_loss_mlp": 1.02082527, "epoch": 0.3483436541814465, "flos": 23986715890560.0, "grad_norm": 1.809540463536754, "language_loss": 0.71647716, "learning_rate": 3.0279758038919156e-06, "loss": 0.73880005, "num_input_tokens_seen": 62227135, "step": 2897, "time_per_iteration": 2.524348258972168 }, { "auxiliary_loss_clip": 0.01186352, "auxiliary_loss_mlp": 0.01038265, "balance_loss_clip": 1.06026638, "balance_loss_mlp": 1.0291276, "epoch": 0.34846389707208564, "flos": 22638342244320.0, "grad_norm": 1.8357426696706902, "language_loss": 0.78318328, "learning_rate": 3.0273075260575455e-06, "loss": 0.80542946, "num_input_tokens_seen": 62246035, "step": 2898, "time_per_iteration": 2.5341744422912598 }, { "auxiliary_loss_clip": 0.01170182, "auxiliary_loss_mlp": 0.0103252, "balance_loss_clip": 1.05445933, "balance_loss_mlp": 1.02278042, "epoch": 0.3485841399627247, "flos": 21792860269440.0, "grad_norm": 1.9637304987538227, "language_loss": 0.80904424, "learning_rate": 3.0266390923814396e-06, "loss": 0.83107126, "num_input_tokens_seen": 62264095, "step": 2899, "time_per_iteration": 2.5786499977111816 }, { "auxiliary_loss_clip": 0.01175295, "auxiliary_loss_mlp": 0.01036993, "balance_loss_clip": 1.06144297, "balance_loss_mlp": 1.02901161, "epoch": 0.3487043828533638, "flos": 17019040144800.0, "grad_norm": 1.7517203024172554, "language_loss": 0.81993222, "learning_rate": 3.0259705029650008e-06, "loss": 0.84205502, "num_input_tokens_seen": 62282025, "step": 2900, "time_per_iteration": 4.06995964050293 }, { "auxiliary_loss_clip": 0.01184924, "auxiliary_loss_mlp": 0.01033402, "balance_loss_clip": 1.05727017, "balance_loss_mlp": 1.02498555, "epoch": 0.34882462574400286, "flos": 22601138662080.0, "grad_norm": 1.8217418891953248, "language_loss": 0.72724748, "learning_rate": 3.025301757909652e-06, "loss": 0.74943072, "num_input_tokens_seen": 62302220, "step": 2901, "time_per_iteration": 2.5308053493499756 }, { "auxiliary_loss_clip": 0.01161816, "auxiliary_loss_mlp": 0.00763572, "balance_loss_clip": 1.05928206, "balance_loss_mlp": 1.00028992, "epoch": 0.34894486863464197, "flos": 29861522442240.0, "grad_norm": 1.481815793121943, "language_loss": 0.80232173, "learning_rate": 3.024632857316842e-06, "loss": 0.82157564, "num_input_tokens_seen": 62323535, "step": 2902, "time_per_iteration": 2.6659631729125977 }, { "auxiliary_loss_clip": 0.01187338, "auxiliary_loss_mlp": 0.01029422, "balance_loss_clip": 1.06131482, "balance_loss_mlp": 1.02132487, "epoch": 0.3490651115252811, "flos": 22122269048640.0, "grad_norm": 2.0754774006508594, "language_loss": 0.77501088, "learning_rate": 3.0239638012880412e-06, "loss": 0.79717851, "num_input_tokens_seen": 62343430, "step": 2903, "time_per_iteration": 3.2127082347869873 }, { "auxiliary_loss_clip": 0.01133783, "auxiliary_loss_mlp": 0.01031175, "balance_loss_clip": 1.05150831, "balance_loss_mlp": 1.02253222, "epoch": 0.34918535441592014, "flos": 12676685634720.0, "grad_norm": 2.142851067360576, "language_loss": 0.81475157, "learning_rate": 3.0232945899247466e-06, "loss": 0.83640116, "num_input_tokens_seen": 62360365, "step": 2904, "time_per_iteration": 2.59560489654541 }, { "auxiliary_loss_clip": 0.01185328, "auxiliary_loss_mlp": 0.01030991, "balance_loss_clip": 1.05785668, "balance_loss_mlp": 1.02234244, "epoch": 0.34930559730655925, "flos": 23185620889920.0, "grad_norm": 1.7209812009124044, "language_loss": 0.77517939, "learning_rate": 3.022625223328476e-06, "loss": 0.79734266, "num_input_tokens_seen": 62382105, "step": 2905, "time_per_iteration": 2.546527147293091 }, { "auxiliary_loss_clip": 0.01192799, "auxiliary_loss_mlp": 0.01035785, "balance_loss_clip": 1.060992, "balance_loss_mlp": 1.02708912, "epoch": 0.34942584019719836, "flos": 22855026851040.0, "grad_norm": 3.7665932052084807, "language_loss": 0.68810642, "learning_rate": 3.0219557016007723e-06, "loss": 0.71039224, "num_input_tokens_seen": 62402235, "step": 2906, "time_per_iteration": 2.538112163543701 }, { "auxiliary_loss_clip": 0.01178811, "auxiliary_loss_mlp": 0.0103578, "balance_loss_clip": 1.05725026, "balance_loss_mlp": 1.02704775, "epoch": 0.3495460830878374, "flos": 24426047319360.0, "grad_norm": 2.4476874689487684, "language_loss": 0.69414061, "learning_rate": 3.021286024843202e-06, "loss": 0.71628654, "num_input_tokens_seen": 62420430, "step": 2907, "time_per_iteration": 2.5441579818725586 }, { "auxiliary_loss_clip": 0.01097549, "auxiliary_loss_mlp": 0.01001049, "balance_loss_clip": 1.0258615, "balance_loss_mlp": 0.99922526, "epoch": 0.3496663259784765, "flos": 70008758646240.0, "grad_norm": 1.0718253756835272, "language_loss": 0.64798868, "learning_rate": 3.0206161931573526e-06, "loss": 0.66897464, "num_input_tokens_seen": 62472980, "step": 2908, "time_per_iteration": 3.7037289142608643 }, { "auxiliary_loss_clip": 0.01167313, "auxiliary_loss_mlp": 0.01024245, "balance_loss_clip": 1.05402887, "balance_loss_mlp": 1.01624024, "epoch": 0.34978656886911563, "flos": 28692809405280.0, "grad_norm": 1.7007008516462214, "language_loss": 0.92757851, "learning_rate": 3.0199462066448388e-06, "loss": 0.94949412, "num_input_tokens_seen": 62495175, "step": 2909, "time_per_iteration": 2.6220452785491943 }, { "auxiliary_loss_clip": 0.01187923, "auxiliary_loss_mlp": 0.01031592, "balance_loss_clip": 1.06121087, "balance_loss_mlp": 1.02275264, "epoch": 0.3499068117597547, "flos": 21142159944000.0, "grad_norm": 1.675715402038092, "language_loss": 0.69499046, "learning_rate": 3.019276065407296e-06, "loss": 0.71718562, "num_input_tokens_seen": 62514295, "step": 2910, "time_per_iteration": 2.524379014968872 }, { "auxiliary_loss_clip": 0.01145634, "auxiliary_loss_mlp": 0.01032136, "balance_loss_clip": 1.05380642, "balance_loss_mlp": 1.02317786, "epoch": 0.3500270546503938, "flos": 22782702870240.0, "grad_norm": 1.6691796190093946, "language_loss": 0.80648214, "learning_rate": 3.018605769546385e-06, "loss": 0.82825983, "num_input_tokens_seen": 62534850, "step": 2911, "time_per_iteration": 2.661916971206665 }, { "auxiliary_loss_clip": 0.0118323, "auxiliary_loss_mlp": 0.01034188, "balance_loss_clip": 1.05525434, "balance_loss_mlp": 1.02507401, "epoch": 0.3501472975410329, "flos": 22894062198240.0, "grad_norm": 1.8077163950520854, "language_loss": 0.79523325, "learning_rate": 3.017935319163788e-06, "loss": 0.81740743, "num_input_tokens_seen": 62553810, "step": 2912, "time_per_iteration": 2.5314087867736816 }, { "auxiliary_loss_clip": 0.01188983, "auxiliary_loss_mlp": 0.01029937, "balance_loss_clip": 1.06138754, "balance_loss_mlp": 1.02046633, "epoch": 0.35026754043167196, "flos": 25446592532640.0, "grad_norm": 1.5831439044480033, "language_loss": 0.70369631, "learning_rate": 3.017264714361213e-06, "loss": 0.72588551, "num_input_tokens_seen": 62573460, "step": 2913, "time_per_iteration": 2.55950665473938 }, { "auxiliary_loss_clip": 0.01167692, "auxiliary_loss_mlp": 0.00764134, "balance_loss_clip": 1.05391431, "balance_loss_mlp": 1.00042367, "epoch": 0.3503877833223111, "flos": 19573761413760.0, "grad_norm": 2.544517683580567, "language_loss": 0.82175779, "learning_rate": 3.016593955240389e-06, "loss": 0.84107602, "num_input_tokens_seen": 62592150, "step": 2914, "time_per_iteration": 2.5546770095825195 }, { "auxiliary_loss_clip": 0.01084085, "auxiliary_loss_mlp": 0.01001056, "balance_loss_clip": 1.02469552, "balance_loss_mlp": 0.99923176, "epoch": 0.3505080262129502, "flos": 65072087049600.0, "grad_norm": 0.8255167973476832, "language_loss": 0.63711762, "learning_rate": 3.015923041903071e-06, "loss": 0.657969, "num_input_tokens_seen": 62658275, "step": 2915, "time_per_iteration": 3.166102886199951 }, { "auxiliary_loss_clip": 0.01184503, "auxiliary_loss_mlp": 0.01031432, "balance_loss_clip": 1.05974686, "balance_loss_mlp": 1.02312326, "epoch": 0.35062826910358924, "flos": 29314567049280.0, "grad_norm": 2.0210033834181687, "language_loss": 0.83725071, "learning_rate": 3.0152519744510347e-06, "loss": 0.85941005, "num_input_tokens_seen": 62678075, "step": 2916, "time_per_iteration": 2.5755672454833984 }, { "auxiliary_loss_clip": 0.01156912, "auxiliary_loss_mlp": 0.01030585, "balance_loss_clip": 1.05401063, "balance_loss_mlp": 1.02234745, "epoch": 0.35074851199422835, "flos": 23987721565440.0, "grad_norm": 1.8750004787365229, "language_loss": 0.82422209, "learning_rate": 3.014580752986081e-06, "loss": 0.84609705, "num_input_tokens_seen": 62696950, "step": 2917, "time_per_iteration": 2.619063138961792 }, { "auxiliary_loss_clip": 0.01146736, "auxiliary_loss_mlp": 0.01039666, "balance_loss_clip": 1.05624092, "balance_loss_mlp": 1.03101158, "epoch": 0.3508687548848674, "flos": 15224439013440.0, "grad_norm": 2.060110724342365, "language_loss": 0.78415078, "learning_rate": 3.0139093776100345e-06, "loss": 0.80601478, "num_input_tokens_seen": 62713540, "step": 2918, "time_per_iteration": 2.5817527770996094 }, { "auxiliary_loss_clip": 0.01199061, "auxiliary_loss_mlp": 0.01032481, "balance_loss_clip": 1.06067252, "balance_loss_mlp": 1.02458894, "epoch": 0.3509889977755065, "flos": 21361753824480.0, "grad_norm": 1.755939807514442, "language_loss": 0.7527023, "learning_rate": 3.013237848424741e-06, "loss": 0.77501774, "num_input_tokens_seen": 62732925, "step": 2919, "time_per_iteration": 2.5177056789398193 }, { "auxiliary_loss_clip": 0.01175521, "auxiliary_loss_mlp": 0.01032082, "balance_loss_clip": 1.06081295, "balance_loss_mlp": 1.02326083, "epoch": 0.35110924066614563, "flos": 19135363825920.0, "grad_norm": 2.2037595363351765, "language_loss": 0.75040805, "learning_rate": 3.012566165532072e-06, "loss": 0.77248406, "num_input_tokens_seen": 62751715, "step": 2920, "time_per_iteration": 2.5336101055145264 }, { "auxiliary_loss_clip": 0.01130248, "auxiliary_loss_mlp": 0.01034286, "balance_loss_clip": 1.05034792, "balance_loss_mlp": 1.02531004, "epoch": 0.3512294835567847, "flos": 21980889530400.0, "grad_norm": 2.0151114846605473, "language_loss": 0.76610237, "learning_rate": 3.0118943290339207e-06, "loss": 0.78774774, "num_input_tokens_seen": 62771925, "step": 2921, "time_per_iteration": 2.668788433074951 }, { "auxiliary_loss_clip": 0.01143626, "auxiliary_loss_mlp": 0.01030836, "balance_loss_clip": 1.05003786, "balance_loss_mlp": 1.02178788, "epoch": 0.3513497264474238, "flos": 17817297705600.0, "grad_norm": 2.0986520621083953, "language_loss": 0.68296242, "learning_rate": 3.011222339032204e-06, "loss": 0.70470703, "num_input_tokens_seen": 62790075, "step": 2922, "time_per_iteration": 2.5602686405181885 }, { "auxiliary_loss_clip": 0.01199501, "auxiliary_loss_mlp": 0.01034646, "balance_loss_clip": 1.06163812, "balance_loss_mlp": 1.02625394, "epoch": 0.3514699693380629, "flos": 26943421338240.0, "grad_norm": 2.451347411412146, "language_loss": 0.69127733, "learning_rate": 3.0105501956288626e-06, "loss": 0.71361881, "num_input_tokens_seen": 62810545, "step": 2923, "time_per_iteration": 2.5755929946899414 }, { "auxiliary_loss_clip": 0.01191672, "auxiliary_loss_mlp": 0.01033429, "balance_loss_clip": 1.05917799, "balance_loss_mlp": 1.02437544, "epoch": 0.35159021222870196, "flos": 15267568894080.0, "grad_norm": 2.614905420831368, "language_loss": 0.725353, "learning_rate": 3.0098778989258602e-06, "loss": 0.74760401, "num_input_tokens_seen": 62829155, "step": 2924, "time_per_iteration": 2.502073287963867 }, { "auxiliary_loss_clip": 0.01152543, "auxiliary_loss_mlp": 0.01034375, "balance_loss_clip": 1.0550977, "balance_loss_mlp": 1.02609587, "epoch": 0.35171045511934107, "flos": 13984156251840.0, "grad_norm": 1.9262409412502641, "language_loss": 0.88113511, "learning_rate": 3.009205449025183e-06, "loss": 0.90300429, "num_input_tokens_seen": 62845350, "step": 2925, "time_per_iteration": 2.564305305480957 }, { "auxiliary_loss_clip": 0.01147778, "auxiliary_loss_mlp": 0.01032825, "balance_loss_clip": 1.05038786, "balance_loss_mlp": 1.02380729, "epoch": 0.3518306980099802, "flos": 14283437089920.0, "grad_norm": 1.8997862177565557, "language_loss": 0.63239515, "learning_rate": 3.008532846028842e-06, "loss": 0.65420115, "num_input_tokens_seen": 62862110, "step": 2926, "time_per_iteration": 3.300067663192749 }, { "auxiliary_loss_clip": 0.01201663, "auxiliary_loss_mlp": 0.01035148, "balance_loss_clip": 1.06158805, "balance_loss_mlp": 1.02612352, "epoch": 0.35195094090061924, "flos": 27052877067360.0, "grad_norm": 2.4434147169815885, "language_loss": 0.72333556, "learning_rate": 3.0078600900388694e-06, "loss": 0.7457037, "num_input_tokens_seen": 62882415, "step": 2927, "time_per_iteration": 3.276232957839966 }, { "auxiliary_loss_clip": 0.01144014, "auxiliary_loss_mlp": 0.01028021, "balance_loss_clip": 1.04998541, "balance_loss_mlp": 1.01959896, "epoch": 0.35207118379125835, "flos": 25629270166560.0, "grad_norm": 1.7674018228932475, "language_loss": 0.73711753, "learning_rate": 3.007187181157323e-06, "loss": 0.75883782, "num_input_tokens_seen": 62902425, "step": 2928, "time_per_iteration": 2.631028175354004 }, { "auxiliary_loss_clip": 0.01111333, "auxiliary_loss_mlp": 0.0102701, "balance_loss_clip": 1.04681277, "balance_loss_mlp": 1.01907647, "epoch": 0.35219142668189746, "flos": 18004716378240.0, "grad_norm": 2.2178014338310597, "language_loss": 0.67629349, "learning_rate": 3.006514119486282e-06, "loss": 0.6976769, "num_input_tokens_seen": 62919255, "step": 2929, "time_per_iteration": 3.343348264694214 }, { "auxiliary_loss_clip": 0.01147795, "auxiliary_loss_mlp": 0.01027871, "balance_loss_clip": 1.0539465, "balance_loss_mlp": 1.01912093, "epoch": 0.3523116695725365, "flos": 14028112222560.0, "grad_norm": 4.250089063569533, "language_loss": 0.69123465, "learning_rate": 3.005840905127849e-06, "loss": 0.7129913, "num_input_tokens_seen": 62936160, "step": 2930, "time_per_iteration": 2.5714831352233887 }, { "auxiliary_loss_clip": 0.01201191, "auxiliary_loss_mlp": 0.01030982, "balance_loss_clip": 1.06302309, "balance_loss_mlp": 1.02236366, "epoch": 0.3524319124631756, "flos": 21433970054400.0, "grad_norm": 2.0546869330466593, "language_loss": 0.867226, "learning_rate": 3.0051675381841516e-06, "loss": 0.88954771, "num_input_tokens_seen": 62953470, "step": 2931, "time_per_iteration": 2.496915102005005 }, { "auxiliary_loss_clip": 0.01109309, "auxiliary_loss_mlp": 0.0076349, "balance_loss_clip": 1.04679728, "balance_loss_mlp": 1.0004909, "epoch": 0.3525521553538147, "flos": 26322777120000.0, "grad_norm": 1.5903303741265513, "language_loss": 0.76981086, "learning_rate": 3.0044940187573363e-06, "loss": 0.78853881, "num_input_tokens_seen": 62974480, "step": 2932, "time_per_iteration": 2.7771618366241455 }, { "auxiliary_loss_clip": 0.01187965, "auxiliary_loss_mlp": 0.01033306, "balance_loss_clip": 1.05872369, "balance_loss_mlp": 1.02512193, "epoch": 0.3526723982444538, "flos": 21543318032640.0, "grad_norm": 1.959437132409743, "language_loss": 0.65403384, "learning_rate": 3.003820346949578e-06, "loss": 0.67624652, "num_input_tokens_seen": 62992560, "step": 2933, "time_per_iteration": 2.7149975299835205 }, { "auxiliary_loss_clip": 0.01200452, "auxiliary_loss_mlp": 0.01031491, "balance_loss_clip": 1.06034541, "balance_loss_mlp": 1.02296746, "epoch": 0.3527926411350929, "flos": 23733653791680.0, "grad_norm": 3.040803859744263, "language_loss": 0.7986697, "learning_rate": 3.003146522863071e-06, "loss": 0.82098913, "num_input_tokens_seen": 63013445, "step": 2934, "time_per_iteration": 3.2974979877471924 }, { "auxiliary_loss_clip": 0.01175279, "auxiliary_loss_mlp": 0.0103204, "balance_loss_clip": 1.06337714, "balance_loss_mlp": 1.02343953, "epoch": 0.35291288402573195, "flos": 30445465915680.0, "grad_norm": 3.8187260591661443, "language_loss": 0.85978723, "learning_rate": 3.0024725466000345e-06, "loss": 0.88186049, "num_input_tokens_seen": 63033400, "step": 2935, "time_per_iteration": 2.6285958290100098 }, { "auxiliary_loss_clip": 0.01187288, "auxiliary_loss_mlp": 0.01025718, "balance_loss_clip": 1.06228399, "balance_loss_mlp": 1.01723599, "epoch": 0.35303312691637107, "flos": 23112183483360.0, "grad_norm": 2.5301385917144494, "language_loss": 0.78778327, "learning_rate": 3.0017984182627087e-06, "loss": 0.80991328, "num_input_tokens_seen": 63052725, "step": 2936, "time_per_iteration": 2.53010892868042 }, { "auxiliary_loss_clip": 0.0115395, "auxiliary_loss_mlp": 0.00763624, "balance_loss_clip": 1.0530498, "balance_loss_mlp": 1.00050294, "epoch": 0.3531533698070102, "flos": 21835702814400.0, "grad_norm": 2.0729058429492686, "language_loss": 0.82331383, "learning_rate": 3.00112413795336e-06, "loss": 0.8424896, "num_input_tokens_seen": 63072560, "step": 2937, "time_per_iteration": 2.609630823135376 }, { "auxiliary_loss_clip": 0.01167168, "auxiliary_loss_mlp": 0.01035946, "balance_loss_clip": 1.05248046, "balance_loss_mlp": 1.0270294, "epoch": 0.35327361269764923, "flos": 15778972884960.0, "grad_norm": 2.201567517716075, "language_loss": 0.80174303, "learning_rate": 3.000449705774275e-06, "loss": 0.82377422, "num_input_tokens_seen": 63090800, "step": 2938, "time_per_iteration": 2.5301220417022705 }, { "auxiliary_loss_clip": 0.01188995, "auxiliary_loss_mlp": 0.01028346, "balance_loss_clip": 1.06318235, "balance_loss_mlp": 1.01949477, "epoch": 0.35339385558828834, "flos": 22090417093440.0, "grad_norm": 1.8773159780758688, "language_loss": 0.71649462, "learning_rate": 2.9997751218277654e-06, "loss": 0.73866802, "num_input_tokens_seen": 63108955, "step": 2939, "time_per_iteration": 2.534149646759033 }, { "auxiliary_loss_clip": 0.01204957, "auxiliary_loss_mlp": 0.01023893, "balance_loss_clip": 1.06451273, "balance_loss_mlp": 1.01477933, "epoch": 0.35351409847892745, "flos": 24165011655360.0, "grad_norm": 2.008833095064184, "language_loss": 0.77890176, "learning_rate": 2.999100386216166e-06, "loss": 0.80119026, "num_input_tokens_seen": 63127895, "step": 2940, "time_per_iteration": 2.498992681503296 }, { "auxiliary_loss_clip": 0.01173773, "auxiliary_loss_mlp": 0.01028435, "balance_loss_clip": 1.05962992, "balance_loss_mlp": 1.02042365, "epoch": 0.3536343413695665, "flos": 27052302396000.0, "grad_norm": 2.14864666742558, "language_loss": 0.74442148, "learning_rate": 2.998425499041831e-06, "loss": 0.76644355, "num_input_tokens_seen": 63148410, "step": 2941, "time_per_iteration": 2.6045289039611816 }, { "auxiliary_loss_clip": 0.01088716, "auxiliary_loss_mlp": 0.01002519, "balance_loss_clip": 1.02953672, "balance_loss_mlp": 1.00070703, "epoch": 0.3537545842602056, "flos": 65991078264960.0, "grad_norm": 1.3149857577985453, "language_loss": 0.64613402, "learning_rate": 2.997750460407142e-06, "loss": 0.66704637, "num_input_tokens_seen": 63209765, "step": 2942, "time_per_iteration": 3.1771044731140137 }, { "auxiliary_loss_clip": 0.0116133, "auxiliary_loss_mlp": 0.0103228, "balance_loss_clip": 1.05446291, "balance_loss_mlp": 1.02327371, "epoch": 0.35387482715084473, "flos": 18436900332000.0, "grad_norm": 2.0601220701829828, "language_loss": 0.70272303, "learning_rate": 2.997075270414501e-06, "loss": 0.72465909, "num_input_tokens_seen": 63226980, "step": 2943, "time_per_iteration": 2.5722057819366455 }, { "auxiliary_loss_clip": 0.01075635, "auxiliary_loss_mlp": 0.01003092, "balance_loss_clip": 1.02727234, "balance_loss_mlp": 1.00139904, "epoch": 0.3539950700414838, "flos": 65588591248800.0, "grad_norm": 0.6998939555963019, "language_loss": 0.57776666, "learning_rate": 2.9963999291663347e-06, "loss": 0.5985539, "num_input_tokens_seen": 63292760, "step": 2944, "time_per_iteration": 3.1785409450531006 }, { "auxiliary_loss_clip": 0.01142856, "auxiliary_loss_mlp": 0.01031707, "balance_loss_clip": 1.05960608, "balance_loss_mlp": 1.022838, "epoch": 0.3541153129321229, "flos": 20521659393600.0, "grad_norm": 2.2432024160985273, "language_loss": 0.73787117, "learning_rate": 2.9957244367650915e-06, "loss": 0.75961685, "num_input_tokens_seen": 63309005, "step": 2945, "time_per_iteration": 2.590118408203125 }, { "auxiliary_loss_clip": 0.01135785, "auxiliary_loss_mlp": 0.01034871, "balance_loss_clip": 1.05252576, "balance_loss_mlp": 1.02646089, "epoch": 0.354235555822762, "flos": 19573581828960.0, "grad_norm": 2.3507445709581716, "language_loss": 0.8372876, "learning_rate": 2.9950487933132425e-06, "loss": 0.85899413, "num_input_tokens_seen": 63326420, "step": 2946, "time_per_iteration": 2.628249168395996 }, { "auxiliary_loss_clip": 0.01191969, "auxiliary_loss_mlp": 0.01034442, "balance_loss_clip": 1.06080091, "balance_loss_mlp": 1.02572751, "epoch": 0.35435579871340106, "flos": 20777271596640.0, "grad_norm": 1.94242822353045, "language_loss": 0.71807122, "learning_rate": 2.994372998913283e-06, "loss": 0.74033535, "num_input_tokens_seen": 63344925, "step": 2947, "time_per_iteration": 2.526782274246216 }, { "auxiliary_loss_clip": 0.01174812, "auxiliary_loss_mlp": 0.01027151, "balance_loss_clip": 1.06196475, "balance_loss_mlp": 1.01841879, "epoch": 0.35447604160404017, "flos": 23951810993760.0, "grad_norm": 2.2553563284626765, "language_loss": 0.6233868, "learning_rate": 2.99369705366773e-06, "loss": 0.64540642, "num_input_tokens_seen": 63365170, "step": 2948, "time_per_iteration": 2.587132215499878 }, { "auxiliary_loss_clip": 0.0116697, "auxiliary_loss_mlp": 0.01030248, "balance_loss_clip": 1.05764747, "balance_loss_mlp": 1.02123594, "epoch": 0.3545962844946792, "flos": 23435665964160.0, "grad_norm": 2.0531103491815608, "language_loss": 0.82047141, "learning_rate": 2.9930209576791244e-06, "loss": 0.84244359, "num_input_tokens_seen": 63383645, "step": 2949, "time_per_iteration": 2.557840347290039 }, { "auxiliary_loss_clip": 0.01183728, "auxiliary_loss_mlp": 0.0103097, "balance_loss_clip": 1.05964708, "balance_loss_mlp": 1.02233958, "epoch": 0.35471652738531834, "flos": 22085137300320.0, "grad_norm": 2.076190439285659, "language_loss": 0.6399107, "learning_rate": 2.9923447110500285e-06, "loss": 0.6620577, "num_input_tokens_seen": 63402390, "step": 2950, "time_per_iteration": 2.538555860519409 }, { "auxiliary_loss_clip": 0.01175254, "auxiliary_loss_mlp": 0.01032876, "balance_loss_clip": 1.05802655, "balance_loss_mlp": 1.02488279, "epoch": 0.35483677027595745, "flos": 27341885654880.0, "grad_norm": 1.510134252371497, "language_loss": 0.75496793, "learning_rate": 2.9916683138830295e-06, "loss": 0.77704924, "num_input_tokens_seen": 63423055, "step": 2951, "time_per_iteration": 2.5829222202301025 }, { "auxiliary_loss_clip": 0.01165428, "auxiliary_loss_mlp": 0.01030446, "balance_loss_clip": 1.05692697, "balance_loss_mlp": 1.02233422, "epoch": 0.3549570131665965, "flos": 13516169477280.0, "grad_norm": 2.9311631847595088, "language_loss": 0.80169821, "learning_rate": 2.9909917662807353e-06, "loss": 0.82365692, "num_input_tokens_seen": 63440855, "step": 2952, "time_per_iteration": 4.026094913482666 }, { "auxiliary_loss_clip": 0.01183116, "auxiliary_loss_mlp": 0.01027757, "balance_loss_clip": 1.05860782, "balance_loss_mlp": 1.01980019, "epoch": 0.3550772560572356, "flos": 20887553415840.0, "grad_norm": 2.1477020897020167, "language_loss": 0.69635844, "learning_rate": 2.9903150683457783e-06, "loss": 0.71846724, "num_input_tokens_seen": 63459400, "step": 2953, "time_per_iteration": 2.5245957374572754 }, { "auxiliary_loss_clip": 0.01170696, "auxiliary_loss_mlp": 0.01029226, "balance_loss_clip": 1.05747116, "balance_loss_mlp": 1.02048218, "epoch": 0.3551974989478747, "flos": 20194046462400.0, "grad_norm": 1.9574988205283501, "language_loss": 0.65185571, "learning_rate": 2.9896382201808126e-06, "loss": 0.67385495, "num_input_tokens_seen": 63476800, "step": 2954, "time_per_iteration": 2.5619051456451416 }, { "auxiliary_loss_clip": 0.01202312, "auxiliary_loss_mlp": 0.01027406, "balance_loss_clip": 1.06137156, "balance_loss_mlp": 1.01864457, "epoch": 0.3553177418385138, "flos": 19828834862400.0, "grad_norm": 2.0402898189871426, "language_loss": 0.80999267, "learning_rate": 2.988961221888516e-06, "loss": 0.83228981, "num_input_tokens_seen": 63493475, "step": 2955, "time_per_iteration": 3.2683258056640625 }, { "auxiliary_loss_clip": 0.01145259, "auxiliary_loss_mlp": 0.01028476, "balance_loss_clip": 1.0530175, "balance_loss_mlp": 1.01998782, "epoch": 0.3554379847291529, "flos": 14829135389280.0, "grad_norm": 3.19537058558109, "language_loss": 0.79129422, "learning_rate": 2.988284073571589e-06, "loss": 0.81303155, "num_input_tokens_seen": 63509560, "step": 2956, "time_per_iteration": 2.5747063159942627 }, { "auxiliary_loss_clip": 0.01186313, "auxiliary_loss_mlp": 0.00763656, "balance_loss_clip": 1.05841613, "balance_loss_mlp": 1.00040603, "epoch": 0.355558227619792, "flos": 20485353735360.0, "grad_norm": 2.234782242132192, "language_loss": 0.72742432, "learning_rate": 2.9876067753327528e-06, "loss": 0.74692404, "num_input_tokens_seen": 63527290, "step": 2957, "time_per_iteration": 2.564629554748535 }, { "auxiliary_loss_clip": 0.01189181, "auxiliary_loss_mlp": 0.01030179, "balance_loss_clip": 1.06024837, "balance_loss_mlp": 1.02197742, "epoch": 0.35567847051043106, "flos": 37663617739200.0, "grad_norm": 2.269540492225835, "language_loss": 0.80314386, "learning_rate": 2.986929327274754e-06, "loss": 0.82533741, "num_input_tokens_seen": 63547870, "step": 2958, "time_per_iteration": 2.6637816429138184 }, { "auxiliary_loss_clip": 0.01184591, "auxiliary_loss_mlp": 0.0103262, "balance_loss_clip": 1.06061816, "balance_loss_mlp": 1.02417946, "epoch": 0.35579871340107017, "flos": 26943062168640.0, "grad_norm": 1.6146735552389382, "language_loss": 0.7864272, "learning_rate": 2.9862517295003617e-06, "loss": 0.80859929, "num_input_tokens_seen": 63568285, "step": 2959, "time_per_iteration": 3.356959342956543 }, { "auxiliary_loss_clip": 0.01153249, "auxiliary_loss_mlp": 0.01026097, "balance_loss_clip": 1.05312765, "balance_loss_mlp": 1.0181216, "epoch": 0.3559189562917093, "flos": 28293339413760.0, "grad_norm": 12.744469200514436, "language_loss": 0.72571862, "learning_rate": 2.9855739821123654e-06, "loss": 0.74751198, "num_input_tokens_seen": 63589865, "step": 2960, "time_per_iteration": 2.6362690925598145 }, { "auxiliary_loss_clip": 0.0118273, "auxiliary_loss_mlp": 0.01032116, "balance_loss_clip": 1.06024289, "balance_loss_mlp": 1.0231154, "epoch": 0.35603919918234833, "flos": 25664067312480.0, "grad_norm": 1.7460072979310295, "language_loss": 0.82047403, "learning_rate": 2.98489608521358e-06, "loss": 0.84262246, "num_input_tokens_seen": 63609805, "step": 2961, "time_per_iteration": 2.5429115295410156 }, { "auxiliary_loss_clip": 0.0119017, "auxiliary_loss_mlp": 0.00763329, "balance_loss_clip": 1.05906153, "balance_loss_mlp": 1.00040102, "epoch": 0.35615944207298744, "flos": 23000860072320.0, "grad_norm": 2.067998066806839, "language_loss": 0.79799414, "learning_rate": 2.9842180389068425e-06, "loss": 0.81752908, "num_input_tokens_seen": 63627115, "step": 2962, "time_per_iteration": 2.519848585128784 }, { "auxiliary_loss_clip": 0.0106271, "auxiliary_loss_mlp": 0.01017066, "balance_loss_clip": 1.03103042, "balance_loss_mlp": 1.01564181, "epoch": 0.35627968496362655, "flos": 68251295651520.0, "grad_norm": 0.7640525417906159, "language_loss": 0.5922823, "learning_rate": 2.98353984329501e-06, "loss": 0.61308002, "num_input_tokens_seen": 63691460, "step": 2963, "time_per_iteration": 3.2070531845092773 }, { "auxiliary_loss_clip": 0.01172162, "auxiliary_loss_mlp": 0.01024432, "balance_loss_clip": 1.05940449, "balance_loss_mlp": 1.01593828, "epoch": 0.3563999278542656, "flos": 22641718438560.0, "grad_norm": 1.8479266147757327, "language_loss": 0.70655382, "learning_rate": 2.982861498480965e-06, "loss": 0.72851974, "num_input_tokens_seen": 63713840, "step": 2964, "time_per_iteration": 2.621236562728882 }, { "auxiliary_loss_clip": 0.01148579, "auxiliary_loss_mlp": 0.01027291, "balance_loss_clip": 1.05087674, "balance_loss_mlp": 1.01972747, "epoch": 0.3565201707449047, "flos": 25952537145600.0, "grad_norm": 3.3661180386945864, "language_loss": 0.82624358, "learning_rate": 2.9821830045676122e-06, "loss": 0.84800231, "num_input_tokens_seen": 63733540, "step": 2965, "time_per_iteration": 2.605005979537964 }, { "auxiliary_loss_clip": 0.01200594, "auxiliary_loss_mlp": 0.01031438, "balance_loss_clip": 1.06198525, "balance_loss_mlp": 1.02289629, "epoch": 0.3566404136355438, "flos": 28475729712000.0, "grad_norm": 1.8820256282552446, "language_loss": 0.73175955, "learning_rate": 2.9815043616578793e-06, "loss": 0.75407982, "num_input_tokens_seen": 63754335, "step": 2966, "time_per_iteration": 2.5934746265411377 }, { "auxiliary_loss_clip": 0.01151087, "auxiliary_loss_mlp": 0.01031505, "balance_loss_clip": 1.05353856, "balance_loss_mlp": 1.02345204, "epoch": 0.3567606565261829, "flos": 38363122824960.0, "grad_norm": 2.084685783702984, "language_loss": 0.77123028, "learning_rate": 2.9808255698547145e-06, "loss": 0.79305619, "num_input_tokens_seen": 63777135, "step": 2967, "time_per_iteration": 2.732470989227295 }, { "auxiliary_loss_clip": 0.0118374, "auxiliary_loss_mlp": 0.01027026, "balance_loss_clip": 1.05959678, "balance_loss_mlp": 1.01863337, "epoch": 0.356880899416822, "flos": 21981033198240.0, "grad_norm": 2.051959013286408, "language_loss": 0.79192674, "learning_rate": 2.9801466292610913e-06, "loss": 0.8140344, "num_input_tokens_seen": 63797020, "step": 2968, "time_per_iteration": 2.540628671646118 }, { "auxiliary_loss_clip": 0.01185192, "auxiliary_loss_mlp": 0.01027491, "balance_loss_clip": 1.05933762, "balance_loss_mlp": 1.01954591, "epoch": 0.35700114230746105, "flos": 18989135518080.0, "grad_norm": 2.1028347907248266, "language_loss": 0.80931544, "learning_rate": 2.979467539980003e-06, "loss": 0.83144224, "num_input_tokens_seen": 63813810, "step": 2969, "time_per_iteration": 2.513373613357544 }, { "auxiliary_loss_clip": 0.01189568, "auxiliary_loss_mlp": 0.01031887, "balance_loss_clip": 1.06275058, "balance_loss_mlp": 1.02326834, "epoch": 0.35712138519810016, "flos": 19756115795040.0, "grad_norm": 2.30507010181026, "language_loss": 0.7658326, "learning_rate": 2.978788302114468e-06, "loss": 0.78804719, "num_input_tokens_seen": 63830925, "step": 2970, "time_per_iteration": 2.528799295425415 }, { "auxiliary_loss_clip": 0.01181329, "auxiliary_loss_mlp": 0.01032973, "balance_loss_clip": 1.05842602, "balance_loss_mlp": 1.02452731, "epoch": 0.35724162808873927, "flos": 35183016299040.0, "grad_norm": 2.757081466980696, "language_loss": 0.82879961, "learning_rate": 2.9781089157675255e-06, "loss": 0.85094261, "num_input_tokens_seen": 63849385, "step": 2971, "time_per_iteration": 2.6293792724609375 }, { "auxiliary_loss_clip": 0.01177678, "auxiliary_loss_mlp": 0.01033084, "balance_loss_clip": 1.05722141, "balance_loss_mlp": 1.02501357, "epoch": 0.3573618709793783, "flos": 25556730684000.0, "grad_norm": 1.4573677332164867, "language_loss": 0.88607919, "learning_rate": 2.977429381042238e-06, "loss": 0.90818679, "num_input_tokens_seen": 63870060, "step": 2972, "time_per_iteration": 2.581264019012451 }, { "auxiliary_loss_clip": 0.01168759, "auxiliary_loss_mlp": 0.01030195, "balance_loss_clip": 1.05671811, "balance_loss_mlp": 1.02290833, "epoch": 0.35748211387001744, "flos": 29132356335840.0, "grad_norm": 2.1755711803107856, "language_loss": 0.88891953, "learning_rate": 2.9767496980416913e-06, "loss": 0.91090906, "num_input_tokens_seen": 63889355, "step": 2973, "time_per_iteration": 2.6151604652404785 }, { "auxiliary_loss_clip": 0.0115895, "auxiliary_loss_mlp": 0.01032804, "balance_loss_clip": 1.05024123, "balance_loss_mlp": 1.02380395, "epoch": 0.35760235676065655, "flos": 13954171978560.0, "grad_norm": 2.3797907693032005, "language_loss": 0.81148738, "learning_rate": 2.9760698668689914e-06, "loss": 0.8334049, "num_input_tokens_seen": 63905580, "step": 2974, "time_per_iteration": 2.5509135723114014 }, { "auxiliary_loss_clip": 0.01183101, "auxiliary_loss_mlp": 0.01033975, "balance_loss_clip": 1.05658209, "balance_loss_mlp": 1.02530837, "epoch": 0.3577225996512956, "flos": 44018694665760.0, "grad_norm": 2.5053838394585877, "language_loss": 0.71307898, "learning_rate": 2.975389887627269e-06, "loss": 0.73524976, "num_input_tokens_seen": 63928180, "step": 2975, "time_per_iteration": 2.710848093032837 }, { "auxiliary_loss_clip": 0.01158224, "auxiliary_loss_mlp": 0.01041371, "balance_loss_clip": 1.05540192, "balance_loss_mlp": 1.03364074, "epoch": 0.3578428425419347, "flos": 17055202135200.0, "grad_norm": 2.18371858620439, "language_loss": 0.90352011, "learning_rate": 2.9747097604196764e-06, "loss": 0.92551601, "num_input_tokens_seen": 63944825, "step": 2976, "time_per_iteration": 2.581364870071411 }, { "auxiliary_loss_clip": 0.01051289, "auxiliary_loss_mlp": 0.01002098, "balance_loss_clip": 1.02455521, "balance_loss_mlp": 1.00050008, "epoch": 0.3579630854325738, "flos": 71676562545120.0, "grad_norm": 0.6752607834637013, "language_loss": 0.5672431, "learning_rate": 2.9740294853493875e-06, "loss": 0.58777696, "num_input_tokens_seen": 64016385, "step": 2977, "time_per_iteration": 3.411505937576294 }, { "auxiliary_loss_clip": 0.01144138, "auxiliary_loss_mlp": 0.01030334, "balance_loss_clip": 1.05108356, "balance_loss_mlp": 1.02176905, "epoch": 0.3580833283232129, "flos": 25046655620640.0, "grad_norm": 3.1428848861634706, "language_loss": 0.66957915, "learning_rate": 2.9733490625196008e-06, "loss": 0.69132388, "num_input_tokens_seen": 64036245, "step": 2978, "time_per_iteration": 4.201448678970337 }, { "auxiliary_loss_clip": 0.01141288, "auxiliary_loss_mlp": 0.01031831, "balance_loss_clip": 1.0517658, "balance_loss_mlp": 1.02336669, "epoch": 0.358203571213852, "flos": 13953130386720.0, "grad_norm": 3.1669682852987395, "language_loss": 0.76078653, "learning_rate": 2.9726684920335353e-06, "loss": 0.78251773, "num_input_tokens_seen": 64054110, "step": 2979, "time_per_iteration": 2.630897283554077 }, { "auxiliary_loss_clip": 0.01201488, "auxiliary_loss_mlp": 0.00763832, "balance_loss_clip": 1.0607934, "balance_loss_mlp": 1.0003618, "epoch": 0.35832381410449105, "flos": 20302460599680.0, "grad_norm": 2.264152585145345, "language_loss": 0.82003492, "learning_rate": 2.971987773994432e-06, "loss": 0.83968812, "num_input_tokens_seen": 64070295, "step": 2980, "time_per_iteration": 2.513749361038208 }, { "auxiliary_loss_clip": 0.01171867, "auxiliary_loss_mlp": 0.01031753, "balance_loss_clip": 1.05308259, "balance_loss_mlp": 1.0237, "epoch": 0.35844405699513016, "flos": 16983237324000.0, "grad_norm": 1.726774248212589, "language_loss": 0.83008391, "learning_rate": 2.9713069085055566e-06, "loss": 0.85212016, "num_input_tokens_seen": 64088605, "step": 2981, "time_per_iteration": 3.2995190620422363 }, { "auxiliary_loss_clip": 0.01153197, "auxiliary_loss_mlp": 0.01031811, "balance_loss_clip": 1.05290532, "balance_loss_mlp": 1.0229125, "epoch": 0.35856429988576927, "flos": 23216862256800.0, "grad_norm": 1.5670704016932153, "language_loss": 0.78881359, "learning_rate": 2.9706258956701958e-06, "loss": 0.81066364, "num_input_tokens_seen": 64108595, "step": 2982, "time_per_iteration": 2.609666109085083 }, { "auxiliary_loss_clip": 0.01185745, "auxiliary_loss_mlp": 0.01034835, "balance_loss_clip": 1.05853748, "balance_loss_mlp": 1.02626395, "epoch": 0.3586845427764083, "flos": 23034579709440.0, "grad_norm": 2.186798115180437, "language_loss": 0.77323854, "learning_rate": 2.9699447355916575e-06, "loss": 0.79544437, "num_input_tokens_seen": 64127405, "step": 2983, "time_per_iteration": 2.5520777702331543 }, { "auxiliary_loss_clip": 0.01192524, "auxiliary_loss_mlp": 0.00763625, "balance_loss_clip": 1.055776, "balance_loss_mlp": 1.00024652, "epoch": 0.35880478566704743, "flos": 20010686406240.0, "grad_norm": 2.1036015891540076, "language_loss": 0.73569328, "learning_rate": 2.969263428373275e-06, "loss": 0.75525475, "num_input_tokens_seen": 64145755, "step": 2984, "time_per_iteration": 2.489621639251709 }, { "auxiliary_loss_clip": 0.01166189, "auxiliary_loss_mlp": 0.01027924, "balance_loss_clip": 1.05213356, "balance_loss_mlp": 1.01949024, "epoch": 0.35892502855768654, "flos": 13699098529920.0, "grad_norm": 1.943431719064895, "language_loss": 0.79276776, "learning_rate": 2.9685819741184007e-06, "loss": 0.81470883, "num_input_tokens_seen": 64164195, "step": 2985, "time_per_iteration": 3.2962679862976074 }, { "auxiliary_loss_clip": 0.01146369, "auxiliary_loss_mlp": 0.01027057, "balance_loss_clip": 1.051929, "balance_loss_mlp": 1.01883149, "epoch": 0.3590452714483256, "flos": 18114100273440.0, "grad_norm": 2.5292832977979165, "language_loss": 0.68363011, "learning_rate": 2.967900372930411e-06, "loss": 0.70536441, "num_input_tokens_seen": 64182705, "step": 2986, "time_per_iteration": 2.576646327972412 }, { "auxiliary_loss_clip": 0.01161418, "auxiliary_loss_mlp": 0.01038894, "balance_loss_clip": 1.05307949, "balance_loss_mlp": 1.03031671, "epoch": 0.3591655143389647, "flos": 17749355593920.0, "grad_norm": 2.268733350850202, "language_loss": 0.79550576, "learning_rate": 2.9672186249127046e-06, "loss": 0.81750888, "num_input_tokens_seen": 64202170, "step": 2987, "time_per_iteration": 2.5320260524749756 }, { "auxiliary_loss_clip": 0.01165939, "auxiliary_loss_mlp": 0.01025976, "balance_loss_clip": 1.05522954, "balance_loss_mlp": 1.01833439, "epoch": 0.3592857572296038, "flos": 25224412631040.0, "grad_norm": 1.8196406163635206, "language_loss": 0.7910049, "learning_rate": 2.9665367301687014e-06, "loss": 0.81292403, "num_input_tokens_seen": 64220415, "step": 2988, "time_per_iteration": 2.587845802307129 }, { "auxiliary_loss_clip": 0.0115813, "auxiliary_loss_mlp": 0.01023839, "balance_loss_clip": 1.05195582, "balance_loss_mlp": 1.01563144, "epoch": 0.3594060001202429, "flos": 29384412759840.0, "grad_norm": 1.8039788009196993, "language_loss": 0.7640667, "learning_rate": 2.965854688801845e-06, "loss": 0.78588641, "num_input_tokens_seen": 64242475, "step": 2989, "time_per_iteration": 2.6839938163757324 }, { "auxiliary_loss_clip": 0.01176498, "auxiliary_loss_mlp": 0.01029922, "balance_loss_clip": 1.05153632, "balance_loss_mlp": 1.02181578, "epoch": 0.359526243010882, "flos": 17052903449760.0, "grad_norm": 1.9332013629360898, "language_loss": 0.76404846, "learning_rate": 2.9651725009156005e-06, "loss": 0.78611267, "num_input_tokens_seen": 64260220, "step": 2990, "time_per_iteration": 2.5304949283599854 }, { "auxiliary_loss_clip": 0.01158048, "auxiliary_loss_mlp": 0.01027307, "balance_loss_clip": 1.05210948, "balance_loss_mlp": 1.01920128, "epoch": 0.3596464859015211, "flos": 22965093168480.0, "grad_norm": 2.3196318689527513, "language_loss": 0.73923957, "learning_rate": 2.964490166613454e-06, "loss": 0.76109314, "num_input_tokens_seen": 64280145, "step": 2991, "time_per_iteration": 2.564260482788086 }, { "auxiliary_loss_clip": 0.0109258, "auxiliary_loss_mlp": 0.01001572, "balance_loss_clip": 1.02247834, "balance_loss_mlp": 1.00001013, "epoch": 0.35976672879216015, "flos": 54739472126400.0, "grad_norm": 0.756876763700638, "language_loss": 0.57717776, "learning_rate": 2.963807685998917e-06, "loss": 0.59811926, "num_input_tokens_seen": 64336010, "step": 2992, "time_per_iteration": 2.92183518409729 }, { "auxiliary_loss_clip": 0.01140941, "auxiliary_loss_mlp": 0.01029908, "balance_loss_clip": 1.05109847, "balance_loss_mlp": 1.02221942, "epoch": 0.35988697168279926, "flos": 43139026133280.0, "grad_norm": 3.2271455871724273, "language_loss": 0.77954531, "learning_rate": 2.9631250591755196e-06, "loss": 0.8012538, "num_input_tokens_seen": 64358725, "step": 2993, "time_per_iteration": 2.814687728881836 }, { "auxiliary_loss_clip": 0.01160756, "auxiliary_loss_mlp": 0.0103122, "balance_loss_clip": 1.05422258, "balance_loss_mlp": 1.02268469, "epoch": 0.36000721457343837, "flos": 35845605138240.0, "grad_norm": 1.6748709036303195, "language_loss": 0.57620001, "learning_rate": 2.962442286246817e-06, "loss": 0.5981198, "num_input_tokens_seen": 64381555, "step": 2994, "time_per_iteration": 2.6829495429992676 }, { "auxiliary_loss_clip": 0.01168974, "auxiliary_loss_mlp": 0.01031161, "balance_loss_clip": 1.05511236, "balance_loss_mlp": 1.02275646, "epoch": 0.3601274574640774, "flos": 18291103027680.0, "grad_norm": 1.5971778669636807, "language_loss": 0.69725937, "learning_rate": 2.9617593673163853e-06, "loss": 0.71926075, "num_input_tokens_seen": 64400375, "step": 2995, "time_per_iteration": 2.557326078414917 }, { "auxiliary_loss_clip": 0.01166023, "auxiliary_loss_mlp": 0.01025721, "balance_loss_clip": 1.05065584, "balance_loss_mlp": 1.01826763, "epoch": 0.36024770035471654, "flos": 13333958763840.0, "grad_norm": 2.222159726069968, "language_loss": 0.77287358, "learning_rate": 2.9610763024878216e-06, "loss": 0.79479098, "num_input_tokens_seen": 64415880, "step": 2996, "time_per_iteration": 2.5585076808929443 }, { "auxiliary_loss_clip": 0.01161185, "auxiliary_loss_mlp": 0.01027785, "balance_loss_clip": 1.05239749, "balance_loss_mlp": 1.01980996, "epoch": 0.3603679432453556, "flos": 20267016948480.0, "grad_norm": 1.9012781176135742, "language_loss": 0.91500378, "learning_rate": 2.960393091864747e-06, "loss": 0.93689346, "num_input_tokens_seen": 64434260, "step": 2997, "time_per_iteration": 2.5741822719573975 }, { "auxiliary_loss_clip": 0.01164427, "auxiliary_loss_mlp": 0.01030117, "balance_loss_clip": 1.05301237, "balance_loss_mlp": 1.02199841, "epoch": 0.3604881861359947, "flos": 22451139073440.0, "grad_norm": 1.7664469162376635, "language_loss": 0.74947351, "learning_rate": 2.959709735550804e-06, "loss": 0.77141893, "num_input_tokens_seen": 64453855, "step": 2998, "time_per_iteration": 2.554081678390503 }, { "auxiliary_loss_clip": 0.01140155, "auxiliary_loss_mlp": 0.01026906, "balance_loss_clip": 1.050493, "balance_loss_mlp": 1.01895475, "epoch": 0.3606084290266338, "flos": 22054255103040.0, "grad_norm": 2.0642188323522337, "language_loss": 0.75478673, "learning_rate": 2.9590262336496575e-06, "loss": 0.77645737, "num_input_tokens_seen": 64473585, "step": 2999, "time_per_iteration": 2.619708776473999 }, { "auxiliary_loss_clip": 0.01146721, "auxiliary_loss_mlp": 0.01033552, "balance_loss_clip": 1.05210865, "balance_loss_mlp": 1.0254879, "epoch": 0.36072867191727287, "flos": 15632924161920.0, "grad_norm": 2.4045597548502, "language_loss": 0.85491639, "learning_rate": 2.9583425862649936e-06, "loss": 0.87671912, "num_input_tokens_seen": 64491720, "step": 3000, "time_per_iteration": 2.5657215118408203 }, { "auxiliary_loss_clip": 0.01198858, "auxiliary_loss_mlp": 0.01029484, "balance_loss_clip": 1.06049442, "balance_loss_mlp": 1.0207932, "epoch": 0.360848914807912, "flos": 19677003508800.0, "grad_norm": 2.303534781434989, "language_loss": 0.74508548, "learning_rate": 2.9576587935005215e-06, "loss": 0.76736891, "num_input_tokens_seen": 64509800, "step": 3001, "time_per_iteration": 2.5199687480926514 }, { "auxiliary_loss_clip": 0.0118173, "auxiliary_loss_mlp": 0.01026729, "balance_loss_clip": 1.05516124, "balance_loss_mlp": 1.01849794, "epoch": 0.3609691576985511, "flos": 18877812107040.0, "grad_norm": 2.796632335634421, "language_loss": 0.71777767, "learning_rate": 2.9569748554599713e-06, "loss": 0.73986226, "num_input_tokens_seen": 64525410, "step": 3002, "time_per_iteration": 2.4881198406219482 }, { "auxiliary_loss_clip": 0.01163644, "auxiliary_loss_mlp": 0.01025817, "balance_loss_clip": 1.05292964, "balance_loss_mlp": 1.0183785, "epoch": 0.36108940058919015, "flos": 42224093534400.0, "grad_norm": 1.8942133009242164, "language_loss": 0.73171246, "learning_rate": 2.956290772247097e-06, "loss": 0.75360703, "num_input_tokens_seen": 64544085, "step": 3003, "time_per_iteration": 3.512747287750244 }, { "auxiliary_loss_clip": 0.01127412, "auxiliary_loss_mlp": 0.01032921, "balance_loss_clip": 1.05092096, "balance_loss_mlp": 1.02544665, "epoch": 0.36120964347982926, "flos": 23185153969440.0, "grad_norm": 1.7148692987601928, "language_loss": 0.73252559, "learning_rate": 2.9556065439656724e-06, "loss": 0.75412893, "num_input_tokens_seen": 64563135, "step": 3004, "time_per_iteration": 2.606994152069092 }, { "auxiliary_loss_clip": 0.01112024, "auxiliary_loss_mlp": 0.01027381, "balance_loss_clip": 1.04357767, "balance_loss_mlp": 1.0196445, "epoch": 0.36132988637046837, "flos": 18113058681600.0, "grad_norm": 1.7166421417498132, "language_loss": 0.81612962, "learning_rate": 2.9549221707194952e-06, "loss": 0.83752364, "num_input_tokens_seen": 64581985, "step": 3005, "time_per_iteration": 3.37920880317688 }, { "auxiliary_loss_clip": 0.01185032, "auxiliary_loss_mlp": 0.01025562, "balance_loss_clip": 1.05753767, "balance_loss_mlp": 1.01757538, "epoch": 0.3614501292611074, "flos": 27813104955840.0, "grad_norm": 2.357419875852614, "language_loss": 0.72960532, "learning_rate": 2.954237652612384e-06, "loss": 0.75171125, "num_input_tokens_seen": 64601035, "step": 3006, "time_per_iteration": 3.327770471572876 }, { "auxiliary_loss_clip": 0.01161769, "auxiliary_loss_mlp": 0.0102982, "balance_loss_clip": 1.05312848, "balance_loss_mlp": 1.02231026, "epoch": 0.36157037215174653, "flos": 22634930133120.0, "grad_norm": 1.9102662022746617, "language_loss": 0.84544897, "learning_rate": 2.9535529897481796e-06, "loss": 0.86736476, "num_input_tokens_seen": 64618580, "step": 3007, "time_per_iteration": 2.5806219577789307 }, { "auxiliary_loss_clip": 0.01192936, "auxiliary_loss_mlp": 0.01030662, "balance_loss_clip": 1.05558085, "balance_loss_mlp": 1.0227282, "epoch": 0.36169061504238564, "flos": 12600841791840.0, "grad_norm": 3.2772406572565, "language_loss": 0.76642418, "learning_rate": 2.9528681822307446e-06, "loss": 0.78866017, "num_input_tokens_seen": 64635430, "step": 3008, "time_per_iteration": 2.5005035400390625 }, { "auxiliary_loss_clip": 0.01174264, "auxiliary_loss_mlp": 0.00762817, "balance_loss_clip": 1.05500734, "balance_loss_mlp": 1.00023127, "epoch": 0.3618108579330247, "flos": 26684648442720.0, "grad_norm": 2.839341449872064, "language_loss": 0.8238728, "learning_rate": 2.952183230163964e-06, "loss": 0.8432436, "num_input_tokens_seen": 64655005, "step": 3009, "time_per_iteration": 2.5827722549438477 }, { "auxiliary_loss_clip": 0.01144273, "auxiliary_loss_mlp": 0.01033548, "balance_loss_clip": 1.05054426, "balance_loss_mlp": 1.02572203, "epoch": 0.3619311008236638, "flos": 22817033095680.0, "grad_norm": 1.9807430014384846, "language_loss": 0.73114711, "learning_rate": 2.9514981336517448e-06, "loss": 0.75292528, "num_input_tokens_seen": 64674775, "step": 3010, "time_per_iteration": 2.668915271759033 }, { "auxiliary_loss_clip": 0.01176854, "auxiliary_loss_mlp": 0.01036092, "balance_loss_clip": 1.05545843, "balance_loss_mlp": 1.02799201, "epoch": 0.36205134371430286, "flos": 25919607681600.0, "grad_norm": 1.8275565681810604, "language_loss": 0.81668365, "learning_rate": 2.950812892798015e-06, "loss": 0.83881313, "num_input_tokens_seen": 64695670, "step": 3011, "time_per_iteration": 3.3250153064727783 }, { "auxiliary_loss_clip": 0.01126388, "auxiliary_loss_mlp": 0.00762479, "balance_loss_clip": 1.04882216, "balance_loss_mlp": 1.00034308, "epoch": 0.362171586604942, "flos": 26139596648640.0, "grad_norm": 1.7802687473453938, "language_loss": 0.87454641, "learning_rate": 2.9501275077067256e-06, "loss": 0.89343506, "num_input_tokens_seen": 64716290, "step": 3012, "time_per_iteration": 2.6601202487945557 }, { "auxiliary_loss_clip": 0.01103723, "auxiliary_loss_mlp": 0.01035023, "balance_loss_clip": 1.04420686, "balance_loss_mlp": 1.02756107, "epoch": 0.3622918294955811, "flos": 28074212453760.0, "grad_norm": 1.403858143007169, "language_loss": 0.88364327, "learning_rate": 2.949441978481848e-06, "loss": 0.90503073, "num_input_tokens_seen": 64737190, "step": 3013, "time_per_iteration": 2.7185003757476807 }, { "auxiliary_loss_clip": 0.01152941, "auxiliary_loss_mlp": 0.01036192, "balance_loss_clip": 1.05019307, "balance_loss_mlp": 1.02762699, "epoch": 0.36241207238622014, "flos": 19828008772320.0, "grad_norm": 2.090968010104929, "language_loss": 0.80356693, "learning_rate": 2.9487563052273778e-06, "loss": 0.82545823, "num_input_tokens_seen": 64753950, "step": 3014, "time_per_iteration": 2.59848952293396 }, { "auxiliary_loss_clip": 0.01173766, "auxiliary_loss_mlp": 0.01032985, "balance_loss_clip": 1.05486417, "balance_loss_mlp": 1.02515256, "epoch": 0.36253231527685925, "flos": 21397161558720.0, "grad_norm": 1.8415801880675453, "language_loss": 0.85388672, "learning_rate": 2.94807048804733e-06, "loss": 0.87595427, "num_input_tokens_seen": 64773570, "step": 3015, "time_per_iteration": 2.528419256210327 }, { "auxiliary_loss_clip": 0.01152045, "auxiliary_loss_mlp": 0.01034573, "balance_loss_clip": 1.04927516, "balance_loss_mlp": 1.02598953, "epoch": 0.36265255816749836, "flos": 18362888254080.0, "grad_norm": 1.9674397831107573, "language_loss": 0.90474981, "learning_rate": 2.9473845270457434e-06, "loss": 0.92661595, "num_input_tokens_seen": 64790385, "step": 3016, "time_per_iteration": 2.57247257232666 }, { "auxiliary_loss_clip": 0.01159742, "auxiliary_loss_mlp": 0.01031874, "balance_loss_clip": 1.05276966, "balance_loss_mlp": 1.02339292, "epoch": 0.3627728010581374, "flos": 18660049991520.0, "grad_norm": 2.0581723279382826, "language_loss": 0.6962285, "learning_rate": 2.946698422326677e-06, "loss": 0.71814471, "num_input_tokens_seen": 64807845, "step": 3017, "time_per_iteration": 2.5268659591674805 }, { "auxiliary_loss_clip": 0.01136133, "auxiliary_loss_mlp": 0.01027545, "balance_loss_clip": 1.04904938, "balance_loss_mlp": 1.01946831, "epoch": 0.36289304394877653, "flos": 27524275953120.0, "grad_norm": 2.1004552378776338, "language_loss": 0.79764181, "learning_rate": 2.946012173994213e-06, "loss": 0.8192786, "num_input_tokens_seen": 64827630, "step": 3018, "time_per_iteration": 2.6581246852874756 }, { "auxiliary_loss_clip": 0.01177748, "auxiliary_loss_mlp": 0.01034719, "balance_loss_clip": 1.05772448, "balance_loss_mlp": 1.02688074, "epoch": 0.36301328683941564, "flos": 34533249814560.0, "grad_norm": 1.4701564215460623, "language_loss": 0.67779356, "learning_rate": 2.945325782152454e-06, "loss": 0.69991827, "num_input_tokens_seen": 64850665, "step": 3019, "time_per_iteration": 2.6455376148223877 }, { "auxiliary_loss_clip": 0.01166009, "auxiliary_loss_mlp": 0.01028737, "balance_loss_clip": 1.0511359, "balance_loss_mlp": 1.02051795, "epoch": 0.3631335297300547, "flos": 19025979930720.0, "grad_norm": 2.0485775518076976, "language_loss": 0.78742456, "learning_rate": 2.9446392469055257e-06, "loss": 0.80937207, "num_input_tokens_seen": 64868700, "step": 3020, "time_per_iteration": 2.5440499782562256 }, { "auxiliary_loss_clip": 0.01149194, "auxiliary_loss_mlp": 0.01031214, "balance_loss_clip": 1.05685413, "balance_loss_mlp": 1.02341783, "epoch": 0.3632537726206938, "flos": 19536773333280.0, "grad_norm": 2.280304754954286, "language_loss": 0.79941142, "learning_rate": 2.9439525683575745e-06, "loss": 0.82121551, "num_input_tokens_seen": 64887620, "step": 3021, "time_per_iteration": 2.5916924476623535 }, { "auxiliary_loss_clip": 0.01200555, "auxiliary_loss_mlp": 0.01029529, "balance_loss_clip": 1.05976522, "balance_loss_mlp": 1.02074301, "epoch": 0.3633740155113329, "flos": 21068614786560.0, "grad_norm": 2.0217466464811054, "language_loss": 0.75175989, "learning_rate": 2.9432657466127694e-06, "loss": 0.77406073, "num_input_tokens_seen": 64907190, "step": 3022, "time_per_iteration": 2.497368812561035 }, { "auxiliary_loss_clip": 0.01141029, "auxiliary_loss_mlp": 0.01033192, "balance_loss_clip": 1.0562067, "balance_loss_mlp": 1.02444208, "epoch": 0.36349425840197197, "flos": 20298725235840.0, "grad_norm": 2.0204447255024136, "language_loss": 0.76719236, "learning_rate": 2.9425787817753007e-06, "loss": 0.78893459, "num_input_tokens_seen": 64925850, "step": 3023, "time_per_iteration": 2.644016742706299 }, { "auxiliary_loss_clip": 0.0115484, "auxiliary_loss_mlp": 0.01034623, "balance_loss_clip": 1.05440688, "balance_loss_mlp": 1.02578354, "epoch": 0.3636145012926111, "flos": 29716766729760.0, "grad_norm": 1.5710550722313044, "language_loss": 0.71441925, "learning_rate": 2.94189167394938e-06, "loss": 0.73631382, "num_input_tokens_seen": 64948285, "step": 3024, "time_per_iteration": 2.6327364444732666 }, { "auxiliary_loss_clip": 0.01197541, "auxiliary_loss_mlp": 0.01032752, "balance_loss_clip": 1.05983615, "balance_loss_mlp": 1.02412724, "epoch": 0.3637347441832502, "flos": 21431850953760.0, "grad_norm": 1.7053303712416175, "language_loss": 0.81083214, "learning_rate": 2.941204423239241e-06, "loss": 0.83313513, "num_input_tokens_seen": 64967160, "step": 3025, "time_per_iteration": 2.5025086402893066 }, { "auxiliary_loss_clip": 0.01178523, "auxiliary_loss_mlp": 0.01027721, "balance_loss_clip": 1.05584955, "balance_loss_mlp": 1.01941848, "epoch": 0.36385498707388925, "flos": 29533945428000.0, "grad_norm": 1.6452319771815644, "language_loss": 0.75775582, "learning_rate": 2.9405170297491395e-06, "loss": 0.7798183, "num_input_tokens_seen": 64987155, "step": 3026, "time_per_iteration": 2.5582640171051025 }, { "auxiliary_loss_clip": 0.01116317, "auxiliary_loss_mlp": 0.007639, "balance_loss_clip": 1.05294752, "balance_loss_mlp": 1.00026381, "epoch": 0.36397522996452836, "flos": 22236573567360.0, "grad_norm": 2.1676171250760907, "language_loss": 0.80430996, "learning_rate": 2.939829493583353e-06, "loss": 0.82311213, "num_input_tokens_seen": 65003800, "step": 3027, "time_per_iteration": 2.653930187225342 }, { "auxiliary_loss_clip": 0.01143153, "auxiliary_loss_mlp": 0.01023579, "balance_loss_clip": 1.04798222, "balance_loss_mlp": 1.01559174, "epoch": 0.3640954728551674, "flos": 21506509536960.0, "grad_norm": 2.851325759364033, "language_loss": 0.82957292, "learning_rate": 2.939141814846179e-06, "loss": 0.85124022, "num_input_tokens_seen": 65021215, "step": 3028, "time_per_iteration": 2.572258949279785 }, { "auxiliary_loss_clip": 0.01168342, "auxiliary_loss_mlp": 0.01028926, "balance_loss_clip": 1.05458641, "balance_loss_mlp": 1.02086163, "epoch": 0.3642157157458065, "flos": 17712870350880.0, "grad_norm": 2.858432285064649, "language_loss": 0.82387304, "learning_rate": 2.938453993641938e-06, "loss": 0.84584576, "num_input_tokens_seen": 65039590, "step": 3029, "time_per_iteration": 2.5749800205230713 }, { "auxiliary_loss_clip": 0.01168562, "auxiliary_loss_mlp": 0.01038296, "balance_loss_clip": 1.05852759, "balance_loss_mlp": 1.02968335, "epoch": 0.36433595863644563, "flos": 17639540695200.0, "grad_norm": 4.105054468072792, "language_loss": 0.69738781, "learning_rate": 2.937766030074973e-06, "loss": 0.71945643, "num_input_tokens_seen": 65056845, "step": 3030, "time_per_iteration": 4.06985878944397 }, { "auxiliary_loss_clip": 0.01157586, "auxiliary_loss_mlp": 0.01038583, "balance_loss_clip": 1.05420983, "balance_loss_mlp": 1.02934408, "epoch": 0.3644562015270847, "flos": 26833283186880.0, "grad_norm": 1.7341403788291605, "language_loss": 0.82596403, "learning_rate": 2.937077924249646e-06, "loss": 0.84792572, "num_input_tokens_seen": 65079435, "step": 3031, "time_per_iteration": 2.6325814723968506 }, { "auxiliary_loss_clip": 0.01173552, "auxiliary_loss_mlp": 0.01029662, "balance_loss_clip": 1.0556922, "balance_loss_mlp": 1.02112103, "epoch": 0.3645764444177238, "flos": 14282718750720.0, "grad_norm": 2.311523826514516, "language_loss": 0.75572681, "learning_rate": 2.9363896762703443e-06, "loss": 0.77775902, "num_input_tokens_seen": 65096500, "step": 3032, "time_per_iteration": 2.5746288299560547 }, { "auxiliary_loss_clip": 0.0119847, "auxiliary_loss_mlp": 0.01033732, "balance_loss_clip": 1.05899215, "balance_loss_mlp": 1.02483904, "epoch": 0.3646966873083629, "flos": 20667492614880.0, "grad_norm": 1.9801905287619341, "language_loss": 0.84306091, "learning_rate": 2.9357012862414725e-06, "loss": 0.86538297, "num_input_tokens_seen": 65115860, "step": 3033, "time_per_iteration": 3.3231756687164307 }, { "auxiliary_loss_clip": 0.01184335, "auxiliary_loss_mlp": 0.01032969, "balance_loss_clip": 1.05830562, "balance_loss_mlp": 1.02436185, "epoch": 0.36481693019900197, "flos": 27782617845120.0, "grad_norm": 1.9889040756676606, "language_loss": 0.72004199, "learning_rate": 2.9350127542674593e-06, "loss": 0.7422151, "num_input_tokens_seen": 65138070, "step": 3034, "time_per_iteration": 2.6324992179870605 }, { "auxiliary_loss_clip": 0.01172541, "auxiliary_loss_mlp": 0.01036251, "balance_loss_clip": 1.05690026, "balance_loss_mlp": 1.02757215, "epoch": 0.3649371730896411, "flos": 19712590827840.0, "grad_norm": 5.958275234591769, "language_loss": 0.76515162, "learning_rate": 2.934324080452755e-06, "loss": 0.78723955, "num_input_tokens_seen": 65155860, "step": 3035, "time_per_iteration": 2.5982120037078857 }, { "auxiliary_loss_clip": 0.0114246, "auxiliary_loss_mlp": 0.00764204, "balance_loss_clip": 1.0495677, "balance_loss_mlp": 1.00049496, "epoch": 0.3650574159802802, "flos": 24750499558080.0, "grad_norm": 2.0033235213062532, "language_loss": 0.77849805, "learning_rate": 2.9336352649018307e-06, "loss": 0.79756469, "num_input_tokens_seen": 65175930, "step": 3036, "time_per_iteration": 3.312333106994629 }, { "auxiliary_loss_clip": 0.01168773, "auxiliary_loss_mlp": 0.01032188, "balance_loss_clip": 1.05628657, "balance_loss_mlp": 1.02309847, "epoch": 0.36517765887091924, "flos": 32853491956320.0, "grad_norm": 1.740357604364964, "language_loss": 0.7018792, "learning_rate": 2.9329463077191783e-06, "loss": 0.72388875, "num_input_tokens_seen": 65199305, "step": 3037, "time_per_iteration": 2.65647292137146 }, { "auxiliary_loss_clip": 0.01137649, "auxiliary_loss_mlp": 0.01032116, "balance_loss_clip": 1.05142474, "balance_loss_mlp": 1.02401555, "epoch": 0.36529790176155835, "flos": 20120321720160.0, "grad_norm": 1.9621715953071401, "language_loss": 0.63958359, "learning_rate": 2.9322572090093135e-06, "loss": 0.66128123, "num_input_tokens_seen": 65218010, "step": 3038, "time_per_iteration": 2.5977261066436768 }, { "auxiliary_loss_clip": 0.0113504, "auxiliary_loss_mlp": 0.01035287, "balance_loss_clip": 1.04932141, "balance_loss_mlp": 1.02649534, "epoch": 0.36541814465219746, "flos": 17639576612160.0, "grad_norm": 2.7997543438727805, "language_loss": 0.76738262, "learning_rate": 2.9315679688767713e-06, "loss": 0.78908587, "num_input_tokens_seen": 65236020, "step": 3039, "time_per_iteration": 2.608065128326416 }, { "auxiliary_loss_clip": 0.01158171, "auxiliary_loss_mlp": 0.01033203, "balance_loss_clip": 1.04962754, "balance_loss_mlp": 1.0256691, "epoch": 0.3655383875428365, "flos": 22674360566880.0, "grad_norm": 1.482703044599186, "language_loss": 0.66553903, "learning_rate": 2.9308785874261085e-06, "loss": 0.68745279, "num_input_tokens_seen": 65256210, "step": 3040, "time_per_iteration": 2.5604214668273926 }, { "auxiliary_loss_clip": 0.01196039, "auxiliary_loss_mlp": 0.01034551, "balance_loss_clip": 1.05782104, "balance_loss_mlp": 1.02577102, "epoch": 0.36565863043347563, "flos": 21981176866080.0, "grad_norm": 1.8744360419576298, "language_loss": 0.81837595, "learning_rate": 2.9301890647619045e-06, "loss": 0.84068191, "num_input_tokens_seen": 65275505, "step": 3041, "time_per_iteration": 2.524866819381714 }, { "auxiliary_loss_clip": 0.01175381, "auxiliary_loss_mlp": 0.01028706, "balance_loss_clip": 1.05817378, "balance_loss_mlp": 1.02064788, "epoch": 0.36577887332411474, "flos": 24827636411520.0, "grad_norm": 2.3589867984244606, "language_loss": 0.8060177, "learning_rate": 2.929499400988759e-06, "loss": 0.8280586, "num_input_tokens_seen": 65296665, "step": 3042, "time_per_iteration": 2.581460952758789 }, { "auxiliary_loss_clip": 0.01179383, "auxiliary_loss_mlp": 0.01033786, "balance_loss_clip": 1.05483818, "balance_loss_mlp": 1.02505946, "epoch": 0.3658991162147538, "flos": 28293195745920.0, "grad_norm": 1.7337751437614144, "language_loss": 0.64779019, "learning_rate": 2.9288095962112927e-06, "loss": 0.66992188, "num_input_tokens_seen": 65317370, "step": 3043, "time_per_iteration": 2.592371940612793 }, { "auxiliary_loss_clip": 0.01196294, "auxiliary_loss_mlp": 0.01029177, "balance_loss_clip": 1.05799472, "balance_loss_mlp": 1.02117801, "epoch": 0.3660193591053929, "flos": 17785553501280.0, "grad_norm": 1.743317166360319, "language_loss": 0.8503058, "learning_rate": 2.9281196505341503e-06, "loss": 0.87256056, "num_input_tokens_seen": 65334540, "step": 3044, "time_per_iteration": 2.4588301181793213 }, { "auxiliary_loss_clip": 0.01135075, "auxiliary_loss_mlp": 0.00763457, "balance_loss_clip": 1.05330718, "balance_loss_mlp": 1.00030184, "epoch": 0.36613960199603196, "flos": 10342779422880.0, "grad_norm": 2.010816107835628, "language_loss": 0.78682721, "learning_rate": 2.9274295640619946e-06, "loss": 0.8058126, "num_input_tokens_seen": 65351670, "step": 3045, "time_per_iteration": 2.646984577178955 }, { "auxiliary_loss_clip": 0.01146755, "auxiliary_loss_mlp": 0.01028041, "balance_loss_clip": 1.04676771, "balance_loss_mlp": 1.02016747, "epoch": 0.36625984488667107, "flos": 19755612957600.0, "grad_norm": 1.7439319030335712, "language_loss": 0.78531587, "learning_rate": 2.9267393368995103e-06, "loss": 0.80706388, "num_input_tokens_seen": 65370900, "step": 3046, "time_per_iteration": 2.609058141708374 }, { "auxiliary_loss_clip": 0.01199357, "auxiliary_loss_mlp": 0.01036074, "balance_loss_clip": 1.06007719, "balance_loss_mlp": 1.02688932, "epoch": 0.3663800877773102, "flos": 17674266007200.0, "grad_norm": 2.3641517343013456, "language_loss": 0.74710286, "learning_rate": 2.926048969151407e-06, "loss": 0.76945722, "num_input_tokens_seen": 65388185, "step": 3047, "time_per_iteration": 2.515082359313965 }, { "auxiliary_loss_clip": 0.01134803, "auxiliary_loss_mlp": 0.01033976, "balance_loss_clip": 1.05526876, "balance_loss_mlp": 1.0251013, "epoch": 0.36650033066794924, "flos": 20303609942400.0, "grad_norm": 1.7913774165018377, "language_loss": 0.68755615, "learning_rate": 2.92535846092241e-06, "loss": 0.70924401, "num_input_tokens_seen": 65407200, "step": 3048, "time_per_iteration": 2.6396749019622803 }, { "auxiliary_loss_clip": 0.01170107, "auxiliary_loss_mlp": 0.01030704, "balance_loss_clip": 1.05649066, "balance_loss_mlp": 1.02169192, "epoch": 0.36662057355858835, "flos": 24716241166560.0, "grad_norm": 3.6008281954429173, "language_loss": 0.82660437, "learning_rate": 2.9246678123172704e-06, "loss": 0.84861243, "num_input_tokens_seen": 65427290, "step": 3049, "time_per_iteration": 2.644038677215576 }, { "auxiliary_loss_clip": 0.01196876, "auxiliary_loss_mlp": 0.01026818, "balance_loss_clip": 1.05738997, "balance_loss_mlp": 1.01797891, "epoch": 0.36674081644922746, "flos": 12385270610880.0, "grad_norm": 2.4517040814841606, "language_loss": 0.74129236, "learning_rate": 2.9239770234407596e-06, "loss": 0.7635293, "num_input_tokens_seen": 65445595, "step": 3050, "time_per_iteration": 2.486475706100464 }, { "auxiliary_loss_clip": 0.01181763, "auxiliary_loss_mlp": 0.01034246, "balance_loss_clip": 1.05436742, "balance_loss_mlp": 1.02504253, "epoch": 0.3668610593398665, "flos": 21105926119680.0, "grad_norm": 2.027093280273077, "language_loss": 0.6821034, "learning_rate": 2.9232860943976686e-06, "loss": 0.70426345, "num_input_tokens_seen": 65466330, "step": 3051, "time_per_iteration": 2.577594518661499 }, { "auxiliary_loss_clip": 0.01164506, "auxiliary_loss_mlp": 0.01029621, "balance_loss_clip": 1.05360615, "balance_loss_mlp": 1.02193809, "epoch": 0.3669813022305056, "flos": 26758085849280.0, "grad_norm": 1.7389711932647152, "language_loss": 0.84005451, "learning_rate": 2.9225950252928115e-06, "loss": 0.86199582, "num_input_tokens_seen": 65487180, "step": 3052, "time_per_iteration": 2.632044792175293 }, { "auxiliary_loss_clip": 0.01181036, "auxiliary_loss_mlp": 0.01046571, "balance_loss_clip": 1.05614448, "balance_loss_mlp": 1.03696299, "epoch": 0.36710154512114473, "flos": 19099525088160.0, "grad_norm": 2.1861925765676977, "language_loss": 0.8204295, "learning_rate": 2.9219038162310217e-06, "loss": 0.84270561, "num_input_tokens_seen": 65505380, "step": 3053, "time_per_iteration": 2.54262113571167 }, { "auxiliary_loss_clip": 0.01109931, "auxiliary_loss_mlp": 0.00763972, "balance_loss_clip": 1.0479331, "balance_loss_mlp": 1.00041103, "epoch": 0.3672217880117838, "flos": 20812032825600.0, "grad_norm": 2.030565162424716, "language_loss": 0.82406431, "learning_rate": 2.921212467317157e-06, "loss": 0.84280336, "num_input_tokens_seen": 65524825, "step": 3054, "time_per_iteration": 2.6975748538970947 }, { "auxiliary_loss_clip": 0.01150588, "auxiliary_loss_mlp": 0.01030948, "balance_loss_clip": 1.04853213, "balance_loss_mlp": 1.02170336, "epoch": 0.3673420309024229, "flos": 13590397056960.0, "grad_norm": 1.8818753764500344, "language_loss": 0.80114543, "learning_rate": 2.920520978656093e-06, "loss": 0.82296085, "num_input_tokens_seen": 65541790, "step": 3055, "time_per_iteration": 2.578246831893921 }, { "auxiliary_loss_clip": 0.01191489, "auxiliary_loss_mlp": 0.00763335, "balance_loss_clip": 1.05514979, "balance_loss_mlp": 1.00042176, "epoch": 0.367462273793062, "flos": 28986882284160.0, "grad_norm": 1.8868461044026654, "language_loss": 0.76817, "learning_rate": 2.919829350352729e-06, "loss": 0.7877183, "num_input_tokens_seen": 65563395, "step": 3056, "time_per_iteration": 4.082438230514526 }, { "auxiliary_loss_clip": 0.01088433, "auxiliary_loss_mlp": 0.01005305, "balance_loss_clip": 1.01860738, "balance_loss_mlp": 1.00374305, "epoch": 0.36758251668370107, "flos": 62643157338240.0, "grad_norm": 0.7521591184877938, "language_loss": 0.6002363, "learning_rate": 2.919137582511983e-06, "loss": 0.62117374, "num_input_tokens_seen": 65619835, "step": 3057, "time_per_iteration": 3.0376672744750977 }, { "auxiliary_loss_clip": 0.01160308, "auxiliary_loss_mlp": 0.01036059, "balance_loss_clip": 1.05677617, "balance_loss_mlp": 1.02697515, "epoch": 0.3677027595743402, "flos": 12713889216960.0, "grad_norm": 6.6052633601230095, "language_loss": 0.64131689, "learning_rate": 2.918445675238797e-06, "loss": 0.66328061, "num_input_tokens_seen": 65636760, "step": 3058, "time_per_iteration": 2.624366521835327 }, { "auxiliary_loss_clip": 0.01195415, "auxiliary_loss_mlp": 0.0102879, "balance_loss_clip": 1.05629563, "balance_loss_mlp": 1.02008748, "epoch": 0.36782300246497923, "flos": 25046583786720.0, "grad_norm": 1.6719342268401063, "language_loss": 0.70172238, "learning_rate": 2.917753628638132e-06, "loss": 0.72396445, "num_input_tokens_seen": 65657065, "step": 3059, "time_per_iteration": 3.3623299598693848 }, { "auxiliary_loss_clip": 0.01169724, "auxiliary_loss_mlp": 0.01031195, "balance_loss_clip": 1.05701756, "balance_loss_mlp": 1.02274895, "epoch": 0.36794324535561834, "flos": 17419515811200.0, "grad_norm": 2.166057377100904, "language_loss": 0.7009325, "learning_rate": 2.9170614428149716e-06, "loss": 0.72294164, "num_input_tokens_seen": 65675400, "step": 3060, "time_per_iteration": 2.557335138320923 }, { "auxiliary_loss_clip": 0.01150248, "auxiliary_loss_mlp": 0.01033541, "balance_loss_clip": 1.05399823, "balance_loss_mlp": 1.02474976, "epoch": 0.36806348824625745, "flos": 24089131895520.0, "grad_norm": 2.159820271042959, "language_loss": 0.87200022, "learning_rate": 2.9163691178743195e-06, "loss": 0.89383811, "num_input_tokens_seen": 65694050, "step": 3061, "time_per_iteration": 2.582792043685913 }, { "auxiliary_loss_clip": 0.01175659, "auxiliary_loss_mlp": 0.01025896, "balance_loss_clip": 1.05352306, "balance_loss_mlp": 1.01786721, "epoch": 0.3681837311368965, "flos": 20521874895360.0, "grad_norm": 2.7036909440206895, "language_loss": 0.77473307, "learning_rate": 2.9156766539212006e-06, "loss": 0.79674864, "num_input_tokens_seen": 65711695, "step": 3062, "time_per_iteration": 2.515385389328003 }, { "auxiliary_loss_clip": 0.01186741, "auxiliary_loss_mlp": 0.01029294, "balance_loss_clip": 1.05619144, "balance_loss_mlp": 1.02139616, "epoch": 0.3683039740275356, "flos": 21466648099680.0, "grad_norm": 1.9232127263761234, "language_loss": 0.72042775, "learning_rate": 2.9149840510606614e-06, "loss": 0.74258816, "num_input_tokens_seen": 65730350, "step": 3063, "time_per_iteration": 3.2932159900665283 }, { "auxiliary_loss_clip": 0.01074501, "auxiliary_loss_mlp": 0.00754388, "balance_loss_clip": 1.01784921, "balance_loss_mlp": 1.00005674, "epoch": 0.36842421691817473, "flos": 70381009970400.0, "grad_norm": 1.0211071158124025, "language_loss": 0.64153641, "learning_rate": 2.914291309397769e-06, "loss": 0.65982527, "num_input_tokens_seen": 65787820, "step": 3064, "time_per_iteration": 3.2050201892852783 }, { "auxiliary_loss_clip": 0.01113712, "auxiliary_loss_mlp": 0.01031891, "balance_loss_clip": 1.04812002, "balance_loss_mlp": 1.0230217, "epoch": 0.3685444598088138, "flos": 23331382277280.0, "grad_norm": 2.0336530110017317, "language_loss": 0.78468168, "learning_rate": 2.9135984290376117e-06, "loss": 0.80613774, "num_input_tokens_seen": 65806685, "step": 3065, "time_per_iteration": 2.6749985218048096 }, { "auxiliary_loss_clip": 0.01121366, "auxiliary_loss_mlp": 0.01032285, "balance_loss_clip": 1.04990816, "balance_loss_mlp": 1.02360654, "epoch": 0.3686647026994529, "flos": 23070274779360.0, "grad_norm": 1.728572811766545, "language_loss": 0.8279475, "learning_rate": 2.9129054100853e-06, "loss": 0.84948409, "num_input_tokens_seen": 65825525, "step": 3066, "time_per_iteration": 2.677377223968506 }, { "auxiliary_loss_clip": 0.01169055, "auxiliary_loss_mlp": 0.01028838, "balance_loss_clip": 1.05425549, "balance_loss_mlp": 1.02030849, "epoch": 0.368784945590092, "flos": 25119913442400.0, "grad_norm": 1.7640727780211145, "language_loss": 0.76123846, "learning_rate": 2.912212252645963e-06, "loss": 0.78321743, "num_input_tokens_seen": 65848110, "step": 3067, "time_per_iteration": 2.6032421588897705 }, { "auxiliary_loss_clip": 0.0118789, "auxiliary_loss_mlp": 0.01032078, "balance_loss_clip": 1.05649745, "balance_loss_mlp": 1.02185595, "epoch": 0.36890518848073106, "flos": 18442287876000.0, "grad_norm": 2.130363145041467, "language_loss": 0.76003391, "learning_rate": 2.9115189568247523e-06, "loss": 0.7822336, "num_input_tokens_seen": 65865670, "step": 3068, "time_per_iteration": 2.513254165649414 }, { "auxiliary_loss_clip": 0.01129933, "auxiliary_loss_mlp": 0.01032891, "balance_loss_clip": 1.05522001, "balance_loss_mlp": 1.02412927, "epoch": 0.36902543137137017, "flos": 16362449437920.0, "grad_norm": 2.0655297916251216, "language_loss": 0.92162877, "learning_rate": 2.910825522726841e-06, "loss": 0.94325697, "num_input_tokens_seen": 65883195, "step": 3069, "time_per_iteration": 2.5896565914154053 }, { "auxiliary_loss_clip": 0.01130918, "auxiliary_loss_mlp": 0.0102795, "balance_loss_clip": 1.0497458, "balance_loss_mlp": 1.01916993, "epoch": 0.3691456742620093, "flos": 12275599380000.0, "grad_norm": 2.050597671845055, "language_loss": 0.77743345, "learning_rate": 2.9101319504574215e-06, "loss": 0.79902214, "num_input_tokens_seen": 65899635, "step": 3070, "time_per_iteration": 2.6174535751342773 }, { "auxiliary_loss_clip": 0.01171083, "auxiliary_loss_mlp": 0.01031399, "balance_loss_clip": 1.05191576, "balance_loss_mlp": 1.02217174, "epoch": 0.36926591715264834, "flos": 17786415508320.0, "grad_norm": 6.6718355881851465, "language_loss": 0.76466149, "learning_rate": 2.909438240121709e-06, "loss": 0.7866863, "num_input_tokens_seen": 65919910, "step": 3071, "time_per_iteration": 2.5596413612365723 }, { "auxiliary_loss_clip": 0.01163769, "auxiliary_loss_mlp": 0.01031941, "balance_loss_clip": 1.05669546, "balance_loss_mlp": 1.02315545, "epoch": 0.36938616004328745, "flos": 28948313857440.0, "grad_norm": 1.6909789356608709, "language_loss": 0.70340359, "learning_rate": 2.908744391824939e-06, "loss": 0.72536069, "num_input_tokens_seen": 65940930, "step": 3072, "time_per_iteration": 2.599886655807495 }, { "auxiliary_loss_clip": 0.01123473, "auxiliary_loss_mlp": 0.01031363, "balance_loss_clip": 1.04538536, "balance_loss_mlp": 1.02264285, "epoch": 0.36950640293392656, "flos": 29205398655840.0, "grad_norm": 1.756318168086548, "language_loss": 0.79138643, "learning_rate": 2.908050405672367e-06, "loss": 0.81293482, "num_input_tokens_seen": 65960475, "step": 3073, "time_per_iteration": 2.696974039077759 }, { "auxiliary_loss_clip": 0.01172432, "auxiliary_loss_mlp": 0.01033189, "balance_loss_clip": 1.05136728, "balance_loss_mlp": 1.02469516, "epoch": 0.3696266458245656, "flos": 24827780079360.0, "grad_norm": 1.678632172456899, "language_loss": 0.79008198, "learning_rate": 2.9073562817692703e-06, "loss": 0.8121382, "num_input_tokens_seen": 65979160, "step": 3074, "time_per_iteration": 2.582253932952881 }, { "auxiliary_loss_clip": 0.01055632, "auxiliary_loss_mlp": 0.01004809, "balance_loss_clip": 1.02169311, "balance_loss_mlp": 1.00333679, "epoch": 0.3697468887152047, "flos": 59887267589280.0, "grad_norm": 0.7189822512609002, "language_loss": 0.56528735, "learning_rate": 2.9066620202209468e-06, "loss": 0.58589172, "num_input_tokens_seen": 66041650, "step": 3075, "time_per_iteration": 3.1715009212493896 }, { "auxiliary_loss_clip": 0.011449, "auxiliary_loss_mlp": 0.01032195, "balance_loss_clip": 1.05128646, "balance_loss_mlp": 1.0233202, "epoch": 0.3698671316058438, "flos": 26137585298880.0, "grad_norm": 2.1480592834964076, "language_loss": 0.78017926, "learning_rate": 2.905967621132716e-06, "loss": 0.80195022, "num_input_tokens_seen": 66059260, "step": 3076, "time_per_iteration": 2.6249778270721436 }, { "auxiliary_loss_clip": 0.0117463, "auxiliary_loss_mlp": 0.01031776, "balance_loss_clip": 1.05647326, "balance_loss_mlp": 1.02265072, "epoch": 0.3699873744964829, "flos": 24607467859680.0, "grad_norm": 3.418731134081714, "language_loss": 0.75267529, "learning_rate": 2.9052730846099172e-06, "loss": 0.77473938, "num_input_tokens_seen": 66080605, "step": 3077, "time_per_iteration": 2.583348035812378 }, { "auxiliary_loss_clip": 0.01062019, "auxiliary_loss_mlp": 0.01002363, "balance_loss_clip": 1.01437616, "balance_loss_mlp": 1.00068235, "epoch": 0.370107617387122, "flos": 64885350799200.0, "grad_norm": 0.8574047012687938, "language_loss": 0.60954195, "learning_rate": 2.9045784107579123e-06, "loss": 0.63018578, "num_input_tokens_seen": 66140710, "step": 3078, "time_per_iteration": 3.1702263355255127 }, { "auxiliary_loss_clip": 0.01197564, "auxiliary_loss_mlp": 0.01030027, "balance_loss_clip": 1.05910563, "balance_loss_mlp": 1.02139664, "epoch": 0.37022786027776106, "flos": 15961686435840.0, "grad_norm": 1.9605775563734873, "language_loss": 0.67185098, "learning_rate": 2.9038835996820807e-06, "loss": 0.69412684, "num_input_tokens_seen": 66158320, "step": 3079, "time_per_iteration": 2.45981764793396 }, { "auxiliary_loss_clip": 0.01156606, "auxiliary_loss_mlp": 0.01035348, "balance_loss_clip": 1.04992461, "balance_loss_mlp": 1.02714074, "epoch": 0.37034810316840017, "flos": 18546930732480.0, "grad_norm": 1.7038119963823652, "language_loss": 0.79384363, "learning_rate": 2.903188651487826e-06, "loss": 0.81576318, "num_input_tokens_seen": 66176875, "step": 3080, "time_per_iteration": 2.574950695037842 }, { "auxiliary_loss_clip": 0.01183586, "auxiliary_loss_mlp": 0.01043458, "balance_loss_clip": 1.05532157, "balance_loss_mlp": 1.03508401, "epoch": 0.3704683460590393, "flos": 17821930993440.0, "grad_norm": 2.134550087448077, "language_loss": 0.86497676, "learning_rate": 2.902493566280571e-06, "loss": 0.8872472, "num_input_tokens_seen": 66194980, "step": 3081, "time_per_iteration": 2.484591484069824 }, { "auxiliary_loss_clip": 0.01169835, "auxiliary_loss_mlp": 0.01037298, "balance_loss_clip": 1.05874014, "balance_loss_mlp": 1.02823782, "epoch": 0.37058858894967833, "flos": 14134084006560.0, "grad_norm": 1.9473263971761254, "language_loss": 0.81181979, "learning_rate": 2.9017983441657595e-06, "loss": 0.83389115, "num_input_tokens_seen": 66212310, "step": 3082, "time_per_iteration": 4.010421991348267 }, { "auxiliary_loss_clip": 0.01140912, "auxiliary_loss_mlp": 0.01031479, "balance_loss_clip": 1.04920244, "balance_loss_mlp": 1.02284861, "epoch": 0.37070883184031744, "flos": 13954495231200.0, "grad_norm": 2.0298762992044836, "language_loss": 0.75169647, "learning_rate": 2.9011029852488564e-06, "loss": 0.77342033, "num_input_tokens_seen": 66229545, "step": 3083, "time_per_iteration": 2.605228900909424 }, { "auxiliary_loss_clip": 0.01083643, "auxiliary_loss_mlp": 0.01000554, "balance_loss_clip": 1.01486349, "balance_loss_mlp": 0.99906427, "epoch": 0.37082907473095655, "flos": 52315427121600.0, "grad_norm": 1.0105493369951242, "language_loss": 0.62526584, "learning_rate": 2.9004074896353465e-06, "loss": 0.64610779, "num_input_tokens_seen": 66283545, "step": 3084, "time_per_iteration": 2.972794771194458 }, { "auxiliary_loss_clip": 0.0119779, "auxiliary_loss_mlp": 0.01031955, "balance_loss_clip": 1.06078672, "balance_loss_mlp": 1.0237236, "epoch": 0.3709493176215956, "flos": 15998099844960.0, "grad_norm": 2.3513832265802708, "language_loss": 0.81658602, "learning_rate": 2.8997118574307362e-06, "loss": 0.83888352, "num_input_tokens_seen": 66300500, "step": 3085, "time_per_iteration": 3.1975104808807373 }, { "auxiliary_loss_clip": 0.01162698, "auxiliary_loss_mlp": 0.01035435, "balance_loss_clip": 1.05621541, "balance_loss_mlp": 1.02666688, "epoch": 0.3710695605122347, "flos": 20959841479680.0, "grad_norm": 2.1267491648373484, "language_loss": 0.74834031, "learning_rate": 2.899016088740553e-06, "loss": 0.77032167, "num_input_tokens_seen": 66318610, "step": 3086, "time_per_iteration": 2.562715768814087 }, { "auxiliary_loss_clip": 0.01139923, "auxiliary_loss_mlp": 0.01031094, "balance_loss_clip": 1.05437422, "balance_loss_mlp": 1.02271402, "epoch": 0.37118980340287383, "flos": 14355581485920.0, "grad_norm": 1.8860145986487777, "language_loss": 0.78986961, "learning_rate": 2.898320183670344e-06, "loss": 0.81157976, "num_input_tokens_seen": 66336025, "step": 3087, "time_per_iteration": 2.603511333465576 }, { "auxiliary_loss_clip": 0.01138557, "auxiliary_loss_mlp": 0.01033359, "balance_loss_clip": 1.05278301, "balance_loss_mlp": 1.02492476, "epoch": 0.3713100462935129, "flos": 25885385207040.0, "grad_norm": 1.7821298511823407, "language_loss": 0.89230639, "learning_rate": 2.8976241423256767e-06, "loss": 0.91402555, "num_input_tokens_seen": 66356120, "step": 3088, "time_per_iteration": 3.393702268600464 }, { "auxiliary_loss_clip": 0.01161501, "auxiliary_loss_mlp": 0.01029239, "balance_loss_clip": 1.05353129, "balance_loss_mlp": 1.02170765, "epoch": 0.371430289184152, "flos": 30518939239200.0, "grad_norm": 1.936966495399612, "language_loss": 0.67939478, "learning_rate": 2.896927964812142e-06, "loss": 0.70130217, "num_input_tokens_seen": 66376685, "step": 3089, "time_per_iteration": 2.5984702110290527 }, { "auxiliary_loss_clip": 0.01166283, "auxiliary_loss_mlp": 0.01026119, "balance_loss_clip": 1.05675077, "balance_loss_mlp": 1.01770926, "epoch": 0.37155053207479105, "flos": 15742236223200.0, "grad_norm": 2.7458945570372397, "language_loss": 0.7462604, "learning_rate": 2.8962316512353465e-06, "loss": 0.76818442, "num_input_tokens_seen": 66394230, "step": 3090, "time_per_iteration": 2.5183002948760986 }, { "auxiliary_loss_clip": 0.01117094, "auxiliary_loss_mlp": 0.01029009, "balance_loss_clip": 1.04759288, "balance_loss_mlp": 1.02042556, "epoch": 0.37167077496543016, "flos": 23404065427680.0, "grad_norm": 1.8442890312626719, "language_loss": 0.75093389, "learning_rate": 2.8955352017009233e-06, "loss": 0.7723949, "num_input_tokens_seen": 66413475, "step": 3091, "time_per_iteration": 2.6641647815704346 }, { "auxiliary_loss_clip": 0.01168669, "auxiliary_loss_mlp": 0.01030507, "balance_loss_clip": 1.0590651, "balance_loss_mlp": 1.02133989, "epoch": 0.3717910178560693, "flos": 22088657162400.0, "grad_norm": 1.9736503894670276, "language_loss": 0.77418709, "learning_rate": 2.8948386163145212e-06, "loss": 0.79617888, "num_input_tokens_seen": 66432685, "step": 3092, "time_per_iteration": 2.5494978427886963 }, { "auxiliary_loss_clip": 0.01185302, "auxiliary_loss_mlp": 0.0103464, "balance_loss_clip": 1.05842054, "balance_loss_mlp": 1.02690339, "epoch": 0.3719112607467083, "flos": 26939973310080.0, "grad_norm": 1.963409659403866, "language_loss": 0.79910505, "learning_rate": 2.8941418951818135e-06, "loss": 0.82130456, "num_input_tokens_seen": 66452245, "step": 3093, "time_per_iteration": 2.535125970840454 }, { "auxiliary_loss_clip": 0.01153787, "auxiliary_loss_mlp": 0.01028572, "balance_loss_clip": 1.05267584, "balance_loss_mlp": 1.02045345, "epoch": 0.37203150363734744, "flos": 12166502820480.0, "grad_norm": 2.157392711969657, "language_loss": 0.71018386, "learning_rate": 2.8934450384084903e-06, "loss": 0.73200738, "num_input_tokens_seen": 66469760, "step": 3094, "time_per_iteration": 2.5601444244384766 }, { "auxiliary_loss_clip": 0.01158905, "auxiliary_loss_mlp": 0.01033362, "balance_loss_clip": 1.05162907, "balance_loss_mlp": 1.02430248, "epoch": 0.37215174652798655, "flos": 23697599552160.0, "grad_norm": 1.9318456620301785, "language_loss": 0.70009315, "learning_rate": 2.8927480461002653e-06, "loss": 0.7220158, "num_input_tokens_seen": 66489730, "step": 3095, "time_per_iteration": 2.559828042984009 }, { "auxiliary_loss_clip": 0.01165957, "auxiliary_loss_mlp": 0.01035583, "balance_loss_clip": 1.05273926, "balance_loss_mlp": 1.02606452, "epoch": 0.3722719894186256, "flos": 17887754004480.0, "grad_norm": 2.359502568973712, "language_loss": 0.86275446, "learning_rate": 2.892050918362872e-06, "loss": 0.88476992, "num_input_tokens_seen": 66504785, "step": 3096, "time_per_iteration": 2.5248641967773438 }, { "auxiliary_loss_clip": 0.01010924, "auxiliary_loss_mlp": 0.0100449, "balance_loss_clip": 1.01123857, "balance_loss_mlp": 1.00301766, "epoch": 0.3723922323092647, "flos": 62419899927840.0, "grad_norm": 0.8556142197103785, "language_loss": 0.55949301, "learning_rate": 2.8913536553020626e-06, "loss": 0.57964712, "num_input_tokens_seen": 66558840, "step": 3097, "time_per_iteration": 3.2100274562835693 }, { "auxiliary_loss_clip": 0.01127818, "auxiliary_loss_mlp": 0.01029836, "balance_loss_clip": 1.04790127, "balance_loss_mlp": 1.02150273, "epoch": 0.3725124751999038, "flos": 23039751751680.0, "grad_norm": 1.8875127283736965, "language_loss": 0.84508842, "learning_rate": 2.8906562570236137e-06, "loss": 0.86666489, "num_input_tokens_seen": 66576750, "step": 3098, "time_per_iteration": 2.616647958755493 }, { "auxiliary_loss_clip": 0.01118576, "auxiliary_loss_mlp": 0.01029686, "balance_loss_clip": 1.04456902, "balance_loss_mlp": 1.02174079, "epoch": 0.3726327180905429, "flos": 20920554713760.0, "grad_norm": 1.5576691729802226, "language_loss": 0.76490295, "learning_rate": 2.889958723633318e-06, "loss": 0.78638554, "num_input_tokens_seen": 66595690, "step": 3099, "time_per_iteration": 2.6426842212677 }, { "auxiliary_loss_clip": 0.01150969, "auxiliary_loss_mlp": 0.01034923, "balance_loss_clip": 1.05290389, "balance_loss_mlp": 1.02679896, "epoch": 0.372752960981182, "flos": 30592161144000.0, "grad_norm": 1.6916253919089383, "language_loss": 0.73920935, "learning_rate": 2.889261055236992e-06, "loss": 0.76106834, "num_input_tokens_seen": 66617905, "step": 3100, "time_per_iteration": 2.656996011734009 }, { "auxiliary_loss_clip": 0.01165022, "auxiliary_loss_mlp": 0.01030691, "balance_loss_clip": 1.05627728, "balance_loss_mlp": 1.02251863, "epoch": 0.3728732038718211, "flos": 25116752749920.0, "grad_norm": 1.907494282383138, "language_loss": 0.82729065, "learning_rate": 2.8885632519404704e-06, "loss": 0.84924781, "num_input_tokens_seen": 66638175, "step": 3101, "time_per_iteration": 2.5708420276641846 }, { "auxiliary_loss_clip": 0.01165744, "auxiliary_loss_mlp": 0.01028105, "balance_loss_clip": 1.055094, "balance_loss_mlp": 1.01946223, "epoch": 0.37299344676246016, "flos": 25302052321920.0, "grad_norm": 1.797519321263086, "language_loss": 0.7557385, "learning_rate": 2.8878653138496107e-06, "loss": 0.77767706, "num_input_tokens_seen": 66658670, "step": 3102, "time_per_iteration": 2.584686756134033 }, { "auxiliary_loss_clip": 0.01118867, "auxiliary_loss_mlp": 0.0103236, "balance_loss_clip": 1.04527557, "balance_loss_mlp": 1.02300239, "epoch": 0.37311368965309927, "flos": 23842534849440.0, "grad_norm": 2.2451775074627927, "language_loss": 0.76315242, "learning_rate": 2.8871672410702878e-06, "loss": 0.78466475, "num_input_tokens_seen": 66676030, "step": 3103, "time_per_iteration": 2.6493477821350098 }, { "auxiliary_loss_clip": 0.01161326, "auxiliary_loss_mlp": 0.01031604, "balance_loss_clip": 1.05244827, "balance_loss_mlp": 1.02319384, "epoch": 0.3732339325437384, "flos": 25811947800480.0, "grad_norm": 1.7489478458308487, "language_loss": 0.81865859, "learning_rate": 2.8864690337084008e-06, "loss": 0.84058797, "num_input_tokens_seen": 66695305, "step": 3104, "time_per_iteration": 2.6142826080322266 }, { "auxiliary_loss_clip": 0.01173489, "auxiliary_loss_mlp": 0.01029352, "balance_loss_clip": 1.05340385, "balance_loss_mlp": 1.0213176, "epoch": 0.37335417543437743, "flos": 26208436684320.0, "grad_norm": 1.7510267416074587, "language_loss": 0.78284103, "learning_rate": 2.885770691869866e-06, "loss": 0.80486941, "num_input_tokens_seen": 66716185, "step": 3105, "time_per_iteration": 2.5538299083709717 }, { "auxiliary_loss_clip": 0.01175696, "auxiliary_loss_mlp": 0.01031676, "balance_loss_clip": 1.05458593, "balance_loss_mlp": 1.02333987, "epoch": 0.37347441832501654, "flos": 24023883555840.0, "grad_norm": 2.0280593579833153, "language_loss": 0.74092758, "learning_rate": 2.8850722156606207e-06, "loss": 0.76300126, "num_input_tokens_seen": 66734575, "step": 3106, "time_per_iteration": 2.5311119556427 }, { "auxiliary_loss_clip": 0.01168997, "auxiliary_loss_mlp": 0.01025782, "balance_loss_clip": 1.05150104, "balance_loss_mlp": 1.01747322, "epoch": 0.3735946612156556, "flos": 19714925430240.0, "grad_norm": 1.7488757032860291, "language_loss": 0.6708771, "learning_rate": 2.8843736051866252e-06, "loss": 0.6928249, "num_input_tokens_seen": 66753500, "step": 3107, "time_per_iteration": 3.2945303916931152 }, { "auxiliary_loss_clip": 0.01134248, "auxiliary_loss_mlp": 0.00762546, "balance_loss_clip": 1.04989648, "balance_loss_mlp": 1.00032902, "epoch": 0.3737149041062947, "flos": 23039608083840.0, "grad_norm": 1.6491461007041675, "language_loss": 0.69469118, "learning_rate": 2.8836748605538557e-06, "loss": 0.71365917, "num_input_tokens_seen": 66775140, "step": 3108, "time_per_iteration": 3.421739101409912 }, { "auxiliary_loss_clip": 0.01170655, "auxiliary_loss_mlp": 0.0102746, "balance_loss_clip": 1.05442274, "balance_loss_mlp": 1.01886487, "epoch": 0.3738351469969338, "flos": 34678113277920.0, "grad_norm": 2.0586337379453346, "language_loss": 0.63460135, "learning_rate": 2.882975981868313e-06, "loss": 0.65658253, "num_input_tokens_seen": 66795525, "step": 3109, "time_per_iteration": 2.6497795581817627 }, { "auxiliary_loss_clip": 0.01181079, "auxiliary_loss_mlp": 0.01036665, "balance_loss_clip": 1.05796766, "balance_loss_mlp": 1.02806997, "epoch": 0.3739553898875729, "flos": 43507973097120.0, "grad_norm": 2.0802551034505448, "language_loss": 0.68792009, "learning_rate": 2.882276969236016e-06, "loss": 0.71009749, "num_input_tokens_seen": 66816885, "step": 3110, "time_per_iteration": 2.6872365474700928 }, { "auxiliary_loss_clip": 0.01161444, "auxiliary_loss_mlp": 0.01032422, "balance_loss_clip": 1.05163217, "balance_loss_mlp": 1.0234127, "epoch": 0.374075632778212, "flos": 12856489911840.0, "grad_norm": 1.9142076173565274, "language_loss": 0.7628113, "learning_rate": 2.881577822763005e-06, "loss": 0.78474998, "num_input_tokens_seen": 66834835, "step": 3111, "time_per_iteration": 3.2782797813415527 }, { "auxiliary_loss_clip": 0.01179557, "auxiliary_loss_mlp": 0.0102901, "balance_loss_clip": 1.05438614, "balance_loss_mlp": 1.02043343, "epoch": 0.3741958756688511, "flos": 26024035036320.0, "grad_norm": 1.95307151929703, "language_loss": 0.87324882, "learning_rate": 2.880878542555338e-06, "loss": 0.89533448, "num_input_tokens_seen": 66852600, "step": 3112, "time_per_iteration": 2.5668394565582275 }, { "auxiliary_loss_clip": 0.01195424, "auxiliary_loss_mlp": 0.0104317, "balance_loss_clip": 1.05694354, "balance_loss_mlp": 1.0349679, "epoch": 0.37431611855949015, "flos": 21433898220480.0, "grad_norm": 2.0008543714803966, "language_loss": 0.80268574, "learning_rate": 2.8801791287190976e-06, "loss": 0.82507163, "num_input_tokens_seen": 66870595, "step": 3113, "time_per_iteration": 2.4653408527374268 }, { "auxiliary_loss_clip": 0.01182798, "auxiliary_loss_mlp": 0.01028304, "balance_loss_clip": 1.05630183, "balance_loss_mlp": 1.01976287, "epoch": 0.37443636145012926, "flos": 24207099944160.0, "grad_norm": 2.698770543072079, "language_loss": 0.86146069, "learning_rate": 2.8794795813603817e-06, "loss": 0.88357174, "num_input_tokens_seen": 66886060, "step": 3114, "time_per_iteration": 3.305539846420288 }, { "auxiliary_loss_clip": 0.01183142, "auxiliary_loss_mlp": 0.01030857, "balance_loss_clip": 1.05477059, "balance_loss_mlp": 1.02254212, "epoch": 0.3745566043407684, "flos": 15378604969440.0, "grad_norm": 1.8738555510912616, "language_loss": 0.81660098, "learning_rate": 2.878779900585314e-06, "loss": 0.83874094, "num_input_tokens_seen": 66903900, "step": 3115, "time_per_iteration": 2.4590065479278564 }, { "auxiliary_loss_clip": 0.01170871, "auxiliary_loss_mlp": 0.01030403, "balance_loss_clip": 1.05326962, "balance_loss_mlp": 1.02218986, "epoch": 0.37467684723140743, "flos": 24608222115840.0, "grad_norm": 2.3498277280920323, "language_loss": 0.75313598, "learning_rate": 2.8780800865000336e-06, "loss": 0.77514875, "num_input_tokens_seen": 66925210, "step": 3116, "time_per_iteration": 2.5850586891174316 }, { "auxiliary_loss_clip": 0.0107442, "auxiliary_loss_mlp": 0.01002896, "balance_loss_clip": 1.01631451, "balance_loss_mlp": 1.00138187, "epoch": 0.37479709012204654, "flos": 64377502587360.0, "grad_norm": 0.9758946602543204, "language_loss": 0.59211904, "learning_rate": 2.877380139210702e-06, "loss": 0.61289221, "num_input_tokens_seen": 66983880, "step": 3117, "time_per_iteration": 3.0556046962738037 }, { "auxiliary_loss_clip": 0.01155871, "auxiliary_loss_mlp": 0.01026378, "balance_loss_clip": 1.05649126, "balance_loss_mlp": 1.01726472, "epoch": 0.37491733301268565, "flos": 23803966422720.0, "grad_norm": 1.944613465722941, "language_loss": 0.7645914, "learning_rate": 2.876680058823501e-06, "loss": 0.78641391, "num_input_tokens_seen": 67004280, "step": 3118, "time_per_iteration": 2.5893449783325195 }, { "auxiliary_loss_clip": 0.01157998, "auxiliary_loss_mlp": 0.01031473, "balance_loss_clip": 1.05354536, "balance_loss_mlp": 1.02188218, "epoch": 0.3750375759033247, "flos": 32160954760800.0, "grad_norm": 1.981252106279678, "language_loss": 0.65543002, "learning_rate": 2.8759798454446314e-06, "loss": 0.67732471, "num_input_tokens_seen": 67027445, "step": 3119, "time_per_iteration": 2.6276299953460693 }, { "auxiliary_loss_clip": 0.01184147, "auxiliary_loss_mlp": 0.01039943, "balance_loss_clip": 1.05686128, "balance_loss_mlp": 1.03181291, "epoch": 0.3751578187939638, "flos": 23367795686400.0, "grad_norm": 1.7708425413532836, "language_loss": 0.80939049, "learning_rate": 2.8752794991803173e-06, "loss": 0.83163142, "num_input_tokens_seen": 67045130, "step": 3120, "time_per_iteration": 2.5398285388946533 }, { "auxiliary_loss_clip": 0.01163204, "auxiliary_loss_mlp": 0.01027374, "balance_loss_clip": 1.05439615, "balance_loss_mlp": 1.01964951, "epoch": 0.37527806168460287, "flos": 14605734311040.0, "grad_norm": 2.909035618365599, "language_loss": 0.75009179, "learning_rate": 2.8745790201367976e-06, "loss": 0.77199757, "num_input_tokens_seen": 67060885, "step": 3121, "time_per_iteration": 2.495305061340332 }, { "auxiliary_loss_clip": 0.01196679, "auxiliary_loss_mlp": 0.01029481, "balance_loss_clip": 1.05858731, "balance_loss_mlp": 1.02102351, "epoch": 0.375398304575242, "flos": 26390826982560.0, "grad_norm": 1.9823050249804113, "language_loss": 0.84275347, "learning_rate": 2.8738784084203373e-06, "loss": 0.86501509, "num_input_tokens_seen": 67080960, "step": 3122, "time_per_iteration": 2.5294430255889893 }, { "auxiliary_loss_clip": 0.01155144, "auxiliary_loss_mlp": 0.0103277, "balance_loss_clip": 1.04831183, "balance_loss_mlp": 1.02523565, "epoch": 0.3755185474658811, "flos": 22236609484320.0, "grad_norm": 1.6794895381805044, "language_loss": 0.78626311, "learning_rate": 2.873177664137216e-06, "loss": 0.80814219, "num_input_tokens_seen": 67101890, "step": 3123, "time_per_iteration": 2.566011667251587 }, { "auxiliary_loss_clip": 0.01153045, "auxiliary_loss_mlp": 0.01027819, "balance_loss_clip": 1.05883574, "balance_loss_mlp": 1.0193193, "epoch": 0.37563879035652015, "flos": 30812940284160.0, "grad_norm": 1.6518380403882462, "language_loss": 0.69358236, "learning_rate": 2.8724767873937384e-06, "loss": 0.71539104, "num_input_tokens_seen": 67126010, "step": 3124, "time_per_iteration": 2.6877946853637695 }, { "auxiliary_loss_clip": 0.01165463, "auxiliary_loss_mlp": 0.01027426, "balance_loss_clip": 1.05406952, "balance_loss_mlp": 1.0196116, "epoch": 0.37575903324715926, "flos": 20773536232800.0, "grad_norm": 2.1475904984443708, "language_loss": 0.87632394, "learning_rate": 2.871775778296225e-06, "loss": 0.89825284, "num_input_tokens_seen": 67143100, "step": 3125, "time_per_iteration": 2.513854742050171 }, { "auxiliary_loss_clip": 0.0118223, "auxiliary_loss_mlp": 0.010359, "balance_loss_clip": 1.05809951, "balance_loss_mlp": 1.02707243, "epoch": 0.37587927613779837, "flos": 18697684577280.0, "grad_norm": 2.600368902013676, "language_loss": 0.78464234, "learning_rate": 2.8710746369510196e-06, "loss": 0.80682361, "num_input_tokens_seen": 67161085, "step": 3126, "time_per_iteration": 2.5112273693084717 }, { "auxiliary_loss_clip": 0.0116342, "auxiliary_loss_mlp": 0.01026419, "balance_loss_clip": 1.05499828, "balance_loss_mlp": 1.01833653, "epoch": 0.3759995190284374, "flos": 13624799116320.0, "grad_norm": 2.2699005010232622, "language_loss": 0.83120716, "learning_rate": 2.8703733634644846e-06, "loss": 0.85310555, "num_input_tokens_seen": 67175840, "step": 3127, "time_per_iteration": 2.5005836486816406 }, { "auxiliary_loss_clip": 0.01193003, "auxiliary_loss_mlp": 0.0103509, "balance_loss_clip": 1.05786347, "balance_loss_mlp": 1.02672791, "epoch": 0.37611976191907653, "flos": 20484850897920.0, "grad_norm": 1.87085698775789, "language_loss": 0.79184812, "learning_rate": 2.869671957943002e-06, "loss": 0.81412905, "num_input_tokens_seen": 67194995, "step": 3128, "time_per_iteration": 2.4877943992614746 }, { "auxiliary_loss_clip": 0.01161316, "auxiliary_loss_mlp": 0.0103083, "balance_loss_clip": 1.05771923, "balance_loss_mlp": 1.02321219, "epoch": 0.37624000480971564, "flos": 21141800774400.0, "grad_norm": 1.844307941941656, "language_loss": 0.7424261, "learning_rate": 2.8689704204929747e-06, "loss": 0.76434755, "num_input_tokens_seen": 67214175, "step": 3129, "time_per_iteration": 2.54190993309021 }, { "auxiliary_loss_clip": 0.01195865, "auxiliary_loss_mlp": 0.01029613, "balance_loss_clip": 1.05859971, "balance_loss_mlp": 1.02153683, "epoch": 0.3763602477003547, "flos": 22564473834240.0, "grad_norm": 1.91046269574706, "language_loss": 0.81101322, "learning_rate": 2.8682687512208253e-06, "loss": 0.83326793, "num_input_tokens_seen": 67233185, "step": 3130, "time_per_iteration": 2.4896833896636963 }, { "auxiliary_loss_clip": 0.01189269, "auxiliary_loss_mlp": 0.01031527, "balance_loss_clip": 1.0579052, "balance_loss_mlp": 1.02318192, "epoch": 0.3764804905909938, "flos": 27526861974240.0, "grad_norm": 2.0956681095441017, "language_loss": 0.80454069, "learning_rate": 2.8675669502329972e-06, "loss": 0.82674861, "num_input_tokens_seen": 67254715, "step": 3131, "time_per_iteration": 2.5539791584014893 }, { "auxiliary_loss_clip": 0.01179259, "auxiliary_loss_mlp": 0.00763245, "balance_loss_clip": 1.05538893, "balance_loss_mlp": 1.00035954, "epoch": 0.3766007334816329, "flos": 22528096342080.0, "grad_norm": 2.1777131756193766, "language_loss": 0.85487211, "learning_rate": 2.866865017635952e-06, "loss": 0.87429714, "num_input_tokens_seen": 67272535, "step": 3132, "time_per_iteration": 2.52309250831604 }, { "auxiliary_loss_clip": 0.01149357, "auxiliary_loss_mlp": 0.01030798, "balance_loss_clip": 1.05658472, "balance_loss_mlp": 1.0220902, "epoch": 0.376720976372272, "flos": 25957170433440.0, "grad_norm": 1.604360243082605, "language_loss": 0.79306078, "learning_rate": 2.866162953536174e-06, "loss": 0.81486237, "num_input_tokens_seen": 67293505, "step": 3133, "time_per_iteration": 3.4006567001342773 }, { "auxiliary_loss_clip": 0.01164728, "auxiliary_loss_mlp": 0.00763429, "balance_loss_clip": 1.05339813, "balance_loss_mlp": 1.00029349, "epoch": 0.3768412192629111, "flos": 18041165704320.0, "grad_norm": 1.4898762072403684, "language_loss": 0.75043225, "learning_rate": 2.8654607580401634e-06, "loss": 0.76971376, "num_input_tokens_seen": 67313240, "step": 3134, "time_per_iteration": 3.286763906478882 }, { "auxiliary_loss_clip": 0.01071734, "auxiliary_loss_mlp": 0.01003666, "balance_loss_clip": 1.01506066, "balance_loss_mlp": 1.00227761, "epoch": 0.3769614621535502, "flos": 62989483005600.0, "grad_norm": 0.8847326296002529, "language_loss": 0.65130377, "learning_rate": 2.8647584312544446e-06, "loss": 0.67205781, "num_input_tokens_seen": 67378445, "step": 3135, "time_per_iteration": 3.115004062652588 }, { "auxiliary_loss_clip": 0.01145396, "auxiliary_loss_mlp": 0.00762649, "balance_loss_clip": 1.04973996, "balance_loss_mlp": 1.00042129, "epoch": 0.37708170504418925, "flos": 23661688980480.0, "grad_norm": 1.4084998990296052, "language_loss": 0.85413349, "learning_rate": 2.864055973285559e-06, "loss": 0.87321389, "num_input_tokens_seen": 67400445, "step": 3136, "time_per_iteration": 2.6064727306365967 }, { "auxiliary_loss_clip": 0.01152283, "auxiliary_loss_mlp": 0.01032855, "balance_loss_clip": 1.05193305, "balance_loss_mlp": 1.02496338, "epoch": 0.37720194793482836, "flos": 24423173962560.0, "grad_norm": 1.7787081645346605, "language_loss": 0.86194247, "learning_rate": 2.8633533842400698e-06, "loss": 0.88379383, "num_input_tokens_seen": 67420645, "step": 3137, "time_per_iteration": 3.342728614807129 }, { "auxiliary_loss_clip": 0.01182954, "auxiliary_loss_mlp": 0.00764145, "balance_loss_clip": 1.05839598, "balance_loss_mlp": 1.00046945, "epoch": 0.3773221908254674, "flos": 20996506307520.0, "grad_norm": 1.8598871179065601, "language_loss": 0.76831508, "learning_rate": 2.862650664224558e-06, "loss": 0.78778601, "num_input_tokens_seen": 67439495, "step": 3138, "time_per_iteration": 2.531919240951538 }, { "auxiliary_loss_clip": 0.0117837, "auxiliary_loss_mlp": 0.01034069, "balance_loss_clip": 1.05870581, "balance_loss_mlp": 1.02637362, "epoch": 0.37744243371610653, "flos": 37631729867040.0, "grad_norm": 1.4331535478342767, "language_loss": 0.69732547, "learning_rate": 2.861947813345627e-06, "loss": 0.71944988, "num_input_tokens_seen": 67462195, "step": 3139, "time_per_iteration": 2.6365020275115967 }, { "auxiliary_loss_clip": 0.01200574, "auxiliary_loss_mlp": 0.00763084, "balance_loss_clip": 1.06068277, "balance_loss_mlp": 1.00035679, "epoch": 0.37756267660674564, "flos": 26140530489600.0, "grad_norm": 1.6741710703887593, "language_loss": 0.72629493, "learning_rate": 2.8612448317098974e-06, "loss": 0.74593151, "num_input_tokens_seen": 67482530, "step": 3140, "time_per_iteration": 3.267821788787842 }, { "auxiliary_loss_clip": 0.01155301, "auxiliary_loss_mlp": 0.00763918, "balance_loss_clip": 1.05466151, "balance_loss_mlp": 1.00052512, "epoch": 0.3776829194973847, "flos": 19427892275520.0, "grad_norm": 2.0287038127040944, "language_loss": 0.8291651, "learning_rate": 2.8605417194240114e-06, "loss": 0.84835726, "num_input_tokens_seen": 67500890, "step": 3141, "time_per_iteration": 2.572645425796509 }, { "auxiliary_loss_clip": 0.01176576, "auxiliary_loss_mlp": 0.01029177, "balance_loss_clip": 1.05672884, "balance_loss_mlp": 1.02122545, "epoch": 0.3778031623880238, "flos": 17382312228960.0, "grad_norm": 1.710222915786342, "language_loss": 0.78809273, "learning_rate": 2.8598384765946315e-06, "loss": 0.81015027, "num_input_tokens_seen": 67519545, "step": 3142, "time_per_iteration": 2.5251715183258057 }, { "auxiliary_loss_clip": 0.0119291, "auxiliary_loss_mlp": 0.0102363, "balance_loss_clip": 1.05691791, "balance_loss_mlp": 1.01560092, "epoch": 0.3779234052786629, "flos": 27125847553440.0, "grad_norm": 1.7142014896255922, "language_loss": 0.7201218, "learning_rate": 2.8591351033284377e-06, "loss": 0.74228722, "num_input_tokens_seen": 67539275, "step": 3143, "time_per_iteration": 2.5240986347198486 }, { "auxiliary_loss_clip": 0.01182494, "auxiliary_loss_mlp": 0.01026507, "balance_loss_clip": 1.05653369, "balance_loss_mlp": 1.01826322, "epoch": 0.37804364816930197, "flos": 19682642471520.0, "grad_norm": 1.9991226459904419, "language_loss": 0.8344999, "learning_rate": 2.8584315997321325e-06, "loss": 0.85658985, "num_input_tokens_seen": 67558280, "step": 3144, "time_per_iteration": 2.5074236392974854 }, { "auxiliary_loss_clip": 0.0119493, "auxiliary_loss_mlp": 0.01030597, "balance_loss_clip": 1.05823386, "balance_loss_mlp": 1.02234745, "epoch": 0.3781638910599411, "flos": 22702908161760.0, "grad_norm": 2.4879268552384057, "language_loss": 0.779194, "learning_rate": 2.8577279659124356e-06, "loss": 0.80144924, "num_input_tokens_seen": 67575955, "step": 3145, "time_per_iteration": 2.47699236869812 }, { "auxiliary_loss_clip": 0.0117546, "auxiliary_loss_mlp": 0.01026202, "balance_loss_clip": 1.0546515, "balance_loss_mlp": 1.01859641, "epoch": 0.3782841339505802, "flos": 14647607098080.0, "grad_norm": 1.826306337020179, "language_loss": 0.8329829, "learning_rate": 2.857024201976089e-06, "loss": 0.85499954, "num_input_tokens_seen": 67593515, "step": 3146, "time_per_iteration": 2.508514881134033 }, { "auxiliary_loss_clip": 0.01164616, "auxiliary_loss_mlp": 0.01028581, "balance_loss_clip": 1.05713594, "balance_loss_mlp": 1.01998019, "epoch": 0.37840437684121925, "flos": 32818227889920.0, "grad_norm": 1.9480168110889962, "language_loss": 0.73326135, "learning_rate": 2.8563203080298516e-06, "loss": 0.75519335, "num_input_tokens_seen": 67614290, "step": 3147, "time_per_iteration": 2.630077362060547 }, { "auxiliary_loss_clip": 0.01166001, "auxiliary_loss_mlp": 0.00763076, "balance_loss_clip": 1.05758429, "balance_loss_mlp": 1.00035596, "epoch": 0.37852461973185836, "flos": 18369209639040.0, "grad_norm": 2.0627729353683466, "language_loss": 0.89238513, "learning_rate": 2.855616284180505e-06, "loss": 0.91167593, "num_input_tokens_seen": 67631340, "step": 3148, "time_per_iteration": 2.540781259536743 }, { "auxiliary_loss_clip": 0.01073117, "auxiliary_loss_mlp": 0.01001289, "balance_loss_clip": 1.01378798, "balance_loss_mlp": 0.99978715, "epoch": 0.37864486262249747, "flos": 59500691786400.0, "grad_norm": 0.8893946832103468, "language_loss": 0.66169846, "learning_rate": 2.8549121305348477e-06, "loss": 0.68244255, "num_input_tokens_seen": 67691125, "step": 3149, "time_per_iteration": 3.0937469005584717 }, { "auxiliary_loss_clip": 0.01179699, "auxiliary_loss_mlp": 0.01028024, "balance_loss_clip": 1.05522537, "balance_loss_mlp": 1.02038312, "epoch": 0.3787651055131365, "flos": 23363018730720.0, "grad_norm": 2.3082980888491, "language_loss": 0.83387905, "learning_rate": 2.8542078471997006e-06, "loss": 0.85595632, "num_input_tokens_seen": 67708740, "step": 3150, "time_per_iteration": 2.534423589706421 }, { "auxiliary_loss_clip": 0.01181145, "auxiliary_loss_mlp": 0.01030494, "balance_loss_clip": 1.05751002, "balance_loss_mlp": 1.02282262, "epoch": 0.37888534840377563, "flos": 24601397893440.0, "grad_norm": 1.7587156075635146, "language_loss": 0.75683475, "learning_rate": 2.8535034342819013e-06, "loss": 0.77895111, "num_input_tokens_seen": 67726150, "step": 3151, "time_per_iteration": 2.575652599334717 }, { "auxiliary_loss_clip": 0.01192408, "auxiliary_loss_mlp": 0.0102234, "balance_loss_clip": 1.05787516, "balance_loss_mlp": 1.01462698, "epoch": 0.37900559129441475, "flos": 23986895475360.0, "grad_norm": 1.6377255341682047, "language_loss": 0.72408271, "learning_rate": 2.85279889188831e-06, "loss": 0.74623019, "num_input_tokens_seen": 67746525, "step": 3152, "time_per_iteration": 2.4943995475769043 }, { "auxiliary_loss_clip": 0.01150678, "auxiliary_loss_mlp": 0.01028994, "balance_loss_clip": 1.05205846, "balance_loss_mlp": 1.02087021, "epoch": 0.3791258341850538, "flos": 24644671441920.0, "grad_norm": 1.8872439589592447, "language_loss": 0.81186563, "learning_rate": 2.852094220125805e-06, "loss": 0.83366233, "num_input_tokens_seen": 67766035, "step": 3153, "time_per_iteration": 2.6366751194000244 }, { "auxiliary_loss_clip": 0.01184321, "auxiliary_loss_mlp": 0.01032616, "balance_loss_clip": 1.05987918, "balance_loss_mlp": 1.02387214, "epoch": 0.3792460770756929, "flos": 17420844738720.0, "grad_norm": 1.9710102101221056, "language_loss": 0.71008408, "learning_rate": 2.8513894191012846e-06, "loss": 0.73225343, "num_input_tokens_seen": 67785015, "step": 3154, "time_per_iteration": 2.49923038482666 }, { "auxiliary_loss_clip": 0.01197127, "auxiliary_loss_mlp": 0.01030228, "balance_loss_clip": 1.05963421, "balance_loss_mlp": 1.02165639, "epoch": 0.37936631996633197, "flos": 24206561189760.0, "grad_norm": 2.293299418338389, "language_loss": 0.78329432, "learning_rate": 2.8506844889216664e-06, "loss": 0.80556786, "num_input_tokens_seen": 67804400, "step": 3155, "time_per_iteration": 2.569411516189575 }, { "auxiliary_loss_clip": 0.01066794, "auxiliary_loss_mlp": 0.01006679, "balance_loss_clip": 1.01398826, "balance_loss_mlp": 1.00512969, "epoch": 0.3794865628569711, "flos": 70297127116800.0, "grad_norm": 0.9119125104457191, "language_loss": 0.62852228, "learning_rate": 2.849979429693887e-06, "loss": 0.649257, "num_input_tokens_seen": 67865385, "step": 3156, "time_per_iteration": 3.1818177700042725 }, { "auxiliary_loss_clip": 0.01197578, "auxiliary_loss_mlp": 0.01028666, "balance_loss_clip": 1.06067753, "balance_loss_mlp": 1.02015471, "epoch": 0.3796068057476102, "flos": 15779367971520.0, "grad_norm": 1.9666739952153873, "language_loss": 0.74028403, "learning_rate": 2.8492742415249042e-06, "loss": 0.76254642, "num_input_tokens_seen": 67883030, "step": 3157, "time_per_iteration": 2.542501926422119 }, { "auxiliary_loss_clip": 0.01197177, "auxiliary_loss_mlp": 0.01029027, "balance_loss_clip": 1.06129265, "balance_loss_mlp": 1.02067971, "epoch": 0.37972704863824924, "flos": 25191698668800.0, "grad_norm": 1.6583849549450926, "language_loss": 0.7627939, "learning_rate": 2.848568924521694e-06, "loss": 0.78505588, "num_input_tokens_seen": 67903810, "step": 3158, "time_per_iteration": 2.6338534355163574 }, { "auxiliary_loss_clip": 0.01170475, "auxiliary_loss_mlp": 0.01031278, "balance_loss_clip": 1.05219007, "balance_loss_mlp": 1.02239072, "epoch": 0.37984729152888835, "flos": 26210376200160.0, "grad_norm": 1.9865504028474272, "language_loss": 0.73499495, "learning_rate": 2.8478634787912526e-06, "loss": 0.75701243, "num_input_tokens_seen": 67921865, "step": 3159, "time_per_iteration": 3.3591647148132324 }, { "auxiliary_loss_clip": 0.01182551, "auxiliary_loss_mlp": 0.01037084, "balance_loss_clip": 1.05847239, "balance_loss_mlp": 1.02916217, "epoch": 0.37996753441952746, "flos": 25629306083520.0, "grad_norm": 2.134948210641407, "language_loss": 0.76484692, "learning_rate": 2.847157904440596e-06, "loss": 0.78704321, "num_input_tokens_seen": 67941595, "step": 3160, "time_per_iteration": 3.3850362300872803 }, { "auxiliary_loss_clip": 0.01180183, "auxiliary_loss_mlp": 0.01028042, "balance_loss_clip": 1.05901682, "balance_loss_mlp": 1.02011466, "epoch": 0.3800877773101665, "flos": 20118418121280.0, "grad_norm": 1.5744569849042673, "language_loss": 0.74163651, "learning_rate": 2.846452201576759e-06, "loss": 0.76371872, "num_input_tokens_seen": 67960970, "step": 3161, "time_per_iteration": 2.599086046218872 }, { "auxiliary_loss_clip": 0.01063506, "auxiliary_loss_mlp": 0.01000716, "balance_loss_clip": 1.01416636, "balance_loss_mlp": 0.99925578, "epoch": 0.38020802020080563, "flos": 63053617919520.0, "grad_norm": 0.8565137258448889, "language_loss": 0.6277473, "learning_rate": 2.845746370306795e-06, "loss": 0.64838952, "num_input_tokens_seen": 68026160, "step": 3162, "time_per_iteration": 3.257530689239502 }, { "auxiliary_loss_clip": 0.01182686, "auxiliary_loss_mlp": 0.01030043, "balance_loss_clip": 1.05788743, "balance_loss_mlp": 1.02219343, "epoch": 0.38032826309144474, "flos": 21288424168800.0, "grad_norm": 1.795965088810162, "language_loss": 0.78423506, "learning_rate": 2.84504041073778e-06, "loss": 0.80636239, "num_input_tokens_seen": 68044575, "step": 3163, "time_per_iteration": 3.258726119995117 }, { "auxiliary_loss_clip": 0.01159416, "auxiliary_loss_mlp": 0.0103416, "balance_loss_clip": 1.05621314, "balance_loss_mlp": 1.02572572, "epoch": 0.3804485059820838, "flos": 18954122870400.0, "grad_norm": 1.7913815220980198, "language_loss": 0.79284728, "learning_rate": 2.844334322976806e-06, "loss": 0.81478298, "num_input_tokens_seen": 68064790, "step": 3164, "time_per_iteration": 2.5692453384399414 }, { "auxiliary_loss_clip": 0.0114331, "auxiliary_loss_mlp": 0.01038569, "balance_loss_clip": 1.05413723, "balance_loss_mlp": 1.03022408, "epoch": 0.3805687488727229, "flos": 21833763298560.0, "grad_norm": 1.855864927392796, "language_loss": 0.83028692, "learning_rate": 2.8436281071309866e-06, "loss": 0.85210568, "num_input_tokens_seen": 68083330, "step": 3165, "time_per_iteration": 2.6355507373809814 }, { "auxiliary_loss_clip": 0.0104473, "auxiliary_loss_mlp": 0.00999759, "balance_loss_clip": 1.01391935, "balance_loss_mlp": 0.99816173, "epoch": 0.380688991763362, "flos": 58546221002880.0, "grad_norm": 0.8408671063521242, "language_loss": 0.52981597, "learning_rate": 2.842921763307455e-06, "loss": 0.5502609, "num_input_tokens_seen": 68146140, "step": 3166, "time_per_iteration": 3.885821580886841 }, { "auxiliary_loss_clip": 0.01159382, "auxiliary_loss_mlp": 0.01035233, "balance_loss_clip": 1.05288672, "balance_loss_mlp": 1.02743089, "epoch": 0.38080923465400107, "flos": 23799512719680.0, "grad_norm": 2.4206658654344326, "language_loss": 0.82591993, "learning_rate": 2.842215291613361e-06, "loss": 0.84786606, "num_input_tokens_seen": 68164520, "step": 3167, "time_per_iteration": 2.6002564430236816 }, { "auxiliary_loss_clip": 0.01012496, "auxiliary_loss_mlp": 0.01008807, "balance_loss_clip": 1.0174799, "balance_loss_mlp": 1.00735843, "epoch": 0.3809294775446402, "flos": 54969877011840.0, "grad_norm": 0.7891312842872814, "language_loss": 0.59281504, "learning_rate": 2.8415086921558774e-06, "loss": 0.61302811, "num_input_tokens_seen": 68227945, "step": 3168, "time_per_iteration": 3.277601718902588 }, { "auxiliary_loss_clip": 0.01150541, "auxiliary_loss_mlp": 0.01026361, "balance_loss_clip": 1.04808378, "balance_loss_mlp": 1.01833773, "epoch": 0.38104972043527924, "flos": 24643701684000.0, "grad_norm": 1.479526617431762, "language_loss": 0.78608024, "learning_rate": 2.840801965042194e-06, "loss": 0.80784929, "num_input_tokens_seen": 68247405, "step": 3169, "time_per_iteration": 2.7000532150268555 }, { "auxiliary_loss_clip": 0.0115443, "auxiliary_loss_mlp": 0.01029419, "balance_loss_clip": 1.04998899, "balance_loss_mlp": 1.02093768, "epoch": 0.38116996332591835, "flos": 22856786782080.0, "grad_norm": 1.7013082121181988, "language_loss": 0.83955878, "learning_rate": 2.840095110379521e-06, "loss": 0.86139727, "num_input_tokens_seen": 68266925, "step": 3170, "time_per_iteration": 2.5853755474090576 }, { "auxiliary_loss_clip": 0.01030524, "auxiliary_loss_mlp": 0.01004365, "balance_loss_clip": 1.0124867, "balance_loss_mlp": 1.00289237, "epoch": 0.38129020621655746, "flos": 60836171244000.0, "grad_norm": 0.7340607742654607, "language_loss": 0.53972435, "learning_rate": 2.8393881282750884e-06, "loss": 0.56007314, "num_input_tokens_seen": 68329755, "step": 3171, "time_per_iteration": 3.1444249153137207 }, { "auxiliary_loss_clip": 0.01165719, "auxiliary_loss_mlp": 0.01032589, "balance_loss_clip": 1.05680585, "balance_loss_mlp": 1.02394652, "epoch": 0.3814104491071965, "flos": 21648104556960.0, "grad_norm": 1.768277385048002, "language_loss": 0.78464532, "learning_rate": 2.838681018836144e-06, "loss": 0.80662835, "num_input_tokens_seen": 68347075, "step": 3172, "time_per_iteration": 2.58158802986145 }, { "auxiliary_loss_clip": 0.01155189, "auxiliary_loss_mlp": 0.00763459, "balance_loss_clip": 1.05270457, "balance_loss_mlp": 1.00045431, "epoch": 0.3815306919978356, "flos": 19099094084640.0, "grad_norm": 1.9587769336928849, "language_loss": 0.78289568, "learning_rate": 2.837973782169955e-06, "loss": 0.80208218, "num_input_tokens_seen": 68365450, "step": 3173, "time_per_iteration": 2.6044459342956543 }, { "auxiliary_loss_clip": 0.01081303, "auxiliary_loss_mlp": 0.00999566, "balance_loss_clip": 1.01455021, "balance_loss_mlp": 0.99818939, "epoch": 0.38165093488847474, "flos": 67067928494880.0, "grad_norm": 0.8105818423732811, "language_loss": 0.59220219, "learning_rate": 2.8372664183838096e-06, "loss": 0.61301088, "num_input_tokens_seen": 68428470, "step": 3174, "time_per_iteration": 3.146024227142334 }, { "auxiliary_loss_clip": 0.01193735, "auxiliary_loss_mlp": 0.01027495, "balance_loss_clip": 1.05780828, "balance_loss_mlp": 1.01941824, "epoch": 0.3817711777791138, "flos": 22341108672960.0, "grad_norm": 2.1043398693860365, "language_loss": 0.67912257, "learning_rate": 2.836558927585015e-06, "loss": 0.70133483, "num_input_tokens_seen": 68445440, "step": 3175, "time_per_iteration": 2.5129196643829346 }, { "auxiliary_loss_clip": 0.01183607, "auxiliary_loss_mlp": 0.01029999, "balance_loss_clip": 1.05711102, "balance_loss_mlp": 1.02169597, "epoch": 0.3818914206697529, "flos": 22820624791680.0, "grad_norm": 1.7140797710555598, "language_loss": 0.8253727, "learning_rate": 2.8358513098808957e-06, "loss": 0.84750873, "num_input_tokens_seen": 68465755, "step": 3176, "time_per_iteration": 2.5658886432647705 }, { "auxiliary_loss_clip": 0.01126219, "auxiliary_loss_mlp": 0.01028751, "balance_loss_clip": 1.05211949, "balance_loss_mlp": 1.02085364, "epoch": 0.382011663560392, "flos": 24386078131200.0, "grad_norm": 1.734563438474244, "language_loss": 0.7668891, "learning_rate": 2.835143565378798e-06, "loss": 0.7884388, "num_input_tokens_seen": 68486220, "step": 3177, "time_per_iteration": 2.661489486694336 }, { "auxiliary_loss_clip": 0.01118215, "auxiliary_loss_mlp": 0.01030188, "balance_loss_clip": 1.04968882, "balance_loss_mlp": 1.02146804, "epoch": 0.38213190645103107, "flos": 21981571952640.0, "grad_norm": 1.6952699282527484, "language_loss": 0.78111506, "learning_rate": 2.8344356941860847e-06, "loss": 0.80259913, "num_input_tokens_seen": 68505850, "step": 3178, "time_per_iteration": 2.680255889892578 }, { "auxiliary_loss_clip": 0.01148004, "auxiliary_loss_mlp": 0.01037428, "balance_loss_clip": 1.05439937, "balance_loss_mlp": 1.02901793, "epoch": 0.3822521493416702, "flos": 35516950615200.0, "grad_norm": 2.038030334996356, "language_loss": 0.66517556, "learning_rate": 2.8337276964101403e-06, "loss": 0.68702984, "num_input_tokens_seen": 68526290, "step": 3179, "time_per_iteration": 2.7346816062927246 }, { "auxiliary_loss_clip": 0.01182958, "auxiliary_loss_mlp": 0.01031873, "balance_loss_clip": 1.05642176, "balance_loss_mlp": 1.02346289, "epoch": 0.3823723922323093, "flos": 21069907797120.0, "grad_norm": 1.8008434707553005, "language_loss": 0.76247686, "learning_rate": 2.833019572158367e-06, "loss": 0.78462517, "num_input_tokens_seen": 68544725, "step": 3180, "time_per_iteration": 2.5456464290618896 }, { "auxiliary_loss_clip": 0.01167379, "auxiliary_loss_mlp": 0.01026173, "balance_loss_clip": 1.05808592, "balance_loss_mlp": 1.0182873, "epoch": 0.38249263512294834, "flos": 19789153009920.0, "grad_norm": 1.7748088420487929, "language_loss": 0.7985937, "learning_rate": 2.8323113215381872e-06, "loss": 0.82052922, "num_input_tokens_seen": 68563070, "step": 3181, "time_per_iteration": 2.6289732456207275 }, { "auxiliary_loss_clip": 0.01152272, "auxiliary_loss_mlp": 0.01029384, "balance_loss_clip": 1.05500078, "balance_loss_mlp": 1.02004957, "epoch": 0.38261287801358745, "flos": 21433934137440.0, "grad_norm": 1.834718550791531, "language_loss": 0.75997621, "learning_rate": 2.831602944657042e-06, "loss": 0.78179276, "num_input_tokens_seen": 68581150, "step": 3182, "time_per_iteration": 2.5932703018188477 }, { "auxiliary_loss_clip": 0.01172175, "auxiliary_loss_mlp": 0.0103562, "balance_loss_clip": 1.05587101, "balance_loss_mlp": 1.02730525, "epoch": 0.38273312090422656, "flos": 21981571952640.0, "grad_norm": 2.40273617082593, "language_loss": 0.74336213, "learning_rate": 2.830894441622391e-06, "loss": 0.76544011, "num_input_tokens_seen": 68597800, "step": 3183, "time_per_iteration": 2.5836870670318604 }, { "auxiliary_loss_clip": 0.01149847, "auxiliary_loss_mlp": 0.00763665, "balance_loss_clip": 1.04996765, "balance_loss_mlp": 1.00041366, "epoch": 0.3828533637948656, "flos": 24790899749760.0, "grad_norm": 1.8198792623623725, "language_loss": 0.79935658, "learning_rate": 2.8301858125417134e-06, "loss": 0.81849164, "num_input_tokens_seen": 68617640, "step": 3184, "time_per_iteration": 2.628124952316284 }, { "auxiliary_loss_clip": 0.01167112, "auxiliary_loss_mlp": 0.01027905, "balance_loss_clip": 1.05687284, "balance_loss_mlp": 1.02022529, "epoch": 0.38297360668550473, "flos": 22455449108640.0, "grad_norm": 1.6827454087258527, "language_loss": 0.73898232, "learning_rate": 2.8294770575225082e-06, "loss": 0.76093251, "num_input_tokens_seen": 68637770, "step": 3185, "time_per_iteration": 3.3331472873687744 }, { "auxiliary_loss_clip": 0.01182792, "auxiliary_loss_mlp": 0.01029599, "balance_loss_clip": 1.0593636, "balance_loss_mlp": 1.02143312, "epoch": 0.3830938495761438, "flos": 24896907450720.0, "grad_norm": 1.7461846060536674, "language_loss": 0.83978724, "learning_rate": 2.828768176672293e-06, "loss": 0.86191118, "num_input_tokens_seen": 68656885, "step": 3186, "time_per_iteration": 3.5639071464538574 }, { "auxiliary_loss_clip": 0.01149452, "auxiliary_loss_mlp": 0.01033244, "balance_loss_clip": 1.05179906, "balance_loss_mlp": 1.02461934, "epoch": 0.3832140924667829, "flos": 33036241424160.0, "grad_norm": 1.7361915245465265, "language_loss": 0.71606839, "learning_rate": 2.8280591700986044e-06, "loss": 0.73789537, "num_input_tokens_seen": 68678750, "step": 3187, "time_per_iteration": 2.709278106689453 }, { "auxiliary_loss_clip": 0.01170002, "auxiliary_loss_mlp": 0.01025972, "balance_loss_clip": 1.05445075, "balance_loss_mlp": 1.01759148, "epoch": 0.383334335357422, "flos": 31903726294560.0, "grad_norm": 1.7802005585785823, "language_loss": 0.74671173, "learning_rate": 2.827350037908999e-06, "loss": 0.76867139, "num_input_tokens_seen": 68698190, "step": 3188, "time_per_iteration": 2.6454124450683594 }, { "auxiliary_loss_clip": 0.01158828, "auxiliary_loss_mlp": 0.01039964, "balance_loss_clip": 1.05465794, "balance_loss_mlp": 1.03049886, "epoch": 0.38345457824806106, "flos": 19791918615840.0, "grad_norm": 2.0813962359284774, "language_loss": 0.78858435, "learning_rate": 2.8266407802110496e-06, "loss": 0.81057227, "num_input_tokens_seen": 68716445, "step": 3189, "time_per_iteration": 3.3356666564941406 }, { "auxiliary_loss_clip": 0.01114951, "auxiliary_loss_mlp": 0.01031539, "balance_loss_clip": 1.04680526, "balance_loss_mlp": 1.02287865, "epoch": 0.3835748211387002, "flos": 22419394869120.0, "grad_norm": 1.782844560286737, "language_loss": 0.75931442, "learning_rate": 2.8259313971123515e-06, "loss": 0.7807793, "num_input_tokens_seen": 68737565, "step": 3190, "time_per_iteration": 2.7257158756256104 }, { "auxiliary_loss_clip": 0.01178124, "auxiliary_loss_mlp": 0.01031338, "balance_loss_clip": 1.05870867, "balance_loss_mlp": 1.02303541, "epoch": 0.3836950640293393, "flos": 25118440847040.0, "grad_norm": 1.7495063651497134, "language_loss": 0.78245533, "learning_rate": 2.8252218887205166e-06, "loss": 0.80454993, "num_input_tokens_seen": 68758255, "step": 3191, "time_per_iteration": 2.5648562908172607 }, { "auxiliary_loss_clip": 0.01126321, "auxiliary_loss_mlp": 0.01032231, "balance_loss_clip": 1.05177665, "balance_loss_mlp": 1.02417231, "epoch": 0.38381530691997834, "flos": 21799217571360.0, "grad_norm": 1.88792814865262, "language_loss": 0.80710506, "learning_rate": 2.824512255143178e-06, "loss": 0.82869059, "num_input_tokens_seen": 68777490, "step": 3192, "time_per_iteration": 3.4126110076904297 }, { "auxiliary_loss_clip": 0.01153599, "auxiliary_loss_mlp": 0.01026096, "balance_loss_clip": 1.05337799, "balance_loss_mlp": 1.01754308, "epoch": 0.38393554981061745, "flos": 21252693181920.0, "grad_norm": 1.7801419412257447, "language_loss": 0.79061782, "learning_rate": 2.8238024964879855e-06, "loss": 0.81241471, "num_input_tokens_seen": 68798385, "step": 3193, "time_per_iteration": 2.6299054622650146 }, { "auxiliary_loss_clip": 0.01197444, "auxiliary_loss_mlp": 0.01034144, "balance_loss_clip": 1.06072116, "balance_loss_mlp": 1.02563882, "epoch": 0.38405579270125656, "flos": 17019363397440.0, "grad_norm": 2.1460347082739197, "language_loss": 0.77015233, "learning_rate": 2.8230926128626095e-06, "loss": 0.79246819, "num_input_tokens_seen": 68816880, "step": 3194, "time_per_iteration": 2.510606527328491 }, { "auxiliary_loss_clip": 0.01159231, "auxiliary_loss_mlp": 0.01030879, "balance_loss_clip": 1.05253994, "balance_loss_mlp": 1.02255845, "epoch": 0.3841760355918956, "flos": 21835379561760.0, "grad_norm": 1.7646736390336792, "language_loss": 0.7979722, "learning_rate": 2.822382604374738e-06, "loss": 0.81987333, "num_input_tokens_seen": 68835805, "step": 3195, "time_per_iteration": 2.5948994159698486 }, { "auxiliary_loss_clip": 0.01165984, "auxiliary_loss_mlp": 0.01033857, "balance_loss_clip": 1.05734563, "balance_loss_mlp": 1.02550018, "epoch": 0.3842962784825347, "flos": 25915118061600.0, "grad_norm": 3.4382847187545447, "language_loss": 0.65812564, "learning_rate": 2.8216724711320793e-06, "loss": 0.68012398, "num_input_tokens_seen": 68854930, "step": 3196, "time_per_iteration": 2.589174509048462 }, { "auxiliary_loss_clip": 0.01192211, "auxiliary_loss_mlp": 0.00762412, "balance_loss_clip": 1.05723166, "balance_loss_mlp": 1.00035095, "epoch": 0.38441652137317384, "flos": 25337495973120.0, "grad_norm": 1.46072926842059, "language_loss": 0.79525006, "learning_rate": 2.820962213242361e-06, "loss": 0.81479627, "num_input_tokens_seen": 68874260, "step": 3197, "time_per_iteration": 2.5638411045074463 }, { "auxiliary_loss_clip": 0.01177673, "auxiliary_loss_mlp": 0.01031723, "balance_loss_clip": 1.05908966, "balance_loss_mlp": 1.02410305, "epoch": 0.3845367642638129, "flos": 18113489685120.0, "grad_norm": 1.9906972601806592, "language_loss": 0.84469497, "learning_rate": 2.8202518308133264e-06, "loss": 0.86678892, "num_input_tokens_seen": 68891535, "step": 3198, "time_per_iteration": 2.517045497894287 }, { "auxiliary_loss_clip": 0.0119406, "auxiliary_loss_mlp": 0.01026117, "balance_loss_clip": 1.05666184, "balance_loss_mlp": 1.01738453, "epoch": 0.384657007154452, "flos": 25228399413600.0, "grad_norm": 6.19179209130631, "language_loss": 0.73850012, "learning_rate": 2.8195413239527426e-06, "loss": 0.76070189, "num_input_tokens_seen": 68911275, "step": 3199, "time_per_iteration": 2.544447660446167 }, { "auxiliary_loss_clip": 0.01173243, "auxiliary_loss_mlp": 0.01028261, "balance_loss_clip": 1.05340266, "balance_loss_mlp": 1.02036381, "epoch": 0.38477725004509106, "flos": 19865858859840.0, "grad_norm": 1.9399587327675394, "language_loss": 0.80470455, "learning_rate": 2.8188306927683906e-06, "loss": 0.82671964, "num_input_tokens_seen": 68930745, "step": 3200, "time_per_iteration": 2.527104139328003 }, { "auxiliary_loss_clip": 0.01168304, "auxiliary_loss_mlp": 0.01026586, "balance_loss_clip": 1.05710793, "balance_loss_mlp": 1.01887369, "epoch": 0.38489749293573017, "flos": 18259394740320.0, "grad_norm": 2.1211844416130163, "language_loss": 0.7438212, "learning_rate": 2.818119937368074e-06, "loss": 0.76577014, "num_input_tokens_seen": 68949380, "step": 3201, "time_per_iteration": 2.57975435256958 }, { "auxiliary_loss_clip": 0.01186245, "auxiliary_loss_mlp": 0.01029291, "balance_loss_clip": 1.05658817, "balance_loss_mlp": 1.02125049, "epoch": 0.3850177358263693, "flos": 24389166989760.0, "grad_norm": 1.8963672678673655, "language_loss": 0.65680814, "learning_rate": 2.817409057859613e-06, "loss": 0.67896354, "num_input_tokens_seen": 68968370, "step": 3202, "time_per_iteration": 2.560770273208618 }, { "auxiliary_loss_clip": 0.01132396, "auxiliary_loss_mlp": 0.01031859, "balance_loss_clip": 1.05251765, "balance_loss_mlp": 1.02292395, "epoch": 0.38513797871700833, "flos": 17671536318240.0, "grad_norm": 1.8052858755266432, "language_loss": 0.79270506, "learning_rate": 2.8166980543508482e-06, "loss": 0.81434757, "num_input_tokens_seen": 68984260, "step": 3203, "time_per_iteration": 2.7130706310272217 }, { "auxiliary_loss_clip": 0.01197411, "auxiliary_loss_mlp": 0.01028074, "balance_loss_clip": 1.05972934, "balance_loss_mlp": 1.02021825, "epoch": 0.38525822160764744, "flos": 25739587902720.0, "grad_norm": 1.8846484872722402, "language_loss": 0.79725039, "learning_rate": 2.815986926949638e-06, "loss": 0.81950521, "num_input_tokens_seen": 69002760, "step": 3204, "time_per_iteration": 2.5138297080993652 }, { "auxiliary_loss_clip": 0.01179442, "auxiliary_loss_mlp": 0.01029025, "balance_loss_clip": 1.05765069, "balance_loss_mlp": 1.02104425, "epoch": 0.38537846449828655, "flos": 20193651375840.0, "grad_norm": 1.6493898761779031, "language_loss": 0.80298007, "learning_rate": 2.8152756757638597e-06, "loss": 0.82506478, "num_input_tokens_seen": 69021260, "step": 3205, "time_per_iteration": 2.5640838146209717 }, { "auxiliary_loss_clip": 0.0117977, "auxiliary_loss_mlp": 0.01029703, "balance_loss_clip": 1.06049705, "balance_loss_mlp": 1.02170372, "epoch": 0.3854987073889256, "flos": 23039356665120.0, "grad_norm": 1.8773683770170961, "language_loss": 0.84495991, "learning_rate": 2.8145643009014093e-06, "loss": 0.86705464, "num_input_tokens_seen": 69039755, "step": 3206, "time_per_iteration": 2.5304713249206543 }, { "auxiliary_loss_clip": 0.01180714, "auxiliary_loss_mlp": 0.0103035, "balance_loss_clip": 1.05676806, "balance_loss_mlp": 1.02291739, "epoch": 0.3856189502795647, "flos": 20190634351200.0, "grad_norm": 1.8268589958750798, "language_loss": 0.78753126, "learning_rate": 2.813852802470202e-06, "loss": 0.80964184, "num_input_tokens_seen": 69057650, "step": 3207, "time_per_iteration": 2.546668291091919 }, { "auxiliary_loss_clip": 0.01158899, "auxiliary_loss_mlp": 0.01029153, "balance_loss_clip": 1.05415392, "balance_loss_mlp": 1.02054596, "epoch": 0.38573919317020383, "flos": 25702635739200.0, "grad_norm": 2.0430548191802433, "language_loss": 0.72236872, "learning_rate": 2.8131411805781717e-06, "loss": 0.74424922, "num_input_tokens_seen": 69077775, "step": 3208, "time_per_iteration": 2.587857961654663 }, { "auxiliary_loss_clip": 0.01170355, "auxiliary_loss_mlp": 0.0103532, "balance_loss_clip": 1.05906081, "balance_loss_mlp": 1.02618289, "epoch": 0.3858594360608429, "flos": 29821409586240.0, "grad_norm": 2.63829734557329, "language_loss": 0.64122999, "learning_rate": 2.8124294353332707e-06, "loss": 0.66328675, "num_input_tokens_seen": 69096450, "step": 3209, "time_per_iteration": 2.6293785572052 }, { "auxiliary_loss_clip": 0.01157661, "auxiliary_loss_mlp": 0.01028704, "balance_loss_clip": 1.05516839, "balance_loss_mlp": 1.02026415, "epoch": 0.385979678951482, "flos": 24790432829280.0, "grad_norm": 1.6667759392839139, "language_loss": 0.77430689, "learning_rate": 2.8117175668434713e-06, "loss": 0.79617059, "num_input_tokens_seen": 69116110, "step": 3210, "time_per_iteration": 2.6449849605560303 }, { "auxiliary_loss_clip": 0.0119508, "auxiliary_loss_mlp": 0.01026951, "balance_loss_clip": 1.05748737, "balance_loss_mlp": 1.01882696, "epoch": 0.3860999218421211, "flos": 21287885414400.0, "grad_norm": 2.097460612727515, "language_loss": 0.69847316, "learning_rate": 2.811005575216762e-06, "loss": 0.72069347, "num_input_tokens_seen": 69134825, "step": 3211, "time_per_iteration": 3.288923978805542 }, { "auxiliary_loss_clip": 0.0114037, "auxiliary_loss_mlp": 0.01031372, "balance_loss_clip": 1.04858017, "balance_loss_mlp": 1.0235641, "epoch": 0.38622016473276016, "flos": 24536724225120.0, "grad_norm": 1.4735734910759641, "language_loss": 0.78835714, "learning_rate": 2.8102934605611513e-06, "loss": 0.81007457, "num_input_tokens_seen": 69156460, "step": 3212, "time_per_iteration": 3.6177468299865723 }, { "auxiliary_loss_clip": 0.01168719, "auxiliary_loss_mlp": 0.01030333, "balance_loss_clip": 1.0547657, "balance_loss_mlp": 1.02263856, "epoch": 0.3863404076233993, "flos": 20558216470560.0, "grad_norm": 2.0785184885909613, "language_loss": 0.67007148, "learning_rate": 2.8095812229846665e-06, "loss": 0.69206202, "num_input_tokens_seen": 69176420, "step": 3213, "time_per_iteration": 2.578375816345215 }, { "auxiliary_loss_clip": 0.01164277, "auxiliary_loss_mlp": 0.01032856, "balance_loss_clip": 1.053177, "balance_loss_mlp": 1.02445769, "epoch": 0.3864606505140384, "flos": 22346280715200.0, "grad_norm": 2.3715367656956263, "language_loss": 0.69334811, "learning_rate": 2.808868862595355e-06, "loss": 0.7153194, "num_input_tokens_seen": 69196665, "step": 3214, "time_per_iteration": 2.5972602367401123 }, { "auxiliary_loss_clip": 0.01183384, "auxiliary_loss_mlp": 0.0102721, "balance_loss_clip": 1.0565412, "balance_loss_mlp": 1.01950276, "epoch": 0.38658089340467744, "flos": 25703605497120.0, "grad_norm": 2.904462055576333, "language_loss": 0.79430318, "learning_rate": 2.8081563795012795e-06, "loss": 0.81640911, "num_input_tokens_seen": 69216290, "step": 3215, "time_per_iteration": 3.2780375480651855 }, { "auxiliary_loss_clip": 0.01172416, "auxiliary_loss_mlp": 0.01031873, "balance_loss_clip": 1.05392003, "balance_loss_mlp": 1.0233016, "epoch": 0.38670113629531655, "flos": 33802539278880.0, "grad_norm": 1.7563315680482452, "language_loss": 0.73935294, "learning_rate": 2.807443773810524e-06, "loss": 0.76139581, "num_input_tokens_seen": 69237550, "step": 3216, "time_per_iteration": 2.708890438079834 }, { "auxiliary_loss_clip": 0.01152532, "auxiliary_loss_mlp": 0.01034244, "balance_loss_clip": 1.05605638, "balance_loss_mlp": 1.02530336, "epoch": 0.3868213791859556, "flos": 23331525945120.0, "grad_norm": 1.8896901489318436, "language_loss": 0.89619744, "learning_rate": 2.80673104563119e-06, "loss": 0.91806519, "num_input_tokens_seen": 69258175, "step": 3217, "time_per_iteration": 2.618727922439575 }, { "auxiliary_loss_clip": 0.01174393, "auxiliary_loss_mlp": 0.01026292, "balance_loss_clip": 1.05567026, "balance_loss_mlp": 1.01898456, "epoch": 0.3869416220765947, "flos": 18441533619840.0, "grad_norm": 1.9347187524909482, "language_loss": 0.7847271, "learning_rate": 2.8060181950713976e-06, "loss": 0.80673385, "num_input_tokens_seen": 69274965, "step": 3218, "time_per_iteration": 3.289970874786377 }, { "auxiliary_loss_clip": 0.01147482, "auxiliary_loss_mlp": 0.01030606, "balance_loss_clip": 1.05163741, "balance_loss_mlp": 1.02199304, "epoch": 0.3870618649672338, "flos": 15632995995840.0, "grad_norm": 2.7738423728666, "language_loss": 0.80731618, "learning_rate": 2.805305222239286e-06, "loss": 0.82909703, "num_input_tokens_seen": 69292220, "step": 3219, "time_per_iteration": 2.565964698791504 }, { "auxiliary_loss_clip": 0.0116025, "auxiliary_loss_mlp": 0.01033392, "balance_loss_clip": 1.05464768, "balance_loss_mlp": 1.02513719, "epoch": 0.3871821078578729, "flos": 23513808492480.0, "grad_norm": 1.823385732580988, "language_loss": 0.74376714, "learning_rate": 2.8045921272430118e-06, "loss": 0.76570356, "num_input_tokens_seen": 69311900, "step": 3220, "time_per_iteration": 2.6097090244293213 }, { "auxiliary_loss_clip": 0.01185272, "auxiliary_loss_mlp": 0.01036038, "balance_loss_clip": 1.05578673, "balance_loss_mlp": 1.02729392, "epoch": 0.387302350748512, "flos": 17778262358400.0, "grad_norm": 2.126734818471168, "language_loss": 0.76475376, "learning_rate": 2.803878910190753e-06, "loss": 0.7869668, "num_input_tokens_seen": 69328820, "step": 3221, "time_per_iteration": 2.5126590728759766 }, { "auxiliary_loss_clip": 0.01179703, "auxiliary_loss_mlp": 0.01031133, "balance_loss_clip": 1.05321658, "balance_loss_mlp": 1.0229286, "epoch": 0.3874225936391511, "flos": 11503411143840.0, "grad_norm": 2.289942279874975, "language_loss": 0.81640059, "learning_rate": 2.8031655711907017e-06, "loss": 0.83850896, "num_input_tokens_seen": 69342525, "step": 3222, "time_per_iteration": 2.493882179260254 }, { "auxiliary_loss_clip": 0.01181565, "auxiliary_loss_mlp": 0.01031381, "balance_loss_clip": 1.05756068, "balance_loss_mlp": 1.0233103, "epoch": 0.38754283652979016, "flos": 21945158543520.0, "grad_norm": 1.982436080080433, "language_loss": 0.80632901, "learning_rate": 2.8024521103510723e-06, "loss": 0.82845843, "num_input_tokens_seen": 69359295, "step": 3223, "time_per_iteration": 2.535447835922241 }, { "auxiliary_loss_clip": 0.01175222, "auxiliary_loss_mlp": 0.01032165, "balance_loss_clip": 1.05238032, "balance_loss_mlp": 1.02448213, "epoch": 0.38766307942042927, "flos": 21175987332000.0, "grad_norm": 1.6428449558102747, "language_loss": 0.754484, "learning_rate": 2.8017385277800952e-06, "loss": 0.77655792, "num_input_tokens_seen": 69377650, "step": 3224, "time_per_iteration": 2.545259714126587 }, { "auxiliary_loss_clip": 0.01155366, "auxiliary_loss_mlp": 0.01033203, "balance_loss_clip": 1.05412698, "balance_loss_mlp": 1.0248462, "epoch": 0.3877833223110684, "flos": 27417298494240.0, "grad_norm": 2.0441104576871187, "language_loss": 0.75299656, "learning_rate": 2.8010248235860213e-06, "loss": 0.77488226, "num_input_tokens_seen": 69397765, "step": 3225, "time_per_iteration": 2.728196144104004 }, { "auxiliary_loss_clip": 0.01064616, "auxiliary_loss_mlp": 0.00753827, "balance_loss_clip": 1.0161407, "balance_loss_mlp": 1.00010002, "epoch": 0.38790356520170743, "flos": 64500032089920.0, "grad_norm": 0.8569340304589519, "language_loss": 0.62826657, "learning_rate": 2.8003109978771192e-06, "loss": 0.646451, "num_input_tokens_seen": 69458930, "step": 3226, "time_per_iteration": 3.2127065658569336 }, { "auxiliary_loss_clip": 0.01141523, "auxiliary_loss_mlp": 0.01032716, "balance_loss_clip": 1.04939985, "balance_loss_mlp": 1.02419806, "epoch": 0.38802380809234654, "flos": 22345418708160.0, "grad_norm": 2.0272432791829877, "language_loss": 0.79292321, "learning_rate": 2.799597050761674e-06, "loss": 0.81466568, "num_input_tokens_seen": 69475135, "step": 3227, "time_per_iteration": 2.6195902824401855 }, { "auxiliary_loss_clip": 0.01195147, "auxiliary_loss_mlp": 0.01034788, "balance_loss_clip": 1.058478, "balance_loss_mlp": 1.02630007, "epoch": 0.38814405098298566, "flos": 25261364794560.0, "grad_norm": 1.941435368349136, "language_loss": 0.7912246, "learning_rate": 2.7988829823479924e-06, "loss": 0.81352395, "num_input_tokens_seen": 69493525, "step": 3228, "time_per_iteration": 2.5367801189422607 }, { "auxiliary_loss_clip": 0.01157624, "auxiliary_loss_mlp": 0.01027045, "balance_loss_clip": 1.05319977, "balance_loss_mlp": 1.01889133, "epoch": 0.3882642938736247, "flos": 18841183196160.0, "grad_norm": 1.7903295557128918, "language_loss": 0.63873208, "learning_rate": 2.7981687927443976e-06, "loss": 0.66057873, "num_input_tokens_seen": 69510325, "step": 3229, "time_per_iteration": 2.5681498050689697 }, { "auxiliary_loss_clip": 0.01175649, "auxiliary_loss_mlp": 0.01025165, "balance_loss_clip": 1.05318761, "balance_loss_mlp": 1.01748788, "epoch": 0.3883845367642638, "flos": 21652809678720.0, "grad_norm": 1.8843703200565063, "language_loss": 0.85621518, "learning_rate": 2.797454482059231e-06, "loss": 0.8782233, "num_input_tokens_seen": 69530480, "step": 3230, "time_per_iteration": 2.549926519393921 }, { "auxiliary_loss_clip": 0.01195306, "auxiliary_loss_mlp": 0.01028575, "balance_loss_clip": 1.05817318, "balance_loss_mlp": 1.02046239, "epoch": 0.3885047796549029, "flos": 20557534048320.0, "grad_norm": 2.8561969511322753, "language_loss": 0.84250164, "learning_rate": 2.7967400504008537e-06, "loss": 0.86474043, "num_input_tokens_seen": 69549780, "step": 3231, "time_per_iteration": 2.5284264087677 }, { "auxiliary_loss_clip": 0.01034624, "auxiliary_loss_mlp": 0.01002075, "balance_loss_clip": 1.01378608, "balance_loss_mlp": 1.00051332, "epoch": 0.388625022545542, "flos": 64325501217600.0, "grad_norm": 0.7940672095475416, "language_loss": 0.57458186, "learning_rate": 2.7960254978776456e-06, "loss": 0.59494883, "num_input_tokens_seen": 69611870, "step": 3232, "time_per_iteration": 3.174234628677368 }, { "auxiliary_loss_clip": 0.01196937, "auxiliary_loss_mlp": 0.0102899, "balance_loss_clip": 1.05961072, "balance_loss_mlp": 1.02031708, "epoch": 0.3887452654361811, "flos": 18113884771680.0, "grad_norm": 1.8945317416943261, "language_loss": 0.81864846, "learning_rate": 2.7953108245980006e-06, "loss": 0.84090775, "num_input_tokens_seen": 69630385, "step": 3233, "time_per_iteration": 2.503823757171631 }, { "auxiliary_loss_clip": 0.01160607, "auxiliary_loss_mlp": 0.0102775, "balance_loss_clip": 1.05682206, "balance_loss_mlp": 1.01990604, "epoch": 0.38886550832682015, "flos": 24975265480800.0, "grad_norm": 1.78710635683167, "language_loss": 0.74086058, "learning_rate": 2.7945960306703365e-06, "loss": 0.76274413, "num_input_tokens_seen": 69653370, "step": 3234, "time_per_iteration": 2.582723617553711 }, { "auxiliary_loss_clip": 0.01184451, "auxiliary_loss_mlp": 0.0103001, "balance_loss_clip": 1.05810738, "balance_loss_mlp": 1.02153373, "epoch": 0.38898575121745926, "flos": 27199500461760.0, "grad_norm": 1.7255327388261748, "language_loss": 0.652143, "learning_rate": 2.7938811162030865e-06, "loss": 0.67428756, "num_input_tokens_seen": 69673635, "step": 3235, "time_per_iteration": 2.6041693687438965 }, { "auxiliary_loss_clip": 0.0117776, "auxiliary_loss_mlp": 0.01028114, "balance_loss_clip": 1.0565201, "balance_loss_mlp": 1.02125049, "epoch": 0.3891059941080984, "flos": 28763732624640.0, "grad_norm": 1.718213147226177, "language_loss": 0.82331312, "learning_rate": 2.793166081304702e-06, "loss": 0.84537184, "num_input_tokens_seen": 69694130, "step": 3236, "time_per_iteration": 2.581494092941284 }, { "auxiliary_loss_clip": 0.01157914, "auxiliary_loss_mlp": 0.01030211, "balance_loss_clip": 1.05375838, "balance_loss_mlp": 1.02174079, "epoch": 0.38922623699873743, "flos": 22893451609920.0, "grad_norm": 2.182726076622821, "language_loss": 0.82569003, "learning_rate": 2.7924509260836543e-06, "loss": 0.84757131, "num_input_tokens_seen": 69713255, "step": 3237, "time_per_iteration": 3.3623898029327393 }, { "auxiliary_loss_clip": 0.01150229, "auxiliary_loss_mlp": 0.01029226, "balance_loss_clip": 1.0541929, "balance_loss_mlp": 1.02129865, "epoch": 0.38934647988937654, "flos": 19792421453280.0, "grad_norm": 1.54540193005233, "language_loss": 0.68696004, "learning_rate": 2.791735650648431e-06, "loss": 0.7087546, "num_input_tokens_seen": 69732375, "step": 3238, "time_per_iteration": 3.3547656536102295 }, { "auxiliary_loss_clip": 0.01162973, "auxiliary_loss_mlp": 0.01032045, "balance_loss_clip": 1.05584383, "balance_loss_mlp": 1.02367687, "epoch": 0.38946672278001565, "flos": 19202084760960.0, "grad_norm": 1.890604528420001, "language_loss": 0.74628967, "learning_rate": 2.791020255107538e-06, "loss": 0.76823986, "num_input_tokens_seen": 69749745, "step": 3239, "time_per_iteration": 2.534426689147949 }, { "auxiliary_loss_clip": 0.01144203, "auxiliary_loss_mlp": 0.01032222, "balance_loss_clip": 1.04978681, "balance_loss_mlp": 1.02401483, "epoch": 0.3895869656706547, "flos": 24936481552320.0, "grad_norm": 1.7146209363316327, "language_loss": 0.80981344, "learning_rate": 2.7903047395695023e-06, "loss": 0.83157766, "num_input_tokens_seen": 69769645, "step": 3240, "time_per_iteration": 2.637269973754883 }, { "auxiliary_loss_clip": 0.01181199, "auxiliary_loss_mlp": 0.00762794, "balance_loss_clip": 1.05933046, "balance_loss_mlp": 1.00040936, "epoch": 0.3897072085612938, "flos": 24133626620640.0, "grad_norm": 2.0599500991183404, "language_loss": 0.89759457, "learning_rate": 2.789589104142865e-06, "loss": 0.91703463, "num_input_tokens_seen": 69787270, "step": 3241, "time_per_iteration": 3.281764030456543 }, { "auxiliary_loss_clip": 0.01151433, "auxiliary_loss_mlp": 0.01033885, "balance_loss_clip": 1.05210316, "balance_loss_mlp": 1.02624381, "epoch": 0.3898274514519329, "flos": 17166345961440.0, "grad_norm": 1.5898830354649223, "language_loss": 0.76538271, "learning_rate": 2.7888733489361895e-06, "loss": 0.78723586, "num_input_tokens_seen": 69805685, "step": 3242, "time_per_iteration": 2.5728297233581543 }, { "auxiliary_loss_clip": 0.01082565, "auxiliary_loss_mlp": 0.01001407, "balance_loss_clip": 1.0171845, "balance_loss_mlp": 0.99999428, "epoch": 0.389947694342572, "flos": 66074817450720.0, "grad_norm": 0.7269039614792278, "language_loss": 0.58735192, "learning_rate": 2.788157474058054e-06, "loss": 0.60819161, "num_input_tokens_seen": 69867960, "step": 3243, "time_per_iteration": 3.2326862812042236 }, { "auxiliary_loss_clip": 0.01190886, "auxiliary_loss_mlp": 0.01022505, "balance_loss_clip": 1.0575943, "balance_loss_mlp": 1.01522112, "epoch": 0.3900679372332111, "flos": 25740916830240.0, "grad_norm": 1.6648052478752102, "language_loss": 0.69955385, "learning_rate": 2.7874414796170555e-06, "loss": 0.72168779, "num_input_tokens_seen": 69889450, "step": 3244, "time_per_iteration": 3.2924304008483887 }, { "auxiliary_loss_clip": 0.01173372, "auxiliary_loss_mlp": 0.01035127, "balance_loss_clip": 1.05356789, "balance_loss_mlp": 1.02671933, "epoch": 0.3901881801238502, "flos": 11801614473120.0, "grad_norm": 8.100364644527902, "language_loss": 0.84055901, "learning_rate": 2.7867253657218113e-06, "loss": 0.86264402, "num_input_tokens_seen": 69903340, "step": 3245, "time_per_iteration": 2.5324885845184326 }, { "auxiliary_loss_clip": 0.01165934, "auxiliary_loss_mlp": 0.00762912, "balance_loss_clip": 1.05470252, "balance_loss_mlp": 1.00022805, "epoch": 0.39030842301448926, "flos": 27308956190880.0, "grad_norm": 1.5922531147069305, "language_loss": 0.7278555, "learning_rate": 2.7860091324809544e-06, "loss": 0.74714398, "num_input_tokens_seen": 69924400, "step": 3246, "time_per_iteration": 2.607818126678467 }, { "auxiliary_loss_clip": 0.01178287, "auxiliary_loss_mlp": 0.01030448, "balance_loss_clip": 1.05929708, "balance_loss_mlp": 1.02233565, "epoch": 0.39042866590512837, "flos": 27163338471360.0, "grad_norm": 1.5342677146750419, "language_loss": 0.81255591, "learning_rate": 2.7852927800031377e-06, "loss": 0.83464324, "num_input_tokens_seen": 69944565, "step": 3247, "time_per_iteration": 2.6276280879974365 }, { "auxiliary_loss_clip": 0.01167522, "auxiliary_loss_mlp": 0.01030698, "balance_loss_clip": 1.05657923, "balance_loss_mlp": 1.02316999, "epoch": 0.3905489087957674, "flos": 29716120224480.0, "grad_norm": 1.9175580693574998, "language_loss": 0.82701612, "learning_rate": 2.7845763083970298e-06, "loss": 0.84899837, "num_input_tokens_seen": 69964965, "step": 3248, "time_per_iteration": 2.6152873039245605 }, { "auxiliary_loss_clip": 0.01170325, "auxiliary_loss_mlp": 0.01032067, "balance_loss_clip": 1.05372095, "balance_loss_mlp": 1.02391899, "epoch": 0.39066915168640653, "flos": 24498622718880.0, "grad_norm": 3.543996088398923, "language_loss": 0.82352793, "learning_rate": 2.7838597177713205e-06, "loss": 0.84555185, "num_input_tokens_seen": 69986055, "step": 3249, "time_per_iteration": 2.604646682739258 }, { "auxiliary_loss_clip": 0.01110777, "auxiliary_loss_mlp": 0.01029965, "balance_loss_clip": 1.05186987, "balance_loss_mlp": 1.02163267, "epoch": 0.39078939457704565, "flos": 20558575640160.0, "grad_norm": 2.8073934131091494, "language_loss": 0.73288351, "learning_rate": 2.7831430082347143e-06, "loss": 0.75429094, "num_input_tokens_seen": 70005260, "step": 3250, "time_per_iteration": 2.647597551345825 }, { "auxiliary_loss_clip": 0.01181698, "auxiliary_loss_mlp": 0.00762508, "balance_loss_clip": 1.05805838, "balance_loss_mlp": 1.00025916, "epoch": 0.3909096374676847, "flos": 22783421209440.0, "grad_norm": 1.8968048801680755, "language_loss": 0.81626791, "learning_rate": 2.7824261798959373e-06, "loss": 0.83570993, "num_input_tokens_seen": 70023440, "step": 3251, "time_per_iteration": 2.595273733139038 }, { "auxiliary_loss_clip": 0.01167793, "auxiliary_loss_mlp": 0.01036868, "balance_loss_clip": 1.05328107, "balance_loss_mlp": 1.02827287, "epoch": 0.3910298803583238, "flos": 23003122840800.0, "grad_norm": 1.7939013074106063, "language_loss": 0.79456604, "learning_rate": 2.78170923286373e-06, "loss": 0.8166126, "num_input_tokens_seen": 70043040, "step": 3252, "time_per_iteration": 2.5538792610168457 }, { "auxiliary_loss_clip": 0.01099336, "auxiliary_loss_mlp": 0.01034853, "balance_loss_clip": 1.04821432, "balance_loss_mlp": 1.02626407, "epoch": 0.3911501232489629, "flos": 24316268337600.0, "grad_norm": 4.7543468646887215, "language_loss": 0.84407806, "learning_rate": 2.780992167246854e-06, "loss": 0.86541998, "num_input_tokens_seen": 70060565, "step": 3253, "time_per_iteration": 2.7256648540496826 }, { "auxiliary_loss_clip": 0.01061127, "auxiliary_loss_mlp": 0.0100416, "balance_loss_clip": 1.01365566, "balance_loss_mlp": 1.0028367, "epoch": 0.391270366139602, "flos": 60869064791040.0, "grad_norm": 1.0264328926227522, "language_loss": 0.72149658, "learning_rate": 2.7802749831540883e-06, "loss": 0.74214947, "num_input_tokens_seen": 70119465, "step": 3254, "time_per_iteration": 3.1616134643554688 }, { "auxiliary_loss_clip": 0.01142135, "auxiliary_loss_mlp": 0.01028041, "balance_loss_clip": 1.05442894, "balance_loss_mlp": 1.02094853, "epoch": 0.3913906090302411, "flos": 21543497617440.0, "grad_norm": 2.189935753803401, "language_loss": 0.81712425, "learning_rate": 2.7795576806942268e-06, "loss": 0.838826, "num_input_tokens_seen": 70138270, "step": 3255, "time_per_iteration": 2.6575348377227783 }, { "auxiliary_loss_clip": 0.01062417, "auxiliary_loss_mlp": 0.01008694, "balance_loss_clip": 1.02427292, "balance_loss_mlp": 1.00746596, "epoch": 0.3915108519208802, "flos": 49839961806720.0, "grad_norm": 0.7548725983663169, "language_loss": 0.54897022, "learning_rate": 2.778840259976085e-06, "loss": 0.56968141, "num_input_tokens_seen": 70193500, "step": 3256, "time_per_iteration": 3.124807357788086 }, { "auxiliary_loss_clip": 0.01185109, "auxiliary_loss_mlp": 0.01031176, "balance_loss_clip": 1.06087828, "balance_loss_mlp": 1.02259851, "epoch": 0.39163109481151925, "flos": 16506450894240.0, "grad_norm": 2.340760292437256, "language_loss": 0.77088928, "learning_rate": 2.778122721108495e-06, "loss": 0.79305214, "num_input_tokens_seen": 70211730, "step": 3257, "time_per_iteration": 2.532135248184204 }, { "auxiliary_loss_clip": 0.01176795, "auxiliary_loss_mlp": 0.01026065, "balance_loss_clip": 1.057374, "balance_loss_mlp": 1.01823545, "epoch": 0.39175133770215836, "flos": 26067488169600.0, "grad_norm": 1.8200858545581513, "language_loss": 0.87967908, "learning_rate": 2.7774050642003076e-06, "loss": 0.90170765, "num_input_tokens_seen": 70232540, "step": 3258, "time_per_iteration": 2.5900235176086426 }, { "auxiliary_loss_clip": 0.01196297, "auxiliary_loss_mlp": 0.01032347, "balance_loss_clip": 1.05934632, "balance_loss_mlp": 1.02434778, "epoch": 0.3918715805927975, "flos": 21872080306560.0, "grad_norm": 1.8953243901799885, "language_loss": 0.93520868, "learning_rate": 2.7766872893603896e-06, "loss": 0.95749521, "num_input_tokens_seen": 70252515, "step": 3259, "time_per_iteration": 2.543111562728882 }, { "auxiliary_loss_clip": 0.01180159, "auxiliary_loss_mlp": 0.0102985, "balance_loss_clip": 1.05798197, "balance_loss_mlp": 1.02249241, "epoch": 0.39199182348343653, "flos": 20376185341920.0, "grad_norm": 1.6016363170201968, "language_loss": 0.73207986, "learning_rate": 2.7759693966976275e-06, "loss": 0.75418001, "num_input_tokens_seen": 70271020, "step": 3260, "time_per_iteration": 2.521862506866455 }, { "auxiliary_loss_clip": 0.01146986, "auxiliary_loss_mlp": 0.01026012, "balance_loss_clip": 1.05251694, "balance_loss_mlp": 1.01756573, "epoch": 0.39211206637407564, "flos": 21683548208160.0, "grad_norm": 1.8306499341294984, "language_loss": 0.85381216, "learning_rate": 2.7752513863209242e-06, "loss": 0.87554204, "num_input_tokens_seen": 70289600, "step": 3261, "time_per_iteration": 2.575261354446411 }, { "auxiliary_loss_clip": 0.01162731, "auxiliary_loss_mlp": 0.00762497, "balance_loss_clip": 1.05815578, "balance_loss_mlp": 1.000211, "epoch": 0.39223230926471475, "flos": 21066280184160.0, "grad_norm": 1.6025420624247648, "language_loss": 0.84305727, "learning_rate": 2.774533258339203e-06, "loss": 0.86230958, "num_input_tokens_seen": 70307060, "step": 3262, "time_per_iteration": 2.5644049644470215 }, { "auxiliary_loss_clip": 0.01133696, "auxiliary_loss_mlp": 0.01030058, "balance_loss_clip": 1.04822278, "balance_loss_mlp": 1.02148724, "epoch": 0.3923525521553538, "flos": 17603019535200.0, "grad_norm": 2.462805285097147, "language_loss": 0.79896605, "learning_rate": 2.7738150128614014e-06, "loss": 0.82060355, "num_input_tokens_seen": 70324465, "step": 3263, "time_per_iteration": 3.431143283843994 }, { "auxiliary_loss_clip": 0.01138841, "auxiliary_loss_mlp": 0.01035629, "balance_loss_clip": 1.05059516, "balance_loss_mlp": 1.02729023, "epoch": 0.3924727950459929, "flos": 20558288304480.0, "grad_norm": 1.77210194110587, "language_loss": 0.89531493, "learning_rate": 2.7730966499964777e-06, "loss": 0.9170596, "num_input_tokens_seen": 70341415, "step": 3264, "time_per_iteration": 3.322793960571289 }, { "auxiliary_loss_clip": 0.01193526, "auxiliary_loss_mlp": 0.01028007, "balance_loss_clip": 1.05619049, "balance_loss_mlp": 1.01944757, "epoch": 0.39259303793663197, "flos": 16216113379200.0, "grad_norm": 2.3575297899015486, "language_loss": 0.8083117, "learning_rate": 2.772378169853408e-06, "loss": 0.83052707, "num_input_tokens_seen": 70358985, "step": 3265, "time_per_iteration": 2.5414364337921143 }, { "auxiliary_loss_clip": 0.0115181, "auxiliary_loss_mlp": 0.0103134, "balance_loss_clip": 1.05660236, "balance_loss_mlp": 1.02405024, "epoch": 0.3927132808272711, "flos": 16797003911040.0, "grad_norm": 1.780167258818397, "language_loss": 0.74278665, "learning_rate": 2.771659572541183e-06, "loss": 0.76461816, "num_input_tokens_seen": 70376915, "step": 3266, "time_per_iteration": 2.5810751914978027 }, { "auxiliary_loss_clip": 0.01183951, "auxiliary_loss_mlp": 0.01027419, "balance_loss_clip": 1.06065881, "balance_loss_mlp": 1.02040052, "epoch": 0.3928335237179102, "flos": 20267232450240.0, "grad_norm": 2.630293773983193, "language_loss": 0.87315488, "learning_rate": 2.7709408581688143e-06, "loss": 0.89526856, "num_input_tokens_seen": 70396900, "step": 3267, "time_per_iteration": 3.249601125717163 }, { "auxiliary_loss_clip": 0.0115345, "auxiliary_loss_mlp": 0.01024795, "balance_loss_clip": 1.05465412, "balance_loss_mlp": 1.01711798, "epoch": 0.39295376660854925, "flos": 24973254131040.0, "grad_norm": 1.5299373288981823, "language_loss": 0.87594962, "learning_rate": 2.7702220268453307e-06, "loss": 0.89773202, "num_input_tokens_seen": 70417260, "step": 3268, "time_per_iteration": 2.6344332695007324 }, { "auxiliary_loss_clip": 0.01169349, "auxiliary_loss_mlp": 0.01025951, "balance_loss_clip": 1.05666852, "balance_loss_mlp": 1.01807106, "epoch": 0.39307400949918836, "flos": 18697792328160.0, "grad_norm": 2.404151921154209, "language_loss": 0.85067475, "learning_rate": 2.7695030786797785e-06, "loss": 0.87262774, "num_input_tokens_seen": 70433155, "step": 3269, "time_per_iteration": 2.565685987472534 }, { "auxiliary_loss_clip": 0.01130561, "auxiliary_loss_mlp": 0.0103865, "balance_loss_clip": 1.04989457, "balance_loss_mlp": 1.03084195, "epoch": 0.39319425238982747, "flos": 22415479920480.0, "grad_norm": 2.1317941539666867, "language_loss": 0.7461803, "learning_rate": 2.7687840137812206e-06, "loss": 0.76787239, "num_input_tokens_seen": 70451240, "step": 3270, "time_per_iteration": 3.36689829826355 }, { "auxiliary_loss_clip": 0.01064945, "auxiliary_loss_mlp": 0.01004199, "balance_loss_clip": 1.01384425, "balance_loss_mlp": 1.00295377, "epoch": 0.3933144952804665, "flos": 66192965084160.0, "grad_norm": 0.9723325608383544, "language_loss": 0.62065446, "learning_rate": 2.7680648322587395e-06, "loss": 0.64134586, "num_input_tokens_seen": 70516115, "step": 3271, "time_per_iteration": 3.167815685272217 }, { "auxiliary_loss_clip": 0.01190037, "auxiliary_loss_mlp": 0.01028761, "balance_loss_clip": 1.05649447, "balance_loss_mlp": 1.02106595, "epoch": 0.39343473817110564, "flos": 15487162774560.0, "grad_norm": 1.8388772460471967, "language_loss": 0.81032789, "learning_rate": 2.7673455342214334e-06, "loss": 0.8325159, "num_input_tokens_seen": 70533105, "step": 3272, "time_per_iteration": 2.489351511001587 }, { "auxiliary_loss_clip": 0.01181374, "auxiliary_loss_mlp": 0.01027045, "balance_loss_clip": 1.05887675, "balance_loss_mlp": 1.01915288, "epoch": 0.39355498106174475, "flos": 21324909411840.0, "grad_norm": 2.3031437852605747, "language_loss": 0.76322424, "learning_rate": 2.7666261197784198e-06, "loss": 0.78530842, "num_input_tokens_seen": 70551920, "step": 3273, "time_per_iteration": 2.5365335941314697 }, { "auxiliary_loss_clip": 0.01160001, "auxiliary_loss_mlp": 0.01025056, "balance_loss_clip": 1.05619681, "balance_loss_mlp": 1.01770079, "epoch": 0.3936752239523838, "flos": 13296360095040.0, "grad_norm": 1.9080434237557409, "language_loss": 0.76817143, "learning_rate": 2.7659065890388336e-06, "loss": 0.79002202, "num_input_tokens_seen": 70567920, "step": 3274, "time_per_iteration": 2.504559278488159 }, { "auxiliary_loss_clip": 0.01165088, "auxiliary_loss_mlp": 0.0102863, "balance_loss_clip": 1.0537051, "balance_loss_mlp": 1.02055335, "epoch": 0.3937954668430229, "flos": 16800164603520.0, "grad_norm": 1.8151788595882419, "language_loss": 0.84764457, "learning_rate": 2.7651869421118266e-06, "loss": 0.8695817, "num_input_tokens_seen": 70584530, "step": 3275, "time_per_iteration": 2.533985137939453 }, { "auxiliary_loss_clip": 0.0118176, "auxiliary_loss_mlp": 0.01031801, "balance_loss_clip": 1.05871987, "balance_loss_mlp": 1.02392745, "epoch": 0.393915709733662, "flos": 21064232917440.0, "grad_norm": 1.620256607095292, "language_loss": 0.82708162, "learning_rate": 2.76446717910657e-06, "loss": 0.8492173, "num_input_tokens_seen": 70605235, "step": 3276, "time_per_iteration": 2.5472190380096436 }, { "auxiliary_loss_clip": 0.0117379, "auxiliary_loss_mlp": 0.01027792, "balance_loss_clip": 1.05521226, "balance_loss_mlp": 1.01992464, "epoch": 0.3940359526243011, "flos": 17165268452640.0, "grad_norm": 2.2383040655330055, "language_loss": 0.76366913, "learning_rate": 2.763747300132249e-06, "loss": 0.785685, "num_input_tokens_seen": 70622675, "step": 3277, "time_per_iteration": 2.5059592723846436 }, { "auxiliary_loss_clip": 0.01193165, "auxiliary_loss_mlp": 0.01026658, "balance_loss_clip": 1.05880344, "balance_loss_mlp": 1.01887667, "epoch": 0.3941561955149402, "flos": 20995859802240.0, "grad_norm": 1.7063978674841438, "language_loss": 0.86805916, "learning_rate": 2.7630273052980704e-06, "loss": 0.89025742, "num_input_tokens_seen": 70643265, "step": 3278, "time_per_iteration": 2.498309373855591 }, { "auxiliary_loss_clip": 0.01151144, "auxiliary_loss_mlp": 0.01028811, "balance_loss_clip": 1.05120254, "balance_loss_mlp": 1.0211165, "epoch": 0.39427643840557924, "flos": 18843410047680.0, "grad_norm": 2.035047045160157, "language_loss": 0.67300034, "learning_rate": 2.7623071947132554e-06, "loss": 0.6947999, "num_input_tokens_seen": 70660295, "step": 3279, "time_per_iteration": 2.5488216876983643 }, { "auxiliary_loss_clip": 0.01171902, "auxiliary_loss_mlp": 0.01026952, "balance_loss_clip": 1.05480313, "balance_loss_mlp": 1.01867306, "epoch": 0.39439668129621835, "flos": 23258663209920.0, "grad_norm": 1.894580259993956, "language_loss": 0.78689927, "learning_rate": 2.7615869684870458e-06, "loss": 0.80888784, "num_input_tokens_seen": 70679605, "step": 3280, "time_per_iteration": 2.5687170028686523 }, { "auxiliary_loss_clip": 0.01174746, "auxiliary_loss_mlp": 0.01037214, "balance_loss_clip": 1.05525887, "balance_loss_mlp": 1.02886355, "epoch": 0.39451692418685746, "flos": 26652293650080.0, "grad_norm": 1.8809956026652825, "language_loss": 0.84572905, "learning_rate": 2.7608666267286986e-06, "loss": 0.86784863, "num_input_tokens_seen": 70699835, "step": 3281, "time_per_iteration": 2.5677521228790283 }, { "auxiliary_loss_clip": 0.01115244, "auxiliary_loss_mlp": 0.01027576, "balance_loss_clip": 1.04605436, "balance_loss_mlp": 1.01889777, "epoch": 0.3946371670774965, "flos": 18258712318080.0, "grad_norm": 2.97509081413592, "language_loss": 0.86363435, "learning_rate": 2.760146169547489e-06, "loss": 0.88506258, "num_input_tokens_seen": 70716600, "step": 3282, "time_per_iteration": 2.6022863388061523 }, { "auxiliary_loss_clip": 0.0117003, "auxiliary_loss_mlp": 0.01032966, "balance_loss_clip": 1.05862641, "balance_loss_mlp": 1.02475274, "epoch": 0.39475740996813563, "flos": 24206130186240.0, "grad_norm": 1.4602886889645756, "language_loss": 0.76692319, "learning_rate": 2.75942559705271e-06, "loss": 0.78895319, "num_input_tokens_seen": 70736335, "step": 3283, "time_per_iteration": 2.588737726211548 }, { "auxiliary_loss_clip": 0.01175212, "auxiliary_loss_mlp": 0.01031411, "balance_loss_clip": 1.05661893, "balance_loss_mlp": 1.02349567, "epoch": 0.39487765285877474, "flos": 19317861875040.0, "grad_norm": 1.8317035395305938, "language_loss": 0.89117801, "learning_rate": 2.7587049093536713e-06, "loss": 0.91324413, "num_input_tokens_seen": 70752665, "step": 3284, "time_per_iteration": 2.508873224258423 }, { "auxiliary_loss_clip": 0.01183568, "auxiliary_loss_mlp": 0.01026699, "balance_loss_clip": 1.05813718, "balance_loss_mlp": 1.01899171, "epoch": 0.3949978957494138, "flos": 17311748179200.0, "grad_norm": 1.814061362397123, "language_loss": 0.8071754, "learning_rate": 2.757984106559701e-06, "loss": 0.82927805, "num_input_tokens_seen": 70771650, "step": 3285, "time_per_iteration": 2.5240304470062256 }, { "auxiliary_loss_clip": 0.01156572, "auxiliary_loss_mlp": 0.01028021, "balance_loss_clip": 1.05367267, "balance_loss_mlp": 1.02034998, "epoch": 0.3951181386400529, "flos": 36317866031040.0, "grad_norm": 2.2006281737706983, "language_loss": 0.71554947, "learning_rate": 2.7572631887801446e-06, "loss": 0.73739541, "num_input_tokens_seen": 70793275, "step": 3286, "time_per_iteration": 2.67494797706604 }, { "auxiliary_loss_clip": 0.01178207, "auxiliary_loss_mlp": 0.01027397, "balance_loss_clip": 1.05611455, "balance_loss_mlp": 1.01916552, "epoch": 0.395238381530692, "flos": 23110351718400.0, "grad_norm": 1.5365654460256906, "language_loss": 0.76607138, "learning_rate": 2.7565421561243654e-06, "loss": 0.78812742, "num_input_tokens_seen": 70811440, "step": 3287, "time_per_iteration": 2.5458409786224365 }, { "auxiliary_loss_clip": 0.01142564, "auxiliary_loss_mlp": 0.01030645, "balance_loss_clip": 1.05073404, "balance_loss_mlp": 1.02312863, "epoch": 0.3953586244213311, "flos": 24347617455360.0, "grad_norm": 2.235882612013833, "language_loss": 0.82035244, "learning_rate": 2.7558210087017413e-06, "loss": 0.84208447, "num_input_tokens_seen": 70831375, "step": 3288, "time_per_iteration": 2.602283477783203 }, { "auxiliary_loss_clip": 0.01142583, "auxiliary_loss_mlp": 0.01031039, "balance_loss_clip": 1.05345309, "balance_loss_mlp": 1.02300441, "epoch": 0.3954788673119702, "flos": 23440083750240.0, "grad_norm": 2.195917659502704, "language_loss": 0.73298776, "learning_rate": 2.7550997466216724e-06, "loss": 0.75472403, "num_input_tokens_seen": 70849170, "step": 3289, "time_per_iteration": 3.352875232696533 }, { "auxiliary_loss_clip": 0.01160727, "auxiliary_loss_mlp": 0.01031426, "balance_loss_clip": 1.05604029, "balance_loss_mlp": 1.02348661, "epoch": 0.3955991102026093, "flos": 17494066643520.0, "grad_norm": 2.1368424862304307, "language_loss": 0.81477964, "learning_rate": 2.7543783699935714e-06, "loss": 0.83670115, "num_input_tokens_seen": 70867200, "step": 3290, "time_per_iteration": 3.257194995880127 }, { "auxiliary_loss_clip": 0.01176799, "auxiliary_loss_mlp": 0.01027048, "balance_loss_clip": 1.05733109, "balance_loss_mlp": 1.0191865, "epoch": 0.39571935309324835, "flos": 18221329151040.0, "grad_norm": 2.360642034849856, "language_loss": 0.86006427, "learning_rate": 2.753656878926872e-06, "loss": 0.88210273, "num_input_tokens_seen": 70883080, "step": 3291, "time_per_iteration": 2.5057313442230225 }, { "auxiliary_loss_clip": 0.01150433, "auxiliary_loss_mlp": 0.01030094, "balance_loss_clip": 1.04958415, "balance_loss_mlp": 1.02196956, "epoch": 0.39583959598388746, "flos": 17748816839520.0, "grad_norm": 1.776078830287895, "language_loss": 0.74494678, "learning_rate": 2.752935273531023e-06, "loss": 0.766752, "num_input_tokens_seen": 70901230, "step": 3292, "time_per_iteration": 2.538145065307617 }, { "auxiliary_loss_clip": 0.01180054, "auxiliary_loss_mlp": 0.0102734, "balance_loss_clip": 1.05688238, "balance_loss_mlp": 1.01844096, "epoch": 0.39595983887452657, "flos": 19352371685280.0, "grad_norm": 2.5924471726128364, "language_loss": 0.78510201, "learning_rate": 2.752213553915492e-06, "loss": 0.80717593, "num_input_tokens_seen": 70919585, "step": 3293, "time_per_iteration": 3.296867609024048 }, { "auxiliary_loss_clip": 0.01056202, "auxiliary_loss_mlp": 0.01002656, "balance_loss_clip": 1.01543665, "balance_loss_mlp": 1.00121391, "epoch": 0.3960800817651656, "flos": 60682292623680.0, "grad_norm": 0.8353180180327063, "language_loss": 0.66076738, "learning_rate": 2.751491720189762e-06, "loss": 0.68135595, "num_input_tokens_seen": 70977695, "step": 3294, "time_per_iteration": 3.139005661010742 }, { "auxiliary_loss_clip": 0.01165394, "auxiliary_loss_mlp": 0.00762657, "balance_loss_clip": 1.05594742, "balance_loss_mlp": 1.00019336, "epoch": 0.39620032465580474, "flos": 16836721680480.0, "grad_norm": 2.1715704589721843, "language_loss": 0.91704905, "learning_rate": 2.7507697724633364e-06, "loss": 0.9363296, "num_input_tokens_seen": 70994455, "step": 3295, "time_per_iteration": 3.254849433898926 }, { "auxiliary_loss_clip": 0.01049826, "auxiliary_loss_mlp": 0.01003954, "balance_loss_clip": 1.02239549, "balance_loss_mlp": 1.00277936, "epoch": 0.3963205675464438, "flos": 69071456169600.0, "grad_norm": 0.7780732235212546, "language_loss": 0.54722059, "learning_rate": 2.7500477108457327e-06, "loss": 0.56775832, "num_input_tokens_seen": 71046465, "step": 3296, "time_per_iteration": 3.0110301971435547 }, { "auxiliary_loss_clip": 0.01174894, "auxiliary_loss_mlp": 0.01026969, "balance_loss_clip": 1.05455077, "balance_loss_mlp": 1.01860082, "epoch": 0.3964408104370829, "flos": 25667443506720.0, "grad_norm": 2.0010135344497932, "language_loss": 0.80368727, "learning_rate": 2.7493255354464877e-06, "loss": 0.82570589, "num_input_tokens_seen": 71064275, "step": 3297, "time_per_iteration": 2.5656747817993164 }, { "auxiliary_loss_clip": 0.01063452, "auxiliary_loss_mlp": 0.01029066, "balance_loss_clip": 1.04172647, "balance_loss_mlp": 1.02174044, "epoch": 0.396561053327722, "flos": 24277484409120.0, "grad_norm": 1.848170857137262, "language_loss": 0.76419979, "learning_rate": 2.748603246375156e-06, "loss": 0.78512502, "num_input_tokens_seen": 71082290, "step": 3298, "time_per_iteration": 2.8844974040985107 }, { "auxiliary_loss_clip": 0.01195103, "auxiliary_loss_mlp": 0.01030825, "balance_loss_clip": 1.06013942, "balance_loss_mlp": 1.02255177, "epoch": 0.39668129621836107, "flos": 20522305898880.0, "grad_norm": 2.3161228320500413, "language_loss": 0.69666374, "learning_rate": 2.7478808437413055e-06, "loss": 0.71892303, "num_input_tokens_seen": 71101700, "step": 3299, "time_per_iteration": 2.5951383113861084 }, { "auxiliary_loss_clip": 0.01134774, "auxiliary_loss_mlp": 0.0103098, "balance_loss_clip": 1.0540961, "balance_loss_mlp": 1.02320206, "epoch": 0.3968015391090002, "flos": 27052589731680.0, "grad_norm": 1.8291140774241006, "language_loss": 0.66180223, "learning_rate": 2.7471583276545263e-06, "loss": 0.68345976, "num_input_tokens_seen": 71122360, "step": 3300, "time_per_iteration": 2.8209919929504395 }, { "auxiliary_loss_clip": 0.01166166, "auxiliary_loss_mlp": 0.01029002, "balance_loss_clip": 1.0559963, "balance_loss_mlp": 1.02121127, "epoch": 0.3969217819996393, "flos": 12531822171360.0, "grad_norm": 1.7083002308214121, "language_loss": 0.70744944, "learning_rate": 2.7464356982244224e-06, "loss": 0.72940105, "num_input_tokens_seen": 71140360, "step": 3301, "time_per_iteration": 2.5818889141082764 }, { "auxiliary_loss_clip": 0.01070981, "auxiliary_loss_mlp": 0.01004874, "balance_loss_clip": 1.0217849, "balance_loss_mlp": 1.00375319, "epoch": 0.39704202489027834, "flos": 66241410674880.0, "grad_norm": 0.77654734115179, "language_loss": 0.61750948, "learning_rate": 2.745712955560617e-06, "loss": 0.63826805, "num_input_tokens_seen": 71196565, "step": 3302, "time_per_iteration": 3.0832059383392334 }, { "auxiliary_loss_clip": 0.01120912, "auxiliary_loss_mlp": 0.01037099, "balance_loss_clip": 1.05060315, "balance_loss_mlp": 1.02889776, "epoch": 0.39716226778091746, "flos": 16982985905280.0, "grad_norm": 2.52394878253303, "language_loss": 0.7721386, "learning_rate": 2.7449900997727496e-06, "loss": 0.7937187, "num_input_tokens_seen": 71214675, "step": 3303, "time_per_iteration": 2.678124189376831 }, { "auxiliary_loss_clip": 0.01160865, "auxiliary_loss_mlp": 0.01030975, "balance_loss_clip": 1.05595469, "balance_loss_mlp": 1.02372098, "epoch": 0.39728251067155657, "flos": 23477143664640.0, "grad_norm": 1.822717546044567, "language_loss": 0.84183788, "learning_rate": 2.744267130970476e-06, "loss": 0.8637563, "num_input_tokens_seen": 71234400, "step": 3304, "time_per_iteration": 2.5686559677124023 }, { "auxiliary_loss_clip": 0.01157024, "auxiliary_loss_mlp": 0.01027589, "balance_loss_clip": 1.05361891, "balance_loss_mlp": 1.01919627, "epoch": 0.3974027535621956, "flos": 20704444778400.0, "grad_norm": 1.9180102620213648, "language_loss": 0.76977473, "learning_rate": 2.7435440492634697e-06, "loss": 0.79162085, "num_input_tokens_seen": 71253725, "step": 3305, "time_per_iteration": 2.597339153289795 }, { "auxiliary_loss_clip": 0.01167547, "auxiliary_loss_mlp": 0.01030779, "balance_loss_clip": 1.05637896, "balance_loss_mlp": 1.02143908, "epoch": 0.39752299645283473, "flos": 21543282115680.0, "grad_norm": 1.8614513314201067, "language_loss": 0.67222476, "learning_rate": 2.7428208547614228e-06, "loss": 0.69420803, "num_input_tokens_seen": 71273220, "step": 3306, "time_per_iteration": 2.5574538707733154 }, { "auxiliary_loss_clip": 0.01178769, "auxiliary_loss_mlp": 0.01033002, "balance_loss_clip": 1.05682313, "balance_loss_mlp": 1.02467525, "epoch": 0.39764323934347384, "flos": 19208298395040.0, "grad_norm": 2.337120704325356, "language_loss": 0.77625918, "learning_rate": 2.742097547574043e-06, "loss": 0.79837686, "num_input_tokens_seen": 71291445, "step": 3307, "time_per_iteration": 2.539245367050171 }, { "auxiliary_loss_clip": 0.01170529, "auxiliary_loss_mlp": 0.0076272, "balance_loss_clip": 1.05516803, "balance_loss_mlp": 1.00027108, "epoch": 0.3977634822341129, "flos": 20850206165760.0, "grad_norm": 1.7692364920324775, "language_loss": 0.77634931, "learning_rate": 2.7413741278110544e-06, "loss": 0.79568183, "num_input_tokens_seen": 71310135, "step": 3308, "time_per_iteration": 2.570409059524536 }, { "auxiliary_loss_clip": 0.01169454, "auxiliary_loss_mlp": 0.01027863, "balance_loss_clip": 1.05731773, "balance_loss_mlp": 1.01952481, "epoch": 0.397883725124752, "flos": 39786047303520.0, "grad_norm": 2.3037413934460744, "language_loss": 0.6897943, "learning_rate": 2.7406505955822016e-06, "loss": 0.71176744, "num_input_tokens_seen": 71331160, "step": 3309, "time_per_iteration": 2.727358818054199 }, { "auxiliary_loss_clip": 0.0116133, "auxiliary_loss_mlp": 0.01034263, "balance_loss_clip": 1.05260086, "balance_loss_mlp": 1.02594197, "epoch": 0.39800396801539106, "flos": 17379510706080.0, "grad_norm": 3.330155660226693, "language_loss": 0.66043103, "learning_rate": 2.7399269509972415e-06, "loss": 0.68238699, "num_input_tokens_seen": 71345315, "step": 3310, "time_per_iteration": 2.511655330657959 }, { "auxiliary_loss_clip": 0.01156035, "auxiliary_loss_mlp": 0.01030452, "balance_loss_clip": 1.04975808, "balance_loss_mlp": 1.02156508, "epoch": 0.3981242109060302, "flos": 19202767183200.0, "grad_norm": 2.5403030990260267, "language_loss": 0.84950328, "learning_rate": 2.7392031941659514e-06, "loss": 0.87136817, "num_input_tokens_seen": 71363160, "step": 3311, "time_per_iteration": 2.5704500675201416 }, { "auxiliary_loss_clip": 0.01167053, "auxiliary_loss_mlp": 0.01027782, "balance_loss_clip": 1.05987978, "balance_loss_mlp": 1.01947927, "epoch": 0.3982444537966693, "flos": 24565128152160.0, "grad_norm": 1.703140499856983, "language_loss": 0.85824358, "learning_rate": 2.7384793251981244e-06, "loss": 0.88019192, "num_input_tokens_seen": 71382145, "step": 3312, "time_per_iteration": 2.5599803924560547 }, { "auxiliary_loss_clip": 0.01183215, "auxiliary_loss_mlp": 0.01032372, "balance_loss_clip": 1.05535483, "balance_loss_mlp": 1.02434921, "epoch": 0.39836469668730834, "flos": 26213860145280.0, "grad_norm": 1.7623154492956508, "language_loss": 0.80495387, "learning_rate": 2.737755344203571e-06, "loss": 0.82710975, "num_input_tokens_seen": 71402095, "step": 3313, "time_per_iteration": 2.5812995433807373 }, { "auxiliary_loss_clip": 0.01181924, "auxiliary_loss_mlp": 0.01031617, "balance_loss_clip": 1.0584408, "balance_loss_mlp": 1.02376664, "epoch": 0.39848493957794745, "flos": 27636138118560.0, "grad_norm": 1.5951792683522326, "language_loss": 0.79500157, "learning_rate": 2.7370312512921186e-06, "loss": 0.817137, "num_input_tokens_seen": 71423875, "step": 3314, "time_per_iteration": 3.3269622325897217 }, { "auxiliary_loss_clip": 0.01167641, "auxiliary_loss_mlp": 0.01030014, "balance_loss_clip": 1.05364656, "balance_loss_mlp": 1.02146053, "epoch": 0.39860518246858656, "flos": 12239329638720.0, "grad_norm": 3.016741401213545, "language_loss": 0.76601112, "learning_rate": 2.736307046573611e-06, "loss": 0.78798771, "num_input_tokens_seen": 71439745, "step": 3315, "time_per_iteration": 2.570486068725586 }, { "auxiliary_loss_clip": 0.01190979, "auxiliary_loss_mlp": 0.01031384, "balance_loss_clip": 1.05613804, "balance_loss_mlp": 1.02424359, "epoch": 0.3987254253592256, "flos": 22379138345280.0, "grad_norm": 1.970511796137028, "language_loss": 0.81810629, "learning_rate": 2.73558273015791e-06, "loss": 0.84032995, "num_input_tokens_seen": 71459575, "step": 3316, "time_per_iteration": 3.2417306900024414 }, { "auxiliary_loss_clip": 0.01195495, "auxiliary_loss_mlp": 0.01029684, "balance_loss_clip": 1.05907691, "balance_loss_mlp": 1.02147675, "epoch": 0.3988456682498647, "flos": 23514023994240.0, "grad_norm": 2.184981910868552, "language_loss": 0.70477581, "learning_rate": 2.734858302154894e-06, "loss": 0.72702754, "num_input_tokens_seen": 71481075, "step": 3317, "time_per_iteration": 2.5188956260681152 }, { "auxiliary_loss_clip": 0.01161594, "auxiliary_loss_mlp": 0.01030644, "balance_loss_clip": 1.05609226, "balance_loss_mlp": 1.02248979, "epoch": 0.39896591114050384, "flos": 19208765315520.0, "grad_norm": 2.0252673325143604, "language_loss": 0.75905585, "learning_rate": 2.734133762674457e-06, "loss": 0.7809782, "num_input_tokens_seen": 71500665, "step": 3318, "time_per_iteration": 2.5375590324401855 }, { "auxiliary_loss_clip": 0.0116551, "auxiliary_loss_mlp": 0.01031798, "balance_loss_clip": 1.05584848, "balance_loss_mlp": 1.02207005, "epoch": 0.3990861540311429, "flos": 28401035211840.0, "grad_norm": 1.965583675899709, "language_loss": 0.71139848, "learning_rate": 2.7334091118265124e-06, "loss": 0.7333715, "num_input_tokens_seen": 71522560, "step": 3319, "time_per_iteration": 3.369403123855591 }, { "auxiliary_loss_clip": 0.01075119, "auxiliary_loss_mlp": 0.01007818, "balance_loss_clip": 1.01785088, "balance_loss_mlp": 1.00644135, "epoch": 0.399206396921782, "flos": 61758574570560.0, "grad_norm": 0.6792167820640709, "language_loss": 0.57852578, "learning_rate": 2.732684349720989e-06, "loss": 0.59935516, "num_input_tokens_seen": 71590520, "step": 3320, "time_per_iteration": 3.11625599861145 }, { "auxiliary_loss_clip": 0.01149632, "auxiliary_loss_mlp": 0.01032825, "balance_loss_clip": 1.05173576, "balance_loss_mlp": 1.02436733, "epoch": 0.3993266398124211, "flos": 28074571623360.0, "grad_norm": 1.593974037266874, "language_loss": 0.75401282, "learning_rate": 2.7319594764678318e-06, "loss": 0.7758373, "num_input_tokens_seen": 71612620, "step": 3321, "time_per_iteration": 3.370532274246216 }, { "auxiliary_loss_clip": 0.01138639, "auxiliary_loss_mlp": 0.01031234, "balance_loss_clip": 1.05370069, "balance_loss_mlp": 1.0226512, "epoch": 0.39944688270306017, "flos": 23225087240640.0, "grad_norm": 1.8302552809875983, "language_loss": 0.83700955, "learning_rate": 2.7312344921770044e-06, "loss": 0.85870832, "num_input_tokens_seen": 71634320, "step": 3322, "time_per_iteration": 2.636488437652588 }, { "auxiliary_loss_clip": 0.01161058, "auxiliary_loss_mlp": 0.01030498, "balance_loss_clip": 1.05012369, "balance_loss_mlp": 1.02257657, "epoch": 0.3995671255936993, "flos": 19390437274560.0, "grad_norm": 1.8322000609440907, "language_loss": 0.78365302, "learning_rate": 2.7305093969584857e-06, "loss": 0.80556858, "num_input_tokens_seen": 71653145, "step": 3323, "time_per_iteration": 2.5927062034606934 }, { "auxiliary_loss_clip": 0.0117037, "auxiliary_loss_mlp": 0.01036578, "balance_loss_clip": 1.05244184, "balance_loss_mlp": 1.02852583, "epoch": 0.3996873684843384, "flos": 23842642600320.0, "grad_norm": 2.1947565606142607, "language_loss": 0.79941481, "learning_rate": 2.729784190922272e-06, "loss": 0.82148433, "num_input_tokens_seen": 71674580, "step": 3324, "time_per_iteration": 2.5275168418884277 }, { "auxiliary_loss_clip": 0.01059097, "auxiliary_loss_mlp": 0.01004662, "balance_loss_clip": 1.0176686, "balance_loss_mlp": 1.00323129, "epoch": 0.39980761137497745, "flos": 66576889420320.0, "grad_norm": 0.9424755911493258, "language_loss": 0.57149279, "learning_rate": 2.729058874178378e-06, "loss": 0.5921303, "num_input_tokens_seen": 71745260, "step": 3325, "time_per_iteration": 3.239025592803955 }, { "auxiliary_loss_clip": 0.01168317, "auxiliary_loss_mlp": 0.01029998, "balance_loss_clip": 1.05621827, "balance_loss_mlp": 1.02185571, "epoch": 0.39992785426561656, "flos": 28549167118560.0, "grad_norm": 2.1356387792574316, "language_loss": 0.69044, "learning_rate": 2.7283334468368315e-06, "loss": 0.71242321, "num_input_tokens_seen": 71766540, "step": 3326, "time_per_iteration": 2.5946571826934814 }, { "auxiliary_loss_clip": 0.01084102, "auxiliary_loss_mlp": 0.01031032, "balance_loss_clip": 1.03971493, "balance_loss_mlp": 1.02267218, "epoch": 0.4000480971562556, "flos": 15049411692000.0, "grad_norm": 1.7945581073330057, "language_loss": 0.73108053, "learning_rate": 2.72760790900768e-06, "loss": 0.7522319, "num_input_tokens_seen": 71783125, "step": 3327, "time_per_iteration": 2.9460930824279785 }, { "auxiliary_loss_clip": 0.01194153, "auxiliary_loss_mlp": 0.01031655, "balance_loss_clip": 1.05804574, "balance_loss_mlp": 1.02328002, "epoch": 0.4001683400468947, "flos": 23915613086400.0, "grad_norm": 1.7689428498088282, "language_loss": 0.78741884, "learning_rate": 2.7268822608009875e-06, "loss": 0.80967689, "num_input_tokens_seen": 71802500, "step": 3328, "time_per_iteration": 2.6517419815063477 }, { "auxiliary_loss_clip": 0.01154494, "auxiliary_loss_mlp": 0.0103282, "balance_loss_clip": 1.05351615, "balance_loss_mlp": 1.02433777, "epoch": 0.40028858293753383, "flos": 24352681746720.0, "grad_norm": 1.9437222368191491, "language_loss": 0.78510898, "learning_rate": 2.726156502326834e-06, "loss": 0.8069821, "num_input_tokens_seen": 71823800, "step": 3329, "time_per_iteration": 2.6332528591156006 }, { "auxiliary_loss_clip": 0.01031703, "auxiliary_loss_mlp": 0.01012332, "balance_loss_clip": 1.01905537, "balance_loss_mlp": 1.01112843, "epoch": 0.4004088258281729, "flos": 66787037140320.0, "grad_norm": 0.6993078084479494, "language_loss": 0.60209608, "learning_rate": 2.725430633695316e-06, "loss": 0.62253642, "num_input_tokens_seen": 71886880, "step": 3330, "time_per_iteration": 3.260667085647583 }, { "auxiliary_loss_clip": 0.01082787, "auxiliary_loss_mlp": 0.01000751, "balance_loss_clip": 1.01665807, "balance_loss_mlp": 0.99944001, "epoch": 0.400529068718812, "flos": 58598395569120.0, "grad_norm": 0.8892346416558196, "language_loss": 0.57951856, "learning_rate": 2.7247046550165485e-06, "loss": 0.60035384, "num_input_tokens_seen": 71939005, "step": 3331, "time_per_iteration": 2.8995442390441895 }, { "auxiliary_loss_clip": 0.01196399, "auxiliary_loss_mlp": 0.01034407, "balance_loss_clip": 1.06091511, "balance_loss_mlp": 1.02658105, "epoch": 0.4006493116094511, "flos": 25377465161280.0, "grad_norm": 1.9726450416880679, "language_loss": 0.75828373, "learning_rate": 2.7239785664006606e-06, "loss": 0.78059173, "num_input_tokens_seen": 71962545, "step": 3332, "time_per_iteration": 2.561753988265991 }, { "auxiliary_loss_clip": 0.01075005, "auxiliary_loss_mlp": 0.01001947, "balance_loss_clip": 1.01740265, "balance_loss_mlp": 1.00062943, "epoch": 0.40076955450009016, "flos": 60280739448480.0, "grad_norm": 0.7709767604884946, "language_loss": 0.61757898, "learning_rate": 2.7232523679578002e-06, "loss": 0.63834846, "num_input_tokens_seen": 72025625, "step": 3333, "time_per_iteration": 3.1675004959106445 }, { "auxiliary_loss_clip": 0.01176952, "auxiliary_loss_mlp": 0.0103401, "balance_loss_clip": 1.05657077, "balance_loss_mlp": 1.02650535, "epoch": 0.4008897973907293, "flos": 16617271467840.0, "grad_norm": 2.3391066363064734, "language_loss": 0.78918254, "learning_rate": 2.7225260597981295e-06, "loss": 0.81129223, "num_input_tokens_seen": 72043330, "step": 3334, "time_per_iteration": 2.497370719909668 }, { "auxiliary_loss_clip": 0.01147929, "auxiliary_loss_mlp": 0.00763448, "balance_loss_clip": 1.05264664, "balance_loss_mlp": 1.00030804, "epoch": 0.4010100402813684, "flos": 15377347875840.0, "grad_norm": 2.480773029965237, "language_loss": 0.7853508, "learning_rate": 2.721799642031831e-06, "loss": 0.80446458, "num_input_tokens_seen": 72059500, "step": 3335, "time_per_iteration": 2.5953030586242676 }, { "auxiliary_loss_clip": 0.01167984, "auxiliary_loss_mlp": 0.01026638, "balance_loss_clip": 1.05274892, "balance_loss_mlp": 1.01859164, "epoch": 0.40113028317200744, "flos": 13298838365280.0, "grad_norm": 1.77431774693712, "language_loss": 0.77433181, "learning_rate": 2.721073114769101e-06, "loss": 0.796278, "num_input_tokens_seen": 72077175, "step": 3336, "time_per_iteration": 2.556497573852539 }, { "auxiliary_loss_clip": 0.01147563, "auxiliary_loss_mlp": 0.01032179, "balance_loss_clip": 1.05413246, "balance_loss_mlp": 1.02423382, "epoch": 0.40125052606264655, "flos": 20668031369280.0, "grad_norm": 2.014474047004553, "language_loss": 0.75256646, "learning_rate": 2.7203464781201523e-06, "loss": 0.77436393, "num_input_tokens_seen": 72096490, "step": 3337, "time_per_iteration": 2.625824451446533 }, { "auxiliary_loss_clip": 0.01195865, "auxiliary_loss_mlp": 0.01030206, "balance_loss_clip": 1.05979848, "balance_loss_mlp": 1.02261209, "epoch": 0.40137076895328566, "flos": 24607683361440.0, "grad_norm": 1.8874169682971276, "language_loss": 0.77654123, "learning_rate": 2.719619732195215e-06, "loss": 0.7988019, "num_input_tokens_seen": 72118130, "step": 3338, "time_per_iteration": 2.5485501289367676 }, { "auxiliary_loss_clip": 0.01149619, "auxiliary_loss_mlp": 0.01031047, "balance_loss_clip": 1.05293059, "balance_loss_mlp": 1.02277958, "epoch": 0.4014910118439247, "flos": 24206597106720.0, "grad_norm": 1.5585354541465835, "language_loss": 0.72748131, "learning_rate": 2.7188928771045377e-06, "loss": 0.74928796, "num_input_tokens_seen": 72139450, "step": 3339, "time_per_iteration": 2.6425819396972656 }, { "auxiliary_loss_clip": 0.01143697, "auxiliary_loss_mlp": 0.01032275, "balance_loss_clip": 1.05146837, "balance_loss_mlp": 1.02400804, "epoch": 0.4016112547345638, "flos": 26725084551360.0, "grad_norm": 1.6032859053545019, "language_loss": 0.79748553, "learning_rate": 2.7181659129583815e-06, "loss": 0.81924522, "num_input_tokens_seen": 72159040, "step": 3340, "time_per_iteration": 3.645031452178955 }, { "auxiliary_loss_clip": 0.01151717, "auxiliary_loss_mlp": 0.01026264, "balance_loss_clip": 1.04810119, "balance_loss_mlp": 1.01791382, "epoch": 0.4017314976252029, "flos": 21288029082240.0, "grad_norm": 1.7345474781863897, "language_loss": 0.75852877, "learning_rate": 2.7174388398670276e-06, "loss": 0.7803086, "num_input_tokens_seen": 72178220, "step": 3341, "time_per_iteration": 2.5994713306427 }, { "auxiliary_loss_clip": 0.01194044, "auxiliary_loss_mlp": 0.01027395, "balance_loss_clip": 1.05643582, "balance_loss_mlp": 1.0185678, "epoch": 0.401851740515842, "flos": 25484694038880.0, "grad_norm": 2.0786317012190128, "language_loss": 0.92406642, "learning_rate": 2.716711657940773e-06, "loss": 0.94628084, "num_input_tokens_seen": 72199230, "step": 3342, "time_per_iteration": 3.717069387435913 }, { "auxiliary_loss_clip": 0.01049247, "auxiliary_loss_mlp": 0.01003033, "balance_loss_clip": 1.01505291, "balance_loss_mlp": 1.00179315, "epoch": 0.4019719834064811, "flos": 55395344494080.0, "grad_norm": 0.8108431580087828, "language_loss": 0.56524503, "learning_rate": 2.7159843672899284e-06, "loss": 0.58576787, "num_input_tokens_seen": 72263430, "step": 3343, "time_per_iteration": 3.326150894165039 }, { "auxiliary_loss_clip": 0.01180916, "auxiliary_loss_mlp": 0.01034541, "balance_loss_clip": 1.05810559, "balance_loss_mlp": 1.02572572, "epoch": 0.40209222629712016, "flos": 18180102869280.0, "grad_norm": 1.851762799575047, "language_loss": 0.81349099, "learning_rate": 2.715256968024825e-06, "loss": 0.83564556, "num_input_tokens_seen": 72280505, "step": 3344, "time_per_iteration": 2.550163984298706 }, { "auxiliary_loss_clip": 0.01172912, "auxiliary_loss_mlp": 0.01028018, "balance_loss_clip": 1.05671895, "balance_loss_mlp": 1.02023935, "epoch": 0.40221246918775927, "flos": 25961013548160.0, "grad_norm": 1.4601887311379977, "language_loss": 0.82168984, "learning_rate": 2.7145294602558083e-06, "loss": 0.8436991, "num_input_tokens_seen": 72301215, "step": 3345, "time_per_iteration": 3.5662429332733154 }, { "auxiliary_loss_clip": 0.01181572, "auxiliary_loss_mlp": 0.01029231, "balance_loss_clip": 1.05848563, "balance_loss_mlp": 1.02034974, "epoch": 0.4023327120783984, "flos": 33838916771040.0, "grad_norm": 2.028524273008681, "language_loss": 0.7068004, "learning_rate": 2.713801844093241e-06, "loss": 0.72890842, "num_input_tokens_seen": 72322365, "step": 3346, "time_per_iteration": 2.6283130645751953 }, { "auxiliary_loss_clip": 0.01178221, "auxiliary_loss_mlp": 0.01022326, "balance_loss_clip": 1.05472994, "balance_loss_mlp": 1.01457119, "epoch": 0.40245295496903744, "flos": 26900255540640.0, "grad_norm": 2.3186397921037023, "language_loss": 0.87997842, "learning_rate": 2.7130741196475014e-06, "loss": 0.90198392, "num_input_tokens_seen": 72340495, "step": 3347, "time_per_iteration": 3.3007049560546875 }, { "auxiliary_loss_clip": 0.01168481, "auxiliary_loss_mlp": 0.0102647, "balance_loss_clip": 1.05747497, "balance_loss_mlp": 1.01807153, "epoch": 0.40257319785967655, "flos": 36902743345440.0, "grad_norm": 2.4895263209189795, "language_loss": 0.79337186, "learning_rate": 2.7123462870289848e-06, "loss": 0.81532133, "num_input_tokens_seen": 72360545, "step": 3348, "time_per_iteration": 2.6597976684570312 }, { "auxiliary_loss_clip": 0.0116539, "auxiliary_loss_mlp": 0.01027203, "balance_loss_clip": 1.05167675, "balance_loss_mlp": 1.01890564, "epoch": 0.40269344075031566, "flos": 24353184584160.0, "grad_norm": 1.6065995098473347, "language_loss": 0.81414378, "learning_rate": 2.711618346348102e-06, "loss": 0.83606964, "num_input_tokens_seen": 72381070, "step": 3349, "time_per_iteration": 2.5944254398345947 }, { "auxiliary_loss_clip": 0.01157107, "auxiliary_loss_mlp": 0.01032954, "balance_loss_clip": 1.05217934, "balance_loss_mlp": 1.0255034, "epoch": 0.4028136836409547, "flos": 14389660292640.0, "grad_norm": 1.8279384137835597, "language_loss": 0.63900417, "learning_rate": 2.7108902977152825e-06, "loss": 0.66090477, "num_input_tokens_seen": 72398970, "step": 3350, "time_per_iteration": 2.5327627658843994 }, { "auxiliary_loss_clip": 0.01177991, "auxiliary_loss_mlp": 0.01032464, "balance_loss_clip": 1.05539846, "balance_loss_mlp": 1.023947, "epoch": 0.4029339265315938, "flos": 26136040869600.0, "grad_norm": 2.0458981097103184, "language_loss": 0.74595791, "learning_rate": 2.7101621412409704e-06, "loss": 0.76806247, "num_input_tokens_seen": 72418455, "step": 3351, "time_per_iteration": 2.580904483795166 }, { "auxiliary_loss_clip": 0.01194218, "auxiliary_loss_mlp": 0.01030169, "balance_loss_clip": 1.05710721, "balance_loss_mlp": 1.02225339, "epoch": 0.40305416942223293, "flos": 23256328607520.0, "grad_norm": 1.7968609045125365, "language_loss": 0.85456914, "learning_rate": 2.7094338770356256e-06, "loss": 0.87681299, "num_input_tokens_seen": 72437540, "step": 3352, "time_per_iteration": 2.4947903156280518 }, { "auxiliary_loss_clip": 0.01161207, "auxiliary_loss_mlp": 0.01033394, "balance_loss_clip": 1.05349851, "balance_loss_mlp": 1.02504349, "epoch": 0.403174412312872, "flos": 27089649646080.0, "grad_norm": 1.9867958043547318, "language_loss": 0.63716787, "learning_rate": 2.708705505209726e-06, "loss": 0.65911382, "num_input_tokens_seen": 72458315, "step": 3353, "time_per_iteration": 2.6020257472991943 }, { "auxiliary_loss_clip": 0.01125035, "auxiliary_loss_mlp": 0.01041023, "balance_loss_clip": 1.04641354, "balance_loss_mlp": 1.03323841, "epoch": 0.4032946552035111, "flos": 21756339109440.0, "grad_norm": 1.854665146009632, "language_loss": 0.91720206, "learning_rate": 2.7079770258737646e-06, "loss": 0.93886262, "num_input_tokens_seen": 72476225, "step": 3354, "time_per_iteration": 2.613300085067749 }, { "auxiliary_loss_clip": 0.01146617, "auxiliary_loss_mlp": 0.01032334, "balance_loss_clip": 1.05042791, "balance_loss_mlp": 1.02373326, "epoch": 0.4034148980941502, "flos": 17343959304000.0, "grad_norm": 2.046306459702466, "language_loss": 0.75052506, "learning_rate": 2.707248439138251e-06, "loss": 0.77231461, "num_input_tokens_seen": 72492460, "step": 3355, "time_per_iteration": 2.5692970752716064 }, { "auxiliary_loss_clip": 0.01161592, "auxiliary_loss_mlp": 0.01025809, "balance_loss_clip": 1.05733871, "balance_loss_mlp": 1.01789308, "epoch": 0.40353514098478926, "flos": 22017841693920.0, "grad_norm": 1.6848704507900103, "language_loss": 0.65251672, "learning_rate": 2.7065197451137114e-06, "loss": 0.67439073, "num_input_tokens_seen": 72513840, "step": 3356, "time_per_iteration": 2.609208345413208 }, { "auxiliary_loss_clip": 0.01161261, "auxiliary_loss_mlp": 0.01029457, "balance_loss_clip": 1.05386901, "balance_loss_mlp": 1.02102292, "epoch": 0.4036553838754284, "flos": 14246449009440.0, "grad_norm": 1.9229733912522153, "language_loss": 0.6752317, "learning_rate": 2.7057909439106894e-06, "loss": 0.69713885, "num_input_tokens_seen": 72531695, "step": 3357, "time_per_iteration": 2.538396120071411 }, { "auxiliary_loss_clip": 0.01168741, "auxiliary_loss_mlp": 0.00762686, "balance_loss_clip": 1.05254066, "balance_loss_mlp": 1.00019407, "epoch": 0.40377562676606743, "flos": 24790648331040.0, "grad_norm": 1.8037224102403104, "language_loss": 0.783319, "learning_rate": 2.7050620356397417e-06, "loss": 0.80263329, "num_input_tokens_seen": 72550645, "step": 3358, "time_per_iteration": 2.58491849899292 }, { "auxiliary_loss_clip": 0.01192421, "auxiliary_loss_mlp": 0.01032081, "balance_loss_clip": 1.05924749, "balance_loss_mlp": 1.02494037, "epoch": 0.40389586965670654, "flos": 24061230805920.0, "grad_norm": 1.7959430191115775, "language_loss": 0.72130245, "learning_rate": 2.7043330204114437e-06, "loss": 0.74354744, "num_input_tokens_seen": 72569355, "step": 3359, "time_per_iteration": 2.5303690433502197 }, { "auxiliary_loss_clip": 0.01188139, "auxiliary_loss_mlp": 0.010269, "balance_loss_clip": 1.05554342, "balance_loss_mlp": 1.01918995, "epoch": 0.40401611254734565, "flos": 16399617103200.0, "grad_norm": 1.9499832100457644, "language_loss": 0.85918534, "learning_rate": 2.7036038983363862e-06, "loss": 0.88133574, "num_input_tokens_seen": 72585960, "step": 3360, "time_per_iteration": 2.4691848754882812 }, { "auxiliary_loss_clip": 0.0117226, "auxiliary_loss_mlp": 0.0102647, "balance_loss_clip": 1.05517697, "balance_loss_mlp": 1.01834607, "epoch": 0.4041363554379847, "flos": 23988224402880.0, "grad_norm": 1.725616268691513, "language_loss": 0.83773857, "learning_rate": 2.702874669525177e-06, "loss": 0.85972589, "num_input_tokens_seen": 72604440, "step": 3361, "time_per_iteration": 2.5914032459259033 }, { "auxiliary_loss_clip": 0.01146908, "auxiliary_loss_mlp": 0.01028702, "balance_loss_clip": 1.05297875, "balance_loss_mlp": 1.0214777, "epoch": 0.4042565983286238, "flos": 28401969052800.0, "grad_norm": 1.8208108717875042, "language_loss": 0.69393528, "learning_rate": 2.7021453340884394e-06, "loss": 0.71569133, "num_input_tokens_seen": 72622165, "step": 3362, "time_per_iteration": 2.626551866531372 }, { "auxiliary_loss_clip": 0.01152241, "auxiliary_loss_mlp": 0.00763188, "balance_loss_clip": 1.05266905, "balance_loss_mlp": 1.000211, "epoch": 0.40437684121926293, "flos": 17710966752000.0, "grad_norm": 2.4596348155547294, "language_loss": 0.73124921, "learning_rate": 2.7014158921368125e-06, "loss": 0.75040352, "num_input_tokens_seen": 72640490, "step": 3363, "time_per_iteration": 2.5520596504211426 }, { "auxiliary_loss_clip": 0.01191857, "auxiliary_loss_mlp": 0.01031636, "balance_loss_clip": 1.05798888, "balance_loss_mlp": 1.0242691, "epoch": 0.404497084109902, "flos": 24018962932320.0, "grad_norm": 2.072655321074938, "language_loss": 0.85507101, "learning_rate": 2.700686343780953e-06, "loss": 0.87730587, "num_input_tokens_seen": 72660360, "step": 3364, "time_per_iteration": 2.5158755779266357 }, { "auxiliary_loss_clip": 0.01165234, "auxiliary_loss_mlp": 0.01030894, "balance_loss_clip": 1.05476713, "balance_loss_mlp": 1.0233779, "epoch": 0.4046173270005411, "flos": 22929865019040.0, "grad_norm": 1.6155851181520624, "language_loss": 0.88239896, "learning_rate": 2.699956689131532e-06, "loss": 0.90436023, "num_input_tokens_seen": 72680345, "step": 3365, "time_per_iteration": 2.6076934337615967 }, { "auxiliary_loss_clip": 0.01166918, "auxiliary_loss_mlp": 0.01029042, "balance_loss_clip": 1.05617213, "balance_loss_mlp": 1.02116513, "epoch": 0.4047375698911802, "flos": 20668139120160.0, "grad_norm": 2.094580655306568, "language_loss": 0.85133994, "learning_rate": 2.699226928299238e-06, "loss": 0.87329948, "num_input_tokens_seen": 72698365, "step": 3366, "time_per_iteration": 3.3967347145080566 }, { "auxiliary_loss_clip": 0.01177731, "auxiliary_loss_mlp": 0.01030098, "balance_loss_clip": 1.05550122, "balance_loss_mlp": 1.02246308, "epoch": 0.40485781278181926, "flos": 28912870206240.0, "grad_norm": 2.298148147563736, "language_loss": 0.78945029, "learning_rate": 2.698497061394774e-06, "loss": 0.81152862, "num_input_tokens_seen": 72716850, "step": 3367, "time_per_iteration": 2.6013917922973633 }, { "auxiliary_loss_clip": 0.01156686, "auxiliary_loss_mlp": 0.00762897, "balance_loss_clip": 1.05715561, "balance_loss_mlp": 1.00032663, "epoch": 0.40497805567245837, "flos": 23148381390720.0, "grad_norm": 1.7000548478870936, "language_loss": 0.80862689, "learning_rate": 2.6977670885288627e-06, "loss": 0.8278228, "num_input_tokens_seen": 72738250, "step": 3368, "time_per_iteration": 3.428927421569824 }, { "auxiliary_loss_clip": 0.0115164, "auxiliary_loss_mlp": 0.01033413, "balance_loss_clip": 1.05027056, "balance_loss_mlp": 1.02591491, "epoch": 0.4050982985630975, "flos": 16289407117920.0, "grad_norm": 1.7167205586111378, "language_loss": 0.75059819, "learning_rate": 2.6970370098122378e-06, "loss": 0.77244866, "num_input_tokens_seen": 72755235, "step": 3369, "time_per_iteration": 2.5519790649414062 }, { "auxiliary_loss_clip": 0.01193671, "auxiliary_loss_mlp": 0.01030252, "balance_loss_clip": 1.0585463, "balance_loss_mlp": 1.02210963, "epoch": 0.40521854145373654, "flos": 34459489155360.0, "grad_norm": 1.8353887397956994, "language_loss": 0.86569798, "learning_rate": 2.6963068253556535e-06, "loss": 0.88793725, "num_input_tokens_seen": 72776620, "step": 3370, "time_per_iteration": 2.597337245941162 }, { "auxiliary_loss_clip": 0.01181961, "auxiliary_loss_mlp": 0.0102558, "balance_loss_clip": 1.05439818, "balance_loss_mlp": 1.01680017, "epoch": 0.40533878434437565, "flos": 25331102754240.0, "grad_norm": 2.2240541663832647, "language_loss": 0.85563159, "learning_rate": 2.6955765352698763e-06, "loss": 0.87770694, "num_input_tokens_seen": 72796765, "step": 3371, "time_per_iteration": 3.3221049308776855 }, { "auxiliary_loss_clip": 0.01195262, "auxiliary_loss_mlp": 0.01029352, "balance_loss_clip": 1.05790675, "balance_loss_mlp": 1.02097154, "epoch": 0.40545902723501476, "flos": 15012064441920.0, "grad_norm": 2.242138065094389, "language_loss": 0.73215151, "learning_rate": 2.6948461396656923e-06, "loss": 0.75439769, "num_input_tokens_seen": 72814175, "step": 3372, "time_per_iteration": 2.4580328464508057 }, { "auxiliary_loss_clip": 0.01183029, "auxiliary_loss_mlp": 0.0103381, "balance_loss_clip": 1.05724275, "balance_loss_mlp": 1.02569473, "epoch": 0.4055792701256538, "flos": 25521107448000.0, "grad_norm": 4.162109936714176, "language_loss": 0.74931955, "learning_rate": 2.6941156386539013e-06, "loss": 0.77148795, "num_input_tokens_seen": 72834125, "step": 3373, "time_per_iteration": 3.3427255153656006 }, { "auxiliary_loss_clip": 0.01159553, "auxiliary_loss_mlp": 0.01031341, "balance_loss_clip": 1.05651748, "balance_loss_mlp": 1.02325904, "epoch": 0.4056995130162929, "flos": 19574587503840.0, "grad_norm": 2.1119907478450273, "language_loss": 0.80859786, "learning_rate": 2.6933850323453203e-06, "loss": 0.8305068, "num_input_tokens_seen": 72852570, "step": 3374, "time_per_iteration": 2.562319278717041 }, { "auxiliary_loss_clip": 0.01191815, "auxiliary_loss_mlp": 0.0102739, "balance_loss_clip": 1.05907869, "balance_loss_mlp": 1.0201745, "epoch": 0.405819755906932, "flos": 15413797201920.0, "grad_norm": 1.7222866986971521, "language_loss": 0.7461201, "learning_rate": 2.6926543208507806e-06, "loss": 0.7683121, "num_input_tokens_seen": 72871250, "step": 3375, "time_per_iteration": 2.48641300201416 }, { "auxiliary_loss_clip": 0.01176679, "auxiliary_loss_mlp": 0.01027723, "balance_loss_clip": 1.05617189, "balance_loss_mlp": 1.01940823, "epoch": 0.4059399987975711, "flos": 21433934137440.0, "grad_norm": 2.1469981968447347, "language_loss": 0.80309618, "learning_rate": 2.6919235042811316e-06, "loss": 0.82514018, "num_input_tokens_seen": 72890035, "step": 3376, "time_per_iteration": 2.49263072013855 }, { "auxiliary_loss_clip": 0.01149702, "auxiliary_loss_mlp": 0.01029773, "balance_loss_clip": 1.05608225, "balance_loss_mlp": 1.02155375, "epoch": 0.4060602416882102, "flos": 25556946185760.0, "grad_norm": 1.9840491382372816, "language_loss": 0.76231581, "learning_rate": 2.691192582747237e-06, "loss": 0.78411055, "num_input_tokens_seen": 72909665, "step": 3377, "time_per_iteration": 2.6428961753845215 }, { "auxiliary_loss_clip": 0.01193192, "auxiliary_loss_mlp": 0.01033496, "balance_loss_clip": 1.05876243, "balance_loss_mlp": 1.02546084, "epoch": 0.40618048457884925, "flos": 23766762840480.0, "grad_norm": 1.6837053462001632, "language_loss": 0.73787141, "learning_rate": 2.6904615563599765e-06, "loss": 0.76013827, "num_input_tokens_seen": 72929465, "step": 3378, "time_per_iteration": 2.496948719024658 }, { "auxiliary_loss_clip": 0.01144484, "auxiliary_loss_mlp": 0.01035397, "balance_loss_clip": 1.05194509, "balance_loss_mlp": 1.0277195, "epoch": 0.40630072746948837, "flos": 17639684363040.0, "grad_norm": 1.711650044627517, "language_loss": 0.83213925, "learning_rate": 2.6897304252302477e-06, "loss": 0.8539381, "num_input_tokens_seen": 72946785, "step": 3379, "time_per_iteration": 2.625251054763794 }, { "auxiliary_loss_clip": 0.01043759, "auxiliary_loss_mlp": 0.01004901, "balance_loss_clip": 1.01420403, "balance_loss_mlp": 1.0035603, "epoch": 0.4064209703601275, "flos": 60836063493120.0, "grad_norm": 0.7874962923868052, "language_loss": 0.54805267, "learning_rate": 2.688999189468962e-06, "loss": 0.56853926, "num_input_tokens_seen": 73003215, "step": 3380, "time_per_iteration": 3.036417245864868 }, { "auxiliary_loss_clip": 0.01175306, "auxiliary_loss_mlp": 0.01025557, "balance_loss_clip": 1.05665863, "balance_loss_mlp": 1.0172838, "epoch": 0.40654121325076653, "flos": 24024350476320.0, "grad_norm": 2.3063765376566554, "language_loss": 0.7588954, "learning_rate": 2.6882678491870464e-06, "loss": 0.78090405, "num_input_tokens_seen": 73023650, "step": 3381, "time_per_iteration": 2.5384576320648193 }, { "auxiliary_loss_clip": 0.01183402, "auxiliary_loss_mlp": 0.01025951, "balance_loss_clip": 1.05858564, "balance_loss_mlp": 1.01799428, "epoch": 0.40666145614140564, "flos": 27344256174240.0, "grad_norm": 1.5569377533580344, "language_loss": 0.71287125, "learning_rate": 2.6875364044954453e-06, "loss": 0.73496485, "num_input_tokens_seen": 73043880, "step": 3382, "time_per_iteration": 2.5622663497924805 }, { "auxiliary_loss_clip": 0.0115675, "auxiliary_loss_mlp": 0.01032384, "balance_loss_clip": 1.04810667, "balance_loss_mlp": 1.02508807, "epoch": 0.40678169903204475, "flos": 26176728396960.0, "grad_norm": 1.6104168256527245, "language_loss": 0.82468468, "learning_rate": 2.6868048555051185e-06, "loss": 0.84657604, "num_input_tokens_seen": 73065410, "step": 3383, "time_per_iteration": 2.599520444869995 }, { "auxiliary_loss_clip": 0.01168228, "auxiliary_loss_mlp": 0.01032601, "balance_loss_clip": 1.05244505, "balance_loss_mlp": 1.02543402, "epoch": 0.4069019419226838, "flos": 28622424940320.0, "grad_norm": 2.308222241203406, "language_loss": 0.85789454, "learning_rate": 2.686073202327041e-06, "loss": 0.87990284, "num_input_tokens_seen": 73084410, "step": 3384, "time_per_iteration": 2.6179070472717285 }, { "auxiliary_loss_clip": 0.0114817, "auxiliary_loss_mlp": 0.01027083, "balance_loss_clip": 1.04722416, "balance_loss_mlp": 1.01976371, "epoch": 0.4070221848133229, "flos": 25229010001920.0, "grad_norm": 1.6483092459224493, "language_loss": 0.73315859, "learning_rate": 2.6853414450722043e-06, "loss": 0.75491112, "num_input_tokens_seen": 73104075, "step": 3385, "time_per_iteration": 2.591275453567505 }, { "auxiliary_loss_clip": 0.01174976, "auxiliary_loss_mlp": 0.01029232, "balance_loss_clip": 1.05598855, "balance_loss_mlp": 1.02199292, "epoch": 0.40714242770396203, "flos": 18405227961600.0, "grad_norm": 1.613619145022257, "language_loss": 0.85052514, "learning_rate": 2.684609583851616e-06, "loss": 0.87256718, "num_input_tokens_seen": 73122250, "step": 3386, "time_per_iteration": 2.5101559162139893 }, { "auxiliary_loss_clip": 0.0113147, "auxiliary_loss_mlp": 0.01028625, "balance_loss_clip": 1.05009413, "balance_loss_mlp": 1.02065635, "epoch": 0.4072626705946011, "flos": 30228924976800.0, "grad_norm": 1.7632646532571936, "language_loss": 0.80642629, "learning_rate": 2.683877618776297e-06, "loss": 0.82802725, "num_input_tokens_seen": 73144505, "step": 3387, "time_per_iteration": 2.698071241378784 }, { "auxiliary_loss_clip": 0.01155037, "auxiliary_loss_mlp": 0.01035624, "balance_loss_clip": 1.04941928, "balance_loss_mlp": 1.02692175, "epoch": 0.4073829134852402, "flos": 21834553471680.0, "grad_norm": 3.9913029085506224, "language_loss": 0.7403881, "learning_rate": 2.6831455499572876e-06, "loss": 0.76229465, "num_input_tokens_seen": 73162440, "step": 3388, "time_per_iteration": 2.565216541290283 }, { "auxiliary_loss_clip": 0.01191076, "auxiliary_loss_mlp": 0.01025237, "balance_loss_clip": 1.05639911, "balance_loss_mlp": 1.01729178, "epoch": 0.40750315637587925, "flos": 25260215451840.0, "grad_norm": 1.8279413622089467, "language_loss": 0.77613902, "learning_rate": 2.682413377505641e-06, "loss": 0.79830217, "num_input_tokens_seen": 73181245, "step": 3389, "time_per_iteration": 2.5114195346832275 }, { "auxiliary_loss_clip": 0.01176383, "auxiliary_loss_mlp": 0.01030564, "balance_loss_clip": 1.05276382, "balance_loss_mlp": 1.02250576, "epoch": 0.40762339926651836, "flos": 19712770412640.0, "grad_norm": 2.717146853637624, "language_loss": 0.7683903, "learning_rate": 2.6816811015324284e-06, "loss": 0.79045975, "num_input_tokens_seen": 73199295, "step": 3390, "time_per_iteration": 2.5216052532196045 }, { "auxiliary_loss_clip": 0.01080527, "auxiliary_loss_mlp": 0.00999841, "balance_loss_clip": 1.01452708, "balance_loss_mlp": 0.99848843, "epoch": 0.40774364215715747, "flos": 71449318352160.0, "grad_norm": 0.726730920787265, "language_loss": 0.56726629, "learning_rate": 2.6809487221487343e-06, "loss": 0.58806998, "num_input_tokens_seen": 73258780, "step": 3391, "time_per_iteration": 3.0116214752197266 }, { "auxiliary_loss_clip": 0.01166069, "auxiliary_loss_mlp": 0.01031045, "balance_loss_clip": 1.05077696, "balance_loss_mlp": 1.02306414, "epoch": 0.4078638850477965, "flos": 15084137004000.0, "grad_norm": 2.5335951081902994, "language_loss": 0.8175571, "learning_rate": 2.6802162394656605e-06, "loss": 0.83952826, "num_input_tokens_seen": 73275490, "step": 3392, "time_per_iteration": 3.2686424255371094 }, { "auxiliary_loss_clip": 0.01158718, "auxiliary_loss_mlp": 0.01027009, "balance_loss_clip": 1.05117345, "balance_loss_mlp": 1.0191735, "epoch": 0.40798412793843564, "flos": 23842894019040.0, "grad_norm": 1.8287806762059504, "language_loss": 0.71946305, "learning_rate": 2.679483653594324e-06, "loss": 0.74132031, "num_input_tokens_seen": 73297260, "step": 3393, "time_per_iteration": 2.5887577533721924 }, { "auxiliary_loss_clip": 0.0117783, "auxiliary_loss_mlp": 0.01026885, "balance_loss_clip": 1.05370855, "balance_loss_mlp": 1.0192256, "epoch": 0.40810437082907475, "flos": 21065777346720.0, "grad_norm": 2.433480096395059, "language_loss": 0.766307, "learning_rate": 2.678750964645857e-06, "loss": 0.78835416, "num_input_tokens_seen": 73316340, "step": 3394, "time_per_iteration": 3.292724370956421 }, { "auxiliary_loss_clip": 0.01177041, "auxiliary_loss_mlp": 0.01029789, "balance_loss_clip": 1.05734897, "balance_loss_mlp": 1.02079487, "epoch": 0.4082246137197138, "flos": 11321380015200.0, "grad_norm": 2.4293081276361304, "language_loss": 0.83565712, "learning_rate": 2.6780181727314094e-06, "loss": 0.8577255, "num_input_tokens_seen": 73331245, "step": 3395, "time_per_iteration": 2.4750938415527344 }, { "auxiliary_loss_clip": 0.01151115, "auxiliary_loss_mlp": 0.00762559, "balance_loss_clip": 1.05221367, "balance_loss_mlp": 1.00035667, "epoch": 0.4083448566103529, "flos": 19062572924640.0, "grad_norm": 1.7263919120978906, "language_loss": 0.77544612, "learning_rate": 2.6772852779621435e-06, "loss": 0.79458284, "num_input_tokens_seen": 73349105, "step": 3396, "time_per_iteration": 2.554511308670044 }, { "auxiliary_loss_clip": 0.01168463, "auxiliary_loss_mlp": 0.007627, "balance_loss_clip": 1.05431807, "balance_loss_mlp": 1.00025141, "epoch": 0.408465099500992, "flos": 23550257818560.0, "grad_norm": 2.062896988566972, "language_loss": 0.86610901, "learning_rate": 2.676552280449239e-06, "loss": 0.88542068, "num_input_tokens_seen": 73368990, "step": 3397, "time_per_iteration": 3.2445592880249023 }, { "auxiliary_loss_clip": 0.01166516, "auxiliary_loss_mlp": 0.01033668, "balance_loss_clip": 1.05116796, "balance_loss_mlp": 1.02519798, "epoch": 0.4085853423916311, "flos": 12750015290400.0, "grad_norm": 2.470486745947221, "language_loss": 0.75899369, "learning_rate": 2.6758191803038917e-06, "loss": 0.78099555, "num_input_tokens_seen": 73387485, "step": 3398, "time_per_iteration": 2.5278282165527344 }, { "auxiliary_loss_clip": 0.01112834, "auxiliary_loss_mlp": 0.01030009, "balance_loss_clip": 1.0492301, "balance_loss_mlp": 1.02155125, "epoch": 0.4087055852822702, "flos": 24353076833280.0, "grad_norm": 1.6570246151041852, "language_loss": 0.82573497, "learning_rate": 2.6750859776373125e-06, "loss": 0.84716344, "num_input_tokens_seen": 73406940, "step": 3399, "time_per_iteration": 3.4193594455718994 }, { "auxiliary_loss_clip": 0.0102708, "auxiliary_loss_mlp": 0.01002055, "balance_loss_clip": 1.01223755, "balance_loss_mlp": 1.0006243, "epoch": 0.4088258281729093, "flos": 66387962235360.0, "grad_norm": 0.7687428357708975, "language_loss": 0.60429442, "learning_rate": 2.674352672560727e-06, "loss": 0.62458575, "num_input_tokens_seen": 73468385, "step": 3400, "time_per_iteration": 3.2239620685577393 }, { "auxiliary_loss_clip": 0.0114549, "auxiliary_loss_mlp": 0.01036382, "balance_loss_clip": 1.05081487, "balance_loss_mlp": 1.02797198, "epoch": 0.40894607106354836, "flos": 20449263578880.0, "grad_norm": 1.53080559777341, "language_loss": 0.76893991, "learning_rate": 2.673619265185377e-06, "loss": 0.79075861, "num_input_tokens_seen": 73488225, "step": 3401, "time_per_iteration": 2.6275694370269775 }, { "auxiliary_loss_clip": 0.01177104, "auxiliary_loss_mlp": 0.01029705, "balance_loss_clip": 1.05211353, "balance_loss_mlp": 1.02190232, "epoch": 0.40906631395418747, "flos": 27053631323520.0, "grad_norm": 1.6715045681466079, "language_loss": 0.78171003, "learning_rate": 2.672885755622521e-06, "loss": 0.80377805, "num_input_tokens_seen": 73510640, "step": 3402, "time_per_iteration": 2.591601610183716 }, { "auxiliary_loss_clip": 0.01127081, "auxiliary_loss_mlp": 0.01026895, "balance_loss_clip": 1.04551768, "balance_loss_mlp": 1.01889563, "epoch": 0.4091865568448266, "flos": 25484155284480.0, "grad_norm": 2.6806409972962046, "language_loss": 0.69902694, "learning_rate": 2.67215214398343e-06, "loss": 0.72056675, "num_input_tokens_seen": 73530655, "step": 3403, "time_per_iteration": 2.6509668827056885 }, { "auxiliary_loss_clip": 0.01136052, "auxiliary_loss_mlp": 0.01033206, "balance_loss_clip": 1.04802513, "balance_loss_mlp": 1.02508187, "epoch": 0.40930679973546563, "flos": 28657868591520.0, "grad_norm": 1.9731345004461713, "language_loss": 0.78550351, "learning_rate": 2.671418430379393e-06, "loss": 0.80719614, "num_input_tokens_seen": 73549340, "step": 3404, "time_per_iteration": 2.6808197498321533 }, { "auxiliary_loss_clip": 0.01189346, "auxiliary_loss_mlp": 0.01030332, "balance_loss_clip": 1.05465364, "balance_loss_mlp": 1.02233958, "epoch": 0.40942704262610474, "flos": 20886296322240.0, "grad_norm": 1.8501527501552075, "language_loss": 0.83314413, "learning_rate": 2.670684614921715e-06, "loss": 0.85534096, "num_input_tokens_seen": 73568315, "step": 3405, "time_per_iteration": 2.497819423675537 }, { "auxiliary_loss_clip": 0.0116266, "auxiliary_loss_mlp": 0.01033994, "balance_loss_clip": 1.05151629, "balance_loss_mlp": 1.02600098, "epoch": 0.4095472855167438, "flos": 21618084366720.0, "grad_norm": 1.9694129331536177, "language_loss": 0.69325888, "learning_rate": 2.6699506977217128e-06, "loss": 0.7152254, "num_input_tokens_seen": 73588490, "step": 3406, "time_per_iteration": 2.5811147689819336 }, { "auxiliary_loss_clip": 0.01173996, "auxiliary_loss_mlp": 0.01028012, "balance_loss_clip": 1.05498743, "balance_loss_mlp": 1.01996493, "epoch": 0.4096675284073829, "flos": 27926116464000.0, "grad_norm": 2.052110986435794, "language_loss": 0.70100731, "learning_rate": 2.6692166788907233e-06, "loss": 0.72302735, "num_input_tokens_seen": 73608685, "step": 3407, "time_per_iteration": 2.555365562438965 }, { "auxiliary_loss_clip": 0.01164182, "auxiliary_loss_mlp": 0.01030248, "balance_loss_clip": 1.05290926, "balance_loss_mlp": 1.02170706, "epoch": 0.409787771298022, "flos": 19206610297920.0, "grad_norm": 2.1419517974784, "language_loss": 0.76819587, "learning_rate": 2.6684825585400957e-06, "loss": 0.79014021, "num_input_tokens_seen": 73627630, "step": 3408, "time_per_iteration": 2.529127359390259 }, { "auxiliary_loss_clip": 0.01052536, "auxiliary_loss_mlp": 0.01006967, "balance_loss_clip": 1.01377964, "balance_loss_mlp": 1.00567997, "epoch": 0.4099080141886611, "flos": 59269245309120.0, "grad_norm": 0.8180407563829696, "language_loss": 0.65113705, "learning_rate": 2.6677483367811947e-06, "loss": 0.67173207, "num_input_tokens_seen": 73687670, "step": 3409, "time_per_iteration": 3.2401795387268066 }, { "auxiliary_loss_clip": 0.01177406, "auxiliary_loss_mlp": 0.01026571, "balance_loss_clip": 1.05255723, "balance_loss_mlp": 1.01824999, "epoch": 0.4100282570793002, "flos": 21906446448960.0, "grad_norm": 1.7160842173499762, "language_loss": 0.75558168, "learning_rate": 2.6670140137254028e-06, "loss": 0.77762145, "num_input_tokens_seen": 73707145, "step": 3410, "time_per_iteration": 2.5219860076904297 }, { "auxiliary_loss_clip": 0.01128767, "auxiliary_loss_mlp": 0.01031729, "balance_loss_clip": 1.04832458, "balance_loss_mlp": 1.02356923, "epoch": 0.4101484999699393, "flos": 18551599937280.0, "grad_norm": 2.7026226465366348, "language_loss": 0.89954209, "learning_rate": 2.666279589484115e-06, "loss": 0.92114705, "num_input_tokens_seen": 73725045, "step": 3411, "time_per_iteration": 2.563370704650879 }, { "auxiliary_loss_clip": 0.01131097, "auxiliary_loss_mlp": 0.01030253, "balance_loss_clip": 1.04634607, "balance_loss_mlp": 1.02236724, "epoch": 0.41026874286057835, "flos": 19094532630720.0, "grad_norm": 1.9554528512297384, "language_loss": 0.81265545, "learning_rate": 2.6655450641687435e-06, "loss": 0.83426893, "num_input_tokens_seen": 73742610, "step": 3412, "time_per_iteration": 2.6024861335754395 }, { "auxiliary_loss_clip": 0.01190878, "auxiliary_loss_mlp": 0.01037227, "balance_loss_clip": 1.05798042, "balance_loss_mlp": 1.02971351, "epoch": 0.41038898575121746, "flos": 31209572835840.0, "grad_norm": 2.186221665184028, "language_loss": 0.69198847, "learning_rate": 2.664810437890715e-06, "loss": 0.71426952, "num_input_tokens_seen": 73764280, "step": 3413, "time_per_iteration": 2.5584638118743896 }, { "auxiliary_loss_clip": 0.01113703, "auxiliary_loss_mlp": 0.01034094, "balance_loss_clip": 1.05331254, "balance_loss_mlp": 1.02632737, "epoch": 0.41050922864185657, "flos": 14355868821600.0, "grad_norm": 1.788113446233164, "language_loss": 0.79617727, "learning_rate": 2.6640757107614714e-06, "loss": 0.81765521, "num_input_tokens_seen": 73782375, "step": 3414, "time_per_iteration": 2.6362650394439697 }, { "auxiliary_loss_clip": 0.01142288, "auxiliary_loss_mlp": 0.0103512, "balance_loss_clip": 1.0504154, "balance_loss_mlp": 1.02633452, "epoch": 0.4106294715324956, "flos": 30956295235200.0, "grad_norm": 1.9524486435409765, "language_loss": 0.68772286, "learning_rate": 2.6633408828924697e-06, "loss": 0.70949697, "num_input_tokens_seen": 73801240, "step": 3415, "time_per_iteration": 2.637549638748169 }, { "auxiliary_loss_clip": 0.01156401, "auxiliary_loss_mlp": 0.01033347, "balance_loss_clip": 1.05303025, "balance_loss_mlp": 1.02428102, "epoch": 0.41074971442313474, "flos": 24457324603200.0, "grad_norm": 1.461512566736418, "language_loss": 0.69953454, "learning_rate": 2.662605954395185e-06, "loss": 0.72143209, "num_input_tokens_seen": 73821200, "step": 3416, "time_per_iteration": 2.6196539402008057 }, { "auxiliary_loss_clip": 0.01177467, "auxiliary_loss_mlp": 0.01031993, "balance_loss_clip": 1.0524354, "balance_loss_mlp": 1.02423286, "epoch": 0.41086995731377385, "flos": 21542994780000.0, "grad_norm": 1.6895673881103805, "language_loss": 0.83738673, "learning_rate": 2.6618709253811027e-06, "loss": 0.85948133, "num_input_tokens_seen": 73840655, "step": 3417, "time_per_iteration": 2.5648837089538574 }, { "auxiliary_loss_clip": 0.01186718, "auxiliary_loss_mlp": 0.01029036, "balance_loss_clip": 1.05600286, "balance_loss_mlp": 1.02216339, "epoch": 0.4109902002044129, "flos": 20702756681280.0, "grad_norm": 1.5901864732029223, "language_loss": 0.87839228, "learning_rate": 2.6611357959617277e-06, "loss": 0.90054983, "num_input_tokens_seen": 73860275, "step": 3418, "time_per_iteration": 3.2716448307037354 }, { "auxiliary_loss_clip": 0.01141721, "auxiliary_loss_mlp": 0.01029514, "balance_loss_clip": 1.05177546, "balance_loss_mlp": 1.02123499, "epoch": 0.411110443095052, "flos": 18179995118400.0, "grad_norm": 1.881370484515619, "language_loss": 0.90687275, "learning_rate": 2.660400566248578e-06, "loss": 0.92858511, "num_input_tokens_seen": 73878400, "step": 3419, "time_per_iteration": 2.5680904388427734 }, { "auxiliary_loss_clip": 0.01149219, "auxiliary_loss_mlp": 0.01034393, "balance_loss_clip": 1.04975104, "balance_loss_mlp": 1.02569687, "epoch": 0.41123068598569107, "flos": 14575247200320.0, "grad_norm": 2.249308838553358, "language_loss": 0.67019147, "learning_rate": 2.6596652363531876e-06, "loss": 0.69202763, "num_input_tokens_seen": 73894275, "step": 3420, "time_per_iteration": 3.3035264015197754 }, { "auxiliary_loss_clip": 0.0119292, "auxiliary_loss_mlp": 0.01029903, "balance_loss_clip": 1.05871212, "balance_loss_mlp": 1.02206516, "epoch": 0.4113509288763302, "flos": 21177998681760.0, "grad_norm": 1.6225778039384862, "language_loss": 0.78245151, "learning_rate": 2.6589298063871055e-06, "loss": 0.80467975, "num_input_tokens_seen": 73914450, "step": 3421, "time_per_iteration": 2.520251989364624 }, { "auxiliary_loss_clip": 0.01190384, "auxiliary_loss_mlp": 0.01033817, "balance_loss_clip": 1.05735135, "balance_loss_mlp": 1.02566326, "epoch": 0.4114711717669693, "flos": 18442216042080.0, "grad_norm": 1.8022889319160975, "language_loss": 0.69817215, "learning_rate": 2.658194276461895e-06, "loss": 0.72041416, "num_input_tokens_seen": 73932375, "step": 3422, "time_per_iteration": 2.4457342624664307 }, { "auxiliary_loss_clip": 0.01164345, "auxiliary_loss_mlp": 0.01033999, "balance_loss_clip": 1.05225754, "balance_loss_mlp": 1.02483821, "epoch": 0.41159141465760835, "flos": 27233399683680.0, "grad_norm": 1.842318225305179, "language_loss": 0.66963589, "learning_rate": 2.6574586466891368e-06, "loss": 0.69161934, "num_input_tokens_seen": 73952850, "step": 3423, "time_per_iteration": 3.310464859008789 }, { "auxiliary_loss_clip": 0.01165193, "auxiliary_loss_mlp": 0.00762713, "balance_loss_clip": 1.05431509, "balance_loss_mlp": 1.00045335, "epoch": 0.41171165754824746, "flos": 20006879208480.0, "grad_norm": 2.5580985160490513, "language_loss": 0.64643675, "learning_rate": 2.6567229171804247e-06, "loss": 0.66571581, "num_input_tokens_seen": 73970735, "step": 3424, "time_per_iteration": 2.54164719581604 }, { "auxiliary_loss_clip": 0.01157263, "auxiliary_loss_mlp": 0.01027333, "balance_loss_clip": 1.05091143, "balance_loss_mlp": 1.01902986, "epoch": 0.41183190043888657, "flos": 18004321291680.0, "grad_norm": 2.460805697584164, "language_loss": 0.87417531, "learning_rate": 2.655987088047368e-06, "loss": 0.89602131, "num_input_tokens_seen": 73989080, "step": 3425, "time_per_iteration": 3.2944753170013428 }, { "auxiliary_loss_clip": 0.0116164, "auxiliary_loss_mlp": 0.01032316, "balance_loss_clip": 1.05404687, "balance_loss_mlp": 1.02427518, "epoch": 0.4119521433295256, "flos": 27163374388320.0, "grad_norm": 1.954403519047017, "language_loss": 0.78782469, "learning_rate": 2.6552511594015912e-06, "loss": 0.80976427, "num_input_tokens_seen": 74009470, "step": 3426, "time_per_iteration": 2.621838331222534 }, { "auxiliary_loss_clip": 0.01159511, "auxiliary_loss_mlp": 0.01038214, "balance_loss_clip": 1.05031037, "balance_loss_mlp": 1.02967238, "epoch": 0.41207238622016473, "flos": 15122023008480.0, "grad_norm": 2.029342109450828, "language_loss": 0.85444188, "learning_rate": 2.654515131354735e-06, "loss": 0.87641919, "num_input_tokens_seen": 74027735, "step": 3427, "time_per_iteration": 2.5257012844085693 }, { "auxiliary_loss_clip": 0.01149633, "auxiliary_loss_mlp": 0.01027174, "balance_loss_clip": 1.05370796, "balance_loss_mlp": 1.01919329, "epoch": 0.41219262911080384, "flos": 27052877067360.0, "grad_norm": 1.919566609172866, "language_loss": 0.8575362, "learning_rate": 2.653779004018453e-06, "loss": 0.87930423, "num_input_tokens_seen": 74048300, "step": 3428, "time_per_iteration": 2.6385040283203125 }, { "auxiliary_loss_clip": 0.01153157, "auxiliary_loss_mlp": 0.010333, "balance_loss_clip": 1.04974532, "balance_loss_mlp": 1.02550983, "epoch": 0.4123128720014429, "flos": 24686364644160.0, "grad_norm": 1.838419970848746, "language_loss": 0.82337523, "learning_rate": 2.653042777504417e-06, "loss": 0.84523982, "num_input_tokens_seen": 74070890, "step": 3429, "time_per_iteration": 2.6188669204711914 }, { "auxiliary_loss_clip": 0.01170087, "auxiliary_loss_mlp": 0.01032099, "balance_loss_clip": 1.05531251, "balance_loss_mlp": 1.02367663, "epoch": 0.412433114892082, "flos": 26244778259520.0, "grad_norm": 1.7223011099349659, "language_loss": 0.79801923, "learning_rate": 2.6523064519243105e-06, "loss": 0.82004106, "num_input_tokens_seen": 74090460, "step": 3430, "time_per_iteration": 2.596240282058716 }, { "auxiliary_loss_clip": 0.01181602, "auxiliary_loss_mlp": 0.01030617, "balance_loss_clip": 1.05818391, "balance_loss_mlp": 1.02144957, "epoch": 0.4125533577827211, "flos": 21361035485280.0, "grad_norm": 2.3099690310866885, "language_loss": 0.78836548, "learning_rate": 2.6515700273898333e-06, "loss": 0.81048769, "num_input_tokens_seen": 74108335, "step": 3431, "time_per_iteration": 2.5077908039093018 }, { "auxiliary_loss_clip": 0.01151481, "auxiliary_loss_mlp": 0.0103232, "balance_loss_clip": 1.05155766, "balance_loss_mlp": 1.02435708, "epoch": 0.4126736006733602, "flos": 26067560003520.0, "grad_norm": 1.9129124722376611, "language_loss": 0.68949497, "learning_rate": 2.6508335040127018e-06, "loss": 0.71133298, "num_input_tokens_seen": 74128030, "step": 3432, "time_per_iteration": 2.6031007766723633 }, { "auxiliary_loss_clip": 0.0117946, "auxiliary_loss_mlp": 0.01028622, "balance_loss_clip": 1.05486417, "balance_loss_mlp": 1.02101624, "epoch": 0.4127938435639993, "flos": 25666150496160.0, "grad_norm": 2.19210036567463, "language_loss": 0.76815093, "learning_rate": 2.6500968819046446e-06, "loss": 0.79023176, "num_input_tokens_seen": 74148330, "step": 3433, "time_per_iteration": 2.5680911540985107 }, { "auxiliary_loss_clip": 0.01135052, "auxiliary_loss_mlp": 0.01031161, "balance_loss_clip": 1.04789281, "balance_loss_mlp": 1.0231266, "epoch": 0.4129140864546384, "flos": 17995916723040.0, "grad_norm": 2.769672272686257, "language_loss": 0.59038472, "learning_rate": 2.649360161177408e-06, "loss": 0.61204684, "num_input_tokens_seen": 74163390, "step": 3434, "time_per_iteration": 2.554147243499756 }, { "auxiliary_loss_clip": 0.01184625, "auxiliary_loss_mlp": 0.01032004, "balance_loss_clip": 1.05696535, "balance_loss_mlp": 1.02314091, "epoch": 0.41303432934527745, "flos": 23732899535520.0, "grad_norm": 1.7603877964612347, "language_loss": 0.73453009, "learning_rate": 2.6486233419427504e-06, "loss": 0.7566964, "num_input_tokens_seen": 74183205, "step": 3435, "time_per_iteration": 2.55033540725708 }, { "auxiliary_loss_clip": 0.01141031, "auxiliary_loss_mlp": 0.01037785, "balance_loss_clip": 1.05250072, "balance_loss_mlp": 1.0296309, "epoch": 0.41315457223591656, "flos": 19755289704960.0, "grad_norm": 2.435452013956697, "language_loss": 0.74887419, "learning_rate": 2.6478864243124484e-06, "loss": 0.77066237, "num_input_tokens_seen": 74202870, "step": 3436, "time_per_iteration": 2.579155683517456 }, { "auxiliary_loss_clip": 0.0117974, "auxiliary_loss_mlp": 0.01026038, "balance_loss_clip": 1.05488324, "balance_loss_mlp": 1.01829505, "epoch": 0.4132748151265556, "flos": 20923320319680.0, "grad_norm": 1.7272978613803938, "language_loss": 0.85111248, "learning_rate": 2.6471494083982903e-06, "loss": 0.87317026, "num_input_tokens_seen": 74222255, "step": 3437, "time_per_iteration": 2.5169484615325928 }, { "auxiliary_loss_clip": 0.01154027, "auxiliary_loss_mlp": 0.01033353, "balance_loss_clip": 1.05247688, "balance_loss_mlp": 1.02584291, "epoch": 0.4133950580171947, "flos": 32232524485440.0, "grad_norm": 2.1217588727713173, "language_loss": 0.74906242, "learning_rate": 2.6464122943120818e-06, "loss": 0.77093625, "num_input_tokens_seen": 74242480, "step": 3438, "time_per_iteration": 2.654693126678467 }, { "auxiliary_loss_clip": 0.01151404, "auxiliary_loss_mlp": 0.01034377, "balance_loss_clip": 1.05638885, "balance_loss_mlp": 1.02619934, "epoch": 0.41351530090783384, "flos": 23292490597920.0, "grad_norm": 2.425249844474835, "language_loss": 0.82329834, "learning_rate": 2.645675082165642e-06, "loss": 0.84515619, "num_input_tokens_seen": 74258690, "step": 3439, "time_per_iteration": 2.550792932510376 }, { "auxiliary_loss_clip": 0.011664, "auxiliary_loss_mlp": 0.010342, "balance_loss_clip": 1.05770373, "balance_loss_mlp": 1.02610588, "epoch": 0.4136355437984729, "flos": 25593575096640.0, "grad_norm": 1.9584682511332003, "language_loss": 0.75037175, "learning_rate": 2.644937772070806e-06, "loss": 0.77237779, "num_input_tokens_seen": 74277135, "step": 3440, "time_per_iteration": 2.613115072250366 }, { "auxiliary_loss_clip": 0.01192599, "auxiliary_loss_mlp": 0.01034286, "balance_loss_clip": 1.05712843, "balance_loss_mlp": 1.02643585, "epoch": 0.413755786689112, "flos": 19828619360640.0, "grad_norm": 2.7146418408461837, "language_loss": 0.83454943, "learning_rate": 2.6442003641394225e-06, "loss": 0.85681832, "num_input_tokens_seen": 74294730, "step": 3441, "time_per_iteration": 2.478756904602051 }, { "auxiliary_loss_clip": 0.01159989, "auxiliary_loss_mlp": 0.01026279, "balance_loss_clip": 1.05147707, "balance_loss_mlp": 1.0186615, "epoch": 0.4138760295797511, "flos": 26870450852160.0, "grad_norm": 1.4611840745169915, "language_loss": 0.83946025, "learning_rate": 2.643462858483356e-06, "loss": 0.861323, "num_input_tokens_seen": 74315015, "step": 3442, "time_per_iteration": 2.6050212383270264 }, { "auxiliary_loss_clip": 0.01125917, "auxiliary_loss_mlp": 0.0102663, "balance_loss_clip": 1.04574418, "balance_loss_mlp": 1.01819587, "epoch": 0.41399627247039017, "flos": 16399257933600.0, "grad_norm": 1.898229380525871, "language_loss": 0.72851026, "learning_rate": 2.6427252552144856e-06, "loss": 0.75003576, "num_input_tokens_seen": 74333665, "step": 3443, "time_per_iteration": 2.592132806777954 }, { "auxiliary_loss_clip": 0.0119299, "auxiliary_loss_mlp": 0.01031401, "balance_loss_clip": 1.05740535, "balance_loss_mlp": 1.02331221, "epoch": 0.4141165153610293, "flos": 22930224188640.0, "grad_norm": 3.9042920341918768, "language_loss": 0.74908018, "learning_rate": 2.6419875544447044e-06, "loss": 0.7713241, "num_input_tokens_seen": 74355065, "step": 3444, "time_per_iteration": 3.279947280883789 }, { "auxiliary_loss_clip": 0.01192196, "auxiliary_loss_mlp": 0.01031785, "balance_loss_clip": 1.0565033, "balance_loss_mlp": 1.02388775, "epoch": 0.4142367582516684, "flos": 25192560675840.0, "grad_norm": 1.6208959993517975, "language_loss": 0.71762872, "learning_rate": 2.6412497562859218e-06, "loss": 0.73986852, "num_input_tokens_seen": 74376345, "step": 3445, "time_per_iteration": 2.5264956951141357 }, { "auxiliary_loss_clip": 0.01182955, "auxiliary_loss_mlp": 0.01035184, "balance_loss_clip": 1.05426693, "balance_loss_mlp": 1.02710795, "epoch": 0.41435700114230745, "flos": 21690480181440.0, "grad_norm": 2.9829235200951167, "language_loss": 0.76354885, "learning_rate": 2.6405118608500617e-06, "loss": 0.78573024, "num_input_tokens_seen": 74395170, "step": 3446, "time_per_iteration": 3.2661197185516357 }, { "auxiliary_loss_clip": 0.01141748, "auxiliary_loss_mlp": 0.01028545, "balance_loss_clip": 1.05393314, "balance_loss_mlp": 1.02088594, "epoch": 0.41447724403294656, "flos": 25995164188800.0, "grad_norm": 1.6090972112262372, "language_loss": 0.81159973, "learning_rate": 2.6397738682490613e-06, "loss": 0.83330262, "num_input_tokens_seen": 74416070, "step": 3447, "time_per_iteration": 2.6280837059020996 }, { "auxiliary_loss_clip": 0.01195025, "auxiliary_loss_mlp": 0.01037447, "balance_loss_clip": 1.05861902, "balance_loss_mlp": 1.02929306, "epoch": 0.41459748692358567, "flos": 18259682076000.0, "grad_norm": 1.7085055242843208, "language_loss": 0.75099599, "learning_rate": 2.6390357785948734e-06, "loss": 0.77332067, "num_input_tokens_seen": 74433185, "step": 3448, "time_per_iteration": 2.4459879398345947 }, { "auxiliary_loss_clip": 0.01178629, "auxiliary_loss_mlp": 0.01035459, "balance_loss_clip": 1.05662441, "balance_loss_mlp": 1.02735293, "epoch": 0.4147177298142247, "flos": 24168459683520.0, "grad_norm": 1.8727766653934568, "language_loss": 0.80238599, "learning_rate": 2.6382975919994667e-06, "loss": 0.82452679, "num_input_tokens_seen": 74453760, "step": 3449, "time_per_iteration": 3.2799458503723145 }, { "auxiliary_loss_clip": 0.01164378, "auxiliary_loss_mlp": 0.01027141, "balance_loss_clip": 1.05331326, "balance_loss_mlp": 1.01981306, "epoch": 0.41483797270486383, "flos": 20084662567200.0, "grad_norm": 1.6528053427717821, "language_loss": 0.73042488, "learning_rate": 2.637559308574822e-06, "loss": 0.75234008, "num_input_tokens_seen": 74473505, "step": 3450, "time_per_iteration": 2.538128137588501 }, { "auxiliary_loss_clip": 0.01193408, "auxiliary_loss_mlp": 0.01027906, "balance_loss_clip": 1.05727768, "balance_loss_mlp": 1.02034175, "epoch": 0.4149582155955029, "flos": 30081044488800.0, "grad_norm": 1.7963328631083095, "language_loss": 0.70902258, "learning_rate": 2.6368209284329376e-06, "loss": 0.73123574, "num_input_tokens_seen": 74494135, "step": 3451, "time_per_iteration": 3.2661163806915283 }, { "auxiliary_loss_clip": 0.01176873, "auxiliary_loss_mlp": 0.01032373, "balance_loss_clip": 1.05347681, "balance_loss_mlp": 1.02398634, "epoch": 0.415078458486142, "flos": 16764397699680.0, "grad_norm": 1.945516697353127, "language_loss": 0.75516593, "learning_rate": 2.636082451685825e-06, "loss": 0.7772584, "num_input_tokens_seen": 74512335, "step": 3452, "time_per_iteration": 2.4974353313446045 }, { "auxiliary_loss_clip": 0.01168637, "auxiliary_loss_mlp": 0.01031539, "balance_loss_clip": 1.05808425, "balance_loss_mlp": 1.02352238, "epoch": 0.4151987013767811, "flos": 26033696698560.0, "grad_norm": 1.5338516769489359, "language_loss": 0.86571527, "learning_rate": 2.6353438784455094e-06, "loss": 0.88771701, "num_input_tokens_seen": 74535620, "step": 3453, "time_per_iteration": 2.6103599071502686 }, { "auxiliary_loss_clip": 0.01162689, "auxiliary_loss_mlp": 0.01032928, "balance_loss_clip": 1.05491877, "balance_loss_mlp": 1.02382076, "epoch": 0.41531894426742016, "flos": 24608006614080.0, "grad_norm": 2.137353694250531, "language_loss": 0.71604061, "learning_rate": 2.6346052088240326e-06, "loss": 0.73799682, "num_input_tokens_seen": 74555140, "step": 3454, "time_per_iteration": 2.6163058280944824 }, { "auxiliary_loss_clip": 0.01165056, "auxiliary_loss_mlp": 0.01029238, "balance_loss_clip": 1.05297256, "balance_loss_mlp": 1.02165604, "epoch": 0.4154391871580593, "flos": 14975794700640.0, "grad_norm": 1.9121736893783567, "language_loss": 0.76881021, "learning_rate": 2.63386644293345e-06, "loss": 0.79075313, "num_input_tokens_seen": 74571485, "step": 3455, "time_per_iteration": 2.546999454498291 }, { "auxiliary_loss_clip": 0.01143145, "auxiliary_loss_mlp": 0.01027133, "balance_loss_clip": 1.04746819, "balance_loss_mlp": 1.01931834, "epoch": 0.4155594300486984, "flos": 14647176094560.0, "grad_norm": 2.091998892211027, "language_loss": 0.83025461, "learning_rate": 2.633127580885833e-06, "loss": 0.85195744, "num_input_tokens_seen": 74585985, "step": 3456, "time_per_iteration": 2.5644052028656006 }, { "auxiliary_loss_clip": 0.01193896, "auxiliary_loss_mlp": 0.01031876, "balance_loss_clip": 1.05952072, "balance_loss_mlp": 1.02363896, "epoch": 0.41567967293933744, "flos": 29497280600160.0, "grad_norm": 1.9937226986559775, "language_loss": 0.64388084, "learning_rate": 2.632388622793265e-06, "loss": 0.66613853, "num_input_tokens_seen": 74605140, "step": 3457, "time_per_iteration": 2.567978858947754 }, { "auxiliary_loss_clip": 0.01176522, "auxiliary_loss_mlp": 0.01033321, "balance_loss_clip": 1.05556965, "balance_loss_mlp": 1.02601933, "epoch": 0.41579991582997655, "flos": 19238390419200.0, "grad_norm": 2.0708310386826576, "language_loss": 0.67729551, "learning_rate": 2.6316495687678457e-06, "loss": 0.69939399, "num_input_tokens_seen": 74623790, "step": 3458, "time_per_iteration": 2.5229711532592773 }, { "auxiliary_loss_clip": 0.01125351, "auxiliary_loss_mlp": 0.01031356, "balance_loss_clip": 1.04811859, "balance_loss_mlp": 1.02299976, "epoch": 0.41592015872061566, "flos": 24462065641920.0, "grad_norm": 2.4661314839019735, "language_loss": 0.76568735, "learning_rate": 2.6309104189216887e-06, "loss": 0.78725439, "num_input_tokens_seen": 74641355, "step": 3459, "time_per_iteration": 2.6281638145446777 }, { "auxiliary_loss_clip": 0.01136084, "auxiliary_loss_mlp": 0.00763703, "balance_loss_clip": 1.04903269, "balance_loss_mlp": 1.00059724, "epoch": 0.4160404016112547, "flos": 20775655333440.0, "grad_norm": 2.3389566501675367, "language_loss": 0.74432695, "learning_rate": 2.630171173366923e-06, "loss": 0.76332486, "num_input_tokens_seen": 74657155, "step": 3460, "time_per_iteration": 2.57796311378479 }, { "auxiliary_loss_clip": 0.01134243, "auxiliary_loss_mlp": 0.01035891, "balance_loss_clip": 1.05038071, "balance_loss_mlp": 1.0278151, "epoch": 0.41616064450189383, "flos": 13916465558880.0, "grad_norm": 2.4031859884824938, "language_loss": 0.74045491, "learning_rate": 2.629431832215691e-06, "loss": 0.76215625, "num_input_tokens_seen": 74671960, "step": 3461, "time_per_iteration": 2.577038526535034 }, { "auxiliary_loss_clip": 0.01158932, "auxiliary_loss_mlp": 0.01024004, "balance_loss_clip": 1.05286098, "balance_loss_mlp": 1.01615953, "epoch": 0.41628088739253294, "flos": 20010830074080.0, "grad_norm": 1.6735524074786752, "language_loss": 0.87121814, "learning_rate": 2.628692395580151e-06, "loss": 0.89304751, "num_input_tokens_seen": 74692050, "step": 3462, "time_per_iteration": 2.580202341079712 }, { "auxiliary_loss_clip": 0.01101978, "auxiliary_loss_mlp": 0.01034193, "balance_loss_clip": 1.04499614, "balance_loss_mlp": 1.02615261, "epoch": 0.416401130283172, "flos": 29168805661920.0, "grad_norm": 1.6978364919357103, "language_loss": 0.79220384, "learning_rate": 2.6279528635724747e-06, "loss": 0.81356555, "num_input_tokens_seen": 74712205, "step": 3463, "time_per_iteration": 2.6900956630706787 }, { "auxiliary_loss_clip": 0.01179914, "auxiliary_loss_mlp": 0.01035134, "balance_loss_clip": 1.05549943, "balance_loss_mlp": 1.02652109, "epoch": 0.4165213731738111, "flos": 16246815991680.0, "grad_norm": 2.506691279745895, "language_loss": 0.78213143, "learning_rate": 2.627213236304848e-06, "loss": 0.80428195, "num_input_tokens_seen": 74729005, "step": 3464, "time_per_iteration": 2.503690242767334 }, { "auxiliary_loss_clip": 0.01182082, "auxiliary_loss_mlp": 0.01029755, "balance_loss_clip": 1.0559392, "balance_loss_mlp": 1.02164316, "epoch": 0.4166416160644502, "flos": 33765443447520.0, "grad_norm": 1.8188022956401815, "language_loss": 0.70897269, "learning_rate": 2.626473513889472e-06, "loss": 0.73109102, "num_input_tokens_seen": 74751385, "step": 3465, "time_per_iteration": 2.633726119995117 }, { "auxiliary_loss_clip": 0.01171099, "auxiliary_loss_mlp": 0.01033658, "balance_loss_clip": 1.05516815, "balance_loss_mlp": 1.02598119, "epoch": 0.41676185895508927, "flos": 20917501772160.0, "grad_norm": 1.7323021757575288, "language_loss": 0.82383573, "learning_rate": 2.625733696438562e-06, "loss": 0.84588337, "num_input_tokens_seen": 74768890, "step": 3466, "time_per_iteration": 2.5459179878234863 }, { "auxiliary_loss_clip": 0.01160945, "auxiliary_loss_mlp": 0.01034507, "balance_loss_clip": 1.05371583, "balance_loss_mlp": 1.02589417, "epoch": 0.4168821018457284, "flos": 18406125885600.0, "grad_norm": 1.8488541906267997, "language_loss": 0.75223935, "learning_rate": 2.6249937840643476e-06, "loss": 0.77419388, "num_input_tokens_seen": 74787195, "step": 3467, "time_per_iteration": 2.566727638244629 }, { "auxiliary_loss_clip": 0.01194897, "auxiliary_loss_mlp": 0.00763426, "balance_loss_clip": 1.06031752, "balance_loss_mlp": 1.00057781, "epoch": 0.41700234473636744, "flos": 18698402916480.0, "grad_norm": 1.6332804063633195, "language_loss": 0.66289157, "learning_rate": 2.6242537768790733e-06, "loss": 0.68247485, "num_input_tokens_seen": 74806350, "step": 3468, "time_per_iteration": 2.4963302612304688 }, { "auxiliary_loss_clip": 0.01175651, "auxiliary_loss_mlp": 0.01029391, "balance_loss_clip": 1.0536437, "balance_loss_mlp": 1.02136278, "epoch": 0.41712258762700655, "flos": 31033288420800.0, "grad_norm": 1.7143640497795813, "language_loss": 0.6879341, "learning_rate": 2.6235136749949975e-06, "loss": 0.70998454, "num_input_tokens_seen": 74829800, "step": 3469, "time_per_iteration": 2.597458839416504 }, { "auxiliary_loss_clip": 0.01190319, "auxiliary_loss_mlp": 0.01026571, "balance_loss_clip": 1.05675149, "balance_loss_mlp": 1.01844132, "epoch": 0.41724283051764566, "flos": 35914768426560.0, "grad_norm": 2.325920137166767, "language_loss": 0.61131191, "learning_rate": 2.6227734785243924e-06, "loss": 0.63348079, "num_input_tokens_seen": 74849760, "step": 3470, "time_per_iteration": 2.607560634613037 }, { "auxiliary_loss_clip": 0.01112992, "auxiliary_loss_mlp": 0.01027035, "balance_loss_clip": 1.04649568, "balance_loss_mlp": 1.01929235, "epoch": 0.4173630734082847, "flos": 25333652858400.0, "grad_norm": 2.214964602128782, "language_loss": 0.79464138, "learning_rate": 2.6220331875795466e-06, "loss": 0.81604165, "num_input_tokens_seen": 74869110, "step": 3471, "time_per_iteration": 3.4776742458343506 }, { "auxiliary_loss_clip": 0.0117297, "auxiliary_loss_mlp": 0.01030852, "balance_loss_clip": 1.05438292, "balance_loss_mlp": 1.02316928, "epoch": 0.4174833162989238, "flos": 26685402698880.0, "grad_norm": 1.723990047032595, "language_loss": 0.75334883, "learning_rate": 2.62129280227276e-06, "loss": 0.77538705, "num_input_tokens_seen": 74889110, "step": 3472, "time_per_iteration": 3.316481828689575 }, { "auxiliary_loss_clip": 0.01184101, "auxiliary_loss_mlp": 0.01029979, "balance_loss_clip": 1.05785942, "balance_loss_mlp": 1.02228427, "epoch": 0.41760355918956293, "flos": 74739592558080.0, "grad_norm": 2.135614889004511, "language_loss": 0.68408638, "learning_rate": 2.62055232271635e-06, "loss": 0.70622718, "num_input_tokens_seen": 74916260, "step": 3473, "time_per_iteration": 2.9528462886810303 }, { "auxiliary_loss_clip": 0.01137729, "auxiliary_loss_mlp": 0.01029905, "balance_loss_clip": 1.0485301, "balance_loss_mlp": 1.0214771, "epoch": 0.417723802080202, "flos": 14317515896640.0, "grad_norm": 2.0342315912788127, "language_loss": 0.8772409, "learning_rate": 2.619811749022646e-06, "loss": 0.8989172, "num_input_tokens_seen": 74931570, "step": 3474, "time_per_iteration": 2.552889823913574 }, { "auxiliary_loss_clip": 0.01181571, "auxiliary_loss_mlp": 0.0103425, "balance_loss_clip": 1.05840492, "balance_loss_mlp": 1.02568507, "epoch": 0.4178440449708411, "flos": 14643799900320.0, "grad_norm": 2.246448072908748, "language_loss": 0.71119952, "learning_rate": 2.6190710813039917e-06, "loss": 0.73335779, "num_input_tokens_seen": 74944695, "step": 3475, "time_per_iteration": 3.2555947303771973 }, { "auxiliary_loss_clip": 0.01123877, "auxiliary_loss_mlp": 0.00763332, "balance_loss_clip": 1.04249454, "balance_loss_mlp": 1.00048661, "epoch": 0.4179642878614802, "flos": 21507299710080.0, "grad_norm": 3.3219592043969954, "language_loss": 0.8362543, "learning_rate": 2.618330319672747e-06, "loss": 0.85512638, "num_input_tokens_seen": 74964115, "step": 3476, "time_per_iteration": 2.6181511878967285 }, { "auxiliary_loss_clip": 0.01191879, "auxiliary_loss_mlp": 0.01027222, "balance_loss_clip": 1.05781555, "balance_loss_mlp": 1.01936054, "epoch": 0.41808453075211927, "flos": 18442000540320.0, "grad_norm": 2.1537304968907507, "language_loss": 0.9228127, "learning_rate": 2.617589464241284e-06, "loss": 0.94500363, "num_input_tokens_seen": 74978515, "step": 3477, "time_per_iteration": 3.2249293327331543 }, { "auxiliary_loss_clip": 0.01151109, "auxiliary_loss_mlp": 0.01029374, "balance_loss_clip": 1.05365801, "balance_loss_mlp": 1.02259958, "epoch": 0.4182047736427584, "flos": 20301023921280.0, "grad_norm": 1.9504575178071315, "language_loss": 0.74487448, "learning_rate": 2.6168485151219914e-06, "loss": 0.76667935, "num_input_tokens_seen": 74998135, "step": 3478, "time_per_iteration": 2.617460250854492 }, { "auxiliary_loss_clip": 0.01179039, "auxiliary_loss_mlp": 0.01026013, "balance_loss_clip": 1.05805957, "balance_loss_mlp": 1.01761436, "epoch": 0.4183250165333975, "flos": 18876626847360.0, "grad_norm": 2.1732106482602056, "language_loss": 0.71433902, "learning_rate": 2.616107472427269e-06, "loss": 0.73638952, "num_input_tokens_seen": 75012830, "step": 3479, "time_per_iteration": 2.4929981231689453 }, { "auxiliary_loss_clip": 0.0118451, "auxiliary_loss_mlp": 0.01032907, "balance_loss_clip": 1.05605292, "balance_loss_mlp": 1.02443171, "epoch": 0.41844525942403654, "flos": 17740053101280.0, "grad_norm": 2.4044325808254357, "language_loss": 0.76202559, "learning_rate": 2.615366336269533e-06, "loss": 0.78419983, "num_input_tokens_seen": 75026495, "step": 3480, "time_per_iteration": 2.4951319694519043 }, { "auxiliary_loss_clip": 0.01194934, "auxiliary_loss_mlp": 0.0102919, "balance_loss_clip": 1.05753922, "balance_loss_mlp": 1.02001667, "epoch": 0.41856550231467565, "flos": 18361379741760.0, "grad_norm": 2.1295763642421237, "language_loss": 0.80633336, "learning_rate": 2.6146251067612126e-06, "loss": 0.82857454, "num_input_tokens_seen": 75041970, "step": 3481, "time_per_iteration": 2.4343130588531494 }, { "auxiliary_loss_clip": 0.01176857, "auxiliary_loss_mlp": 0.0103334, "balance_loss_clip": 1.0571456, "balance_loss_mlp": 1.02496576, "epoch": 0.41868574520531476, "flos": 22781804946240.0, "grad_norm": 2.2526314797704057, "language_loss": 0.82494158, "learning_rate": 2.6138837840147525e-06, "loss": 0.84704357, "num_input_tokens_seen": 75061005, "step": 3482, "time_per_iteration": 2.5319130420684814 }, { "auxiliary_loss_clip": 0.01148466, "auxiliary_loss_mlp": 0.01030953, "balance_loss_clip": 1.05228221, "balance_loss_mlp": 1.0233233, "epoch": 0.4188059880959538, "flos": 13699170363840.0, "grad_norm": 2.0228176263950512, "language_loss": 0.76187009, "learning_rate": 2.6131423681426103e-06, "loss": 0.78366429, "num_input_tokens_seen": 75076920, "step": 3483, "time_per_iteration": 2.53594708442688 }, { "auxiliary_loss_clip": 0.01190295, "auxiliary_loss_mlp": 0.01032676, "balance_loss_clip": 1.05725765, "balance_loss_mlp": 1.02526736, "epoch": 0.41892623098659293, "flos": 37818286532640.0, "grad_norm": 2.007593388905816, "language_loss": 0.72982574, "learning_rate": 2.6124008592572587e-06, "loss": 0.75205541, "num_input_tokens_seen": 75100905, "step": 3484, "time_per_iteration": 2.634382486343384 }, { "auxiliary_loss_clip": 0.01195537, "auxiliary_loss_mlp": 0.0102891, "balance_loss_clip": 1.05735707, "balance_loss_mlp": 1.02064872, "epoch": 0.419046473877232, "flos": 23258878711680.0, "grad_norm": 2.137643448723444, "language_loss": 0.82051885, "learning_rate": 2.6116592574711835e-06, "loss": 0.84276336, "num_input_tokens_seen": 75119205, "step": 3485, "time_per_iteration": 2.4953436851501465 }, { "auxiliary_loss_clip": 0.01196834, "auxiliary_loss_mlp": 0.01035815, "balance_loss_clip": 1.05861497, "balance_loss_mlp": 1.02788162, "epoch": 0.4191667167678711, "flos": 20741037772320.0, "grad_norm": 1.7203608457647188, "language_loss": 0.84097135, "learning_rate": 2.6109175628968853e-06, "loss": 0.86329782, "num_input_tokens_seen": 75138970, "step": 3486, "time_per_iteration": 2.5080080032348633 }, { "auxiliary_loss_clip": 0.01167314, "auxiliary_loss_mlp": 0.0103433, "balance_loss_clip": 1.05316639, "balance_loss_mlp": 1.02560413, "epoch": 0.4192869596585102, "flos": 23586419808960.0, "grad_norm": 1.9515387019959753, "language_loss": 0.83027077, "learning_rate": 2.610175775646878e-06, "loss": 0.85228717, "num_input_tokens_seen": 75157550, "step": 3487, "time_per_iteration": 2.53588604927063 }, { "auxiliary_loss_clip": 0.01157841, "auxiliary_loss_mlp": 0.01028346, "balance_loss_clip": 1.05078709, "balance_loss_mlp": 1.02043009, "epoch": 0.41940720254914926, "flos": 25081273181760.0, "grad_norm": 3.2339268070024896, "language_loss": 0.7310257, "learning_rate": 2.6094338958336907e-06, "loss": 0.75288761, "num_input_tokens_seen": 75176220, "step": 3488, "time_per_iteration": 2.592481851577759 }, { "auxiliary_loss_clip": 0.01163497, "auxiliary_loss_mlp": 0.01030265, "balance_loss_clip": 1.0557189, "balance_loss_mlp": 1.02264142, "epoch": 0.41952744543978837, "flos": 15554135128320.0, "grad_norm": 1.966449772102665, "language_loss": 0.82432556, "learning_rate": 2.608691923569867e-06, "loss": 0.84626317, "num_input_tokens_seen": 75193095, "step": 3489, "time_per_iteration": 2.514561653137207 }, { "auxiliary_loss_clip": 0.01181433, "auxiliary_loss_mlp": 0.01024785, "balance_loss_clip": 1.05878043, "balance_loss_mlp": 1.01694131, "epoch": 0.4196476883304275, "flos": 24644779192800.0, "grad_norm": 1.5946482251215863, "language_loss": 0.7585513, "learning_rate": 2.6079498589679616e-06, "loss": 0.78061348, "num_input_tokens_seen": 75214185, "step": 3490, "time_per_iteration": 2.5980849266052246 }, { "auxiliary_loss_clip": 0.01113317, "auxiliary_loss_mlp": 0.01037866, "balance_loss_clip": 1.04475689, "balance_loss_mlp": 1.02821553, "epoch": 0.41976793122106654, "flos": 24531336681120.0, "grad_norm": 5.552828395423788, "language_loss": 0.76597881, "learning_rate": 2.6072077021405465e-06, "loss": 0.78749061, "num_input_tokens_seen": 75233020, "step": 3491, "time_per_iteration": 2.658090114593506 }, { "auxiliary_loss_clip": 0.01156364, "auxiliary_loss_mlp": 0.01036411, "balance_loss_clip": 1.05246627, "balance_loss_mlp": 1.02841163, "epoch": 0.41988817411170565, "flos": 21175304909760.0, "grad_norm": 1.7299699653079843, "language_loss": 0.69192952, "learning_rate": 2.6064654532002054e-06, "loss": 0.71385723, "num_input_tokens_seen": 75252030, "step": 3492, "time_per_iteration": 2.600691795349121 }, { "auxiliary_loss_clip": 0.01195163, "auxiliary_loss_mlp": 0.01028805, "balance_loss_clip": 1.05955243, "balance_loss_mlp": 1.02074015, "epoch": 0.42000841700234476, "flos": 31649407102080.0, "grad_norm": 1.477178905298813, "language_loss": 0.75620556, "learning_rate": 2.6057231122595375e-06, "loss": 0.77844524, "num_input_tokens_seen": 75273340, "step": 3493, "time_per_iteration": 2.5956859588623047 }, { "auxiliary_loss_clip": 0.01163288, "auxiliary_loss_mlp": 0.01027672, "balance_loss_clip": 1.05129218, "balance_loss_mlp": 1.01996493, "epoch": 0.4201286598929838, "flos": 21281528112480.0, "grad_norm": 1.5927783378146234, "language_loss": 0.72797233, "learning_rate": 2.604980679431154e-06, "loss": 0.74988186, "num_input_tokens_seen": 75291580, "step": 3494, "time_per_iteration": 2.5723094940185547 }, { "auxiliary_loss_clip": 0.01177419, "auxiliary_loss_mlp": 0.0102808, "balance_loss_clip": 1.05227947, "balance_loss_mlp": 1.02012229, "epoch": 0.4202489027836229, "flos": 18546535645920.0, "grad_norm": 2.8100935727924217, "language_loss": 0.74616766, "learning_rate": 2.604238154827684e-06, "loss": 0.76822263, "num_input_tokens_seen": 75308205, "step": 3495, "time_per_iteration": 2.476644992828369 }, { "auxiliary_loss_clip": 0.01177065, "auxiliary_loss_mlp": 0.01031934, "balance_loss_clip": 1.05482149, "balance_loss_mlp": 1.02428126, "epoch": 0.42036914567426203, "flos": 19317646373280.0, "grad_norm": 1.6608341561556041, "language_loss": 0.7245093, "learning_rate": 2.6034955385617656e-06, "loss": 0.74659926, "num_input_tokens_seen": 75326535, "step": 3496, "time_per_iteration": 2.508732795715332 }, { "auxiliary_loss_clip": 0.01050914, "auxiliary_loss_mlp": 0.0100899, "balance_loss_clip": 1.01662159, "balance_loss_mlp": 1.00770235, "epoch": 0.4204893885649011, "flos": 67842954170880.0, "grad_norm": 0.7269157305699654, "language_loss": 0.61639857, "learning_rate": 2.6027528307460544e-06, "loss": 0.63699758, "num_input_tokens_seen": 75390540, "step": 3497, "time_per_iteration": 4.0150837898254395 }, { "auxiliary_loss_clip": 0.01190148, "auxiliary_loss_mlp": 0.01030877, "balance_loss_clip": 1.05532074, "balance_loss_mlp": 1.02303004, "epoch": 0.4206096314555402, "flos": 21908781051360.0, "grad_norm": 1.8004873045862266, "language_loss": 0.86301494, "learning_rate": 2.602010031493217e-06, "loss": 0.88522524, "num_input_tokens_seen": 75408770, "step": 3498, "time_per_iteration": 3.2594380378723145 }, { "auxiliary_loss_clip": 0.01144825, "auxiliary_loss_mlp": 0.0103009, "balance_loss_clip": 1.05144131, "balance_loss_mlp": 1.02215362, "epoch": 0.42072987434617926, "flos": 29278189557120.0, "grad_norm": 1.9386155898979018, "language_loss": 0.86382562, "learning_rate": 2.6012671409159367e-06, "loss": 0.88557482, "num_input_tokens_seen": 75430105, "step": 3499, "time_per_iteration": 2.6402738094329834 }, { "auxiliary_loss_clip": 0.01158558, "auxiliary_loss_mlp": 0.01032922, "balance_loss_clip": 1.05344081, "balance_loss_mlp": 1.02441669, "epoch": 0.42085011723681837, "flos": 27600730384320.0, "grad_norm": 1.6856564258764701, "language_loss": 0.82001019, "learning_rate": 2.6005241591269097e-06, "loss": 0.84192502, "num_input_tokens_seen": 75449475, "step": 3500, "time_per_iteration": 3.331315517425537 }, { "auxiliary_loss_clip": 0.01143951, "auxiliary_loss_mlp": 0.01025258, "balance_loss_clip": 1.05451238, "balance_loss_mlp": 1.01751518, "epoch": 0.4209703601274575, "flos": 27818636167680.0, "grad_norm": 2.393648839747602, "language_loss": 0.79756927, "learning_rate": 2.5997810862388454e-06, "loss": 0.81926131, "num_input_tokens_seen": 75469315, "step": 3501, "time_per_iteration": 2.6241941452026367 }, { "auxiliary_loss_clip": 0.01163274, "auxiliary_loss_mlp": 0.01027839, "balance_loss_clip": 1.05177093, "balance_loss_mlp": 1.01935172, "epoch": 0.42109060301809653, "flos": 27525533046720.0, "grad_norm": 1.8034500123650674, "language_loss": 0.75558668, "learning_rate": 2.599037922364467e-06, "loss": 0.77749777, "num_input_tokens_seen": 75488215, "step": 3502, "time_per_iteration": 2.5789613723754883 }, { "auxiliary_loss_clip": 0.01141417, "auxiliary_loss_mlp": 0.01028913, "balance_loss_clip": 1.05185163, "balance_loss_mlp": 1.02079511, "epoch": 0.42121084590873564, "flos": 29314279713600.0, "grad_norm": 2.1946741731806076, "language_loss": 0.75126421, "learning_rate": 2.5982946676165112e-06, "loss": 0.77296746, "num_input_tokens_seen": 75507985, "step": 3503, "time_per_iteration": 3.39754056930542 }, { "auxiliary_loss_clip": 0.01045326, "auxiliary_loss_mlp": 0.0100361, "balance_loss_clip": 1.01665688, "balance_loss_mlp": 1.00237024, "epoch": 0.42133108879937475, "flos": 67398845786400.0, "grad_norm": 0.7582089216850949, "language_loss": 0.57635593, "learning_rate": 2.5975513221077313e-06, "loss": 0.59684539, "num_input_tokens_seen": 75571955, "step": 3504, "time_per_iteration": 3.277263879776001 }, { "auxiliary_loss_clip": 0.01153441, "auxiliary_loss_mlp": 0.01037599, "balance_loss_clip": 1.05138397, "balance_loss_mlp": 1.03003573, "epoch": 0.4214513316900138, "flos": 23106041683200.0, "grad_norm": 2.390565623697234, "language_loss": 0.88269305, "learning_rate": 2.5968078859508897e-06, "loss": 0.90460348, "num_input_tokens_seen": 75589155, "step": 3505, "time_per_iteration": 2.5445470809936523 }, { "auxiliary_loss_clip": 0.01177581, "auxiliary_loss_mlp": 0.01025511, "balance_loss_clip": 1.05507183, "balance_loss_mlp": 1.01763749, "epoch": 0.4215715745806529, "flos": 15336193428000.0, "grad_norm": 1.9003062031431508, "language_loss": 0.79746115, "learning_rate": 2.5960643592587673e-06, "loss": 0.81949204, "num_input_tokens_seen": 75606565, "step": 3506, "time_per_iteration": 2.4925031661987305 }, { "auxiliary_loss_clip": 0.01148873, "auxiliary_loss_mlp": 0.0102848, "balance_loss_clip": 1.05311275, "balance_loss_mlp": 1.0201354, "epoch": 0.42169181747129203, "flos": 22127261506080.0, "grad_norm": 3.423260577883868, "language_loss": 0.81297773, "learning_rate": 2.5953207421441553e-06, "loss": 0.83475125, "num_input_tokens_seen": 75625165, "step": 3507, "time_per_iteration": 2.5717530250549316 }, { "auxiliary_loss_clip": 0.01149942, "auxiliary_loss_mlp": 0.01024894, "balance_loss_clip": 1.05175674, "balance_loss_mlp": 1.01682901, "epoch": 0.4218120603619311, "flos": 22630727848800.0, "grad_norm": 2.561203406446309, "language_loss": 0.74919581, "learning_rate": 2.5945770347198603e-06, "loss": 0.77094418, "num_input_tokens_seen": 75643320, "step": 3508, "time_per_iteration": 2.593141555786133 }, { "auxiliary_loss_clip": 0.01157946, "auxiliary_loss_mlp": 0.01030132, "balance_loss_clip": 1.05241752, "balance_loss_mlp": 1.02249098, "epoch": 0.4219323032525702, "flos": 19682822056320.0, "grad_norm": 1.7385297845499572, "language_loss": 0.81858075, "learning_rate": 2.593833237098701e-06, "loss": 0.84046155, "num_input_tokens_seen": 75660920, "step": 3509, "time_per_iteration": 2.528024196624756 }, { "auxiliary_loss_clip": 0.0117408, "auxiliary_loss_mlp": 0.01029526, "balance_loss_clip": 1.05019653, "balance_loss_mlp": 1.02119923, "epoch": 0.4220525461432093, "flos": 30190751636640.0, "grad_norm": 1.6278927405715473, "language_loss": 0.62182945, "learning_rate": 2.593089349393512e-06, "loss": 0.64386547, "num_input_tokens_seen": 75681410, "step": 3510, "time_per_iteration": 2.573979377746582 }, { "auxiliary_loss_clip": 0.01175029, "auxiliary_loss_mlp": 0.01029076, "balance_loss_clip": 1.05491948, "balance_loss_mlp": 1.02090418, "epoch": 0.42217278903384836, "flos": 24315945084960.0, "grad_norm": 2.2927187244640437, "language_loss": 0.84135365, "learning_rate": 2.592345371717141e-06, "loss": 0.86339474, "num_input_tokens_seen": 75700940, "step": 3511, "time_per_iteration": 2.527134895324707 }, { "auxiliary_loss_clip": 0.01179522, "auxiliary_loss_mlp": 0.01036191, "balance_loss_clip": 1.05948925, "balance_loss_mlp": 1.02772689, "epoch": 0.42229303192448747, "flos": 17092477551360.0, "grad_norm": 2.0634449114993743, "language_loss": 0.71848959, "learning_rate": 2.591601304182448e-06, "loss": 0.74064672, "num_input_tokens_seen": 75718910, "step": 3512, "time_per_iteration": 2.499201774597168 }, { "auxiliary_loss_clip": 0.01162621, "auxiliary_loss_mlp": 0.01025706, "balance_loss_clip": 1.05693662, "balance_loss_mlp": 1.01790977, "epoch": 0.4224132748151266, "flos": 22784534635200.0, "grad_norm": 1.7942090734593452, "language_loss": 0.79608428, "learning_rate": 2.5908571469023067e-06, "loss": 0.81796753, "num_input_tokens_seen": 75738395, "step": 3513, "time_per_iteration": 2.54321551322937 }, { "auxiliary_loss_clip": 0.01188521, "auxiliary_loss_mlp": 0.01028596, "balance_loss_clip": 1.05419183, "balance_loss_mlp": 1.02055526, "epoch": 0.42253351770576564, "flos": 17819093553600.0, "grad_norm": 2.967946856495959, "language_loss": 0.75790429, "learning_rate": 2.5901128999896067e-06, "loss": 0.78007543, "num_input_tokens_seen": 75753825, "step": 3514, "time_per_iteration": 2.444397449493408 }, { "auxiliary_loss_clip": 0.01174734, "auxiliary_loss_mlp": 0.0103291, "balance_loss_clip": 1.05463886, "balance_loss_mlp": 1.02547407, "epoch": 0.42265376059640475, "flos": 28512394539840.0, "grad_norm": 1.8436469982486912, "language_loss": 0.68050367, "learning_rate": 2.5893685635572487e-06, "loss": 0.70258009, "num_input_tokens_seen": 75774675, "step": 3515, "time_per_iteration": 2.5456595420837402 }, { "auxiliary_loss_clip": 0.01160105, "auxiliary_loss_mlp": 0.01035098, "balance_loss_clip": 1.05378819, "balance_loss_mlp": 1.0263598, "epoch": 0.4227740034870438, "flos": 16253352878400.0, "grad_norm": 1.945019835664216, "language_loss": 0.69296157, "learning_rate": 2.5886241377181483e-06, "loss": 0.71491361, "num_input_tokens_seen": 75793545, "step": 3516, "time_per_iteration": 2.5088465213775635 }, { "auxiliary_loss_clip": 0.01179948, "auxiliary_loss_mlp": 0.01041781, "balance_loss_clip": 1.05649149, "balance_loss_mlp": 1.0324409, "epoch": 0.4228942463776829, "flos": 25295695020000.0, "grad_norm": 1.9543524765512972, "language_loss": 0.81261218, "learning_rate": 2.587879622585234e-06, "loss": 0.83482951, "num_input_tokens_seen": 75812145, "step": 3517, "time_per_iteration": 2.5301144123077393 }, { "auxiliary_loss_clip": 0.01176257, "auxiliary_loss_mlp": 0.01032683, "balance_loss_clip": 1.05631053, "balance_loss_mlp": 1.02476716, "epoch": 0.423014489268322, "flos": 26395783523040.0, "grad_norm": 2.233916520644883, "language_loss": 0.7591095, "learning_rate": 2.5871350182714486e-06, "loss": 0.78119886, "num_input_tokens_seen": 75833025, "step": 3518, "time_per_iteration": 2.541153907775879 }, { "auxiliary_loss_clip": 0.01187756, "auxiliary_loss_mlp": 0.01030759, "balance_loss_clip": 1.05521393, "balance_loss_mlp": 1.02317691, "epoch": 0.4231347321589611, "flos": 17274005842560.0, "grad_norm": 1.9667531391897144, "language_loss": 0.80230963, "learning_rate": 2.586390324889748e-06, "loss": 0.82449478, "num_input_tokens_seen": 75848925, "step": 3519, "time_per_iteration": 2.441807746887207 }, { "auxiliary_loss_clip": 0.01172315, "auxiliary_loss_mlp": 0.0102958, "balance_loss_clip": 1.05435371, "balance_loss_mlp": 1.02121747, "epoch": 0.4232549750496002, "flos": 22999638895680.0, "grad_norm": 1.7107856391976497, "language_loss": 0.6709277, "learning_rate": 2.5856455425531003e-06, "loss": 0.69294667, "num_input_tokens_seen": 75870400, "step": 3520, "time_per_iteration": 2.5261075496673584 }, { "auxiliary_loss_clip": 0.01173691, "auxiliary_loss_mlp": 0.01031532, "balance_loss_clip": 1.05456173, "balance_loss_mlp": 1.02362275, "epoch": 0.4233752179402393, "flos": 21248347229760.0, "grad_norm": 2.012976917540589, "language_loss": 0.80881399, "learning_rate": 2.5849006713744902e-06, "loss": 0.83086622, "num_input_tokens_seen": 75889195, "step": 3521, "time_per_iteration": 2.4894583225250244 }, { "auxiliary_loss_clip": 0.01160281, "auxiliary_loss_mlp": 0.01035829, "balance_loss_clip": 1.05329514, "balance_loss_mlp": 1.02767515, "epoch": 0.42349546083087836, "flos": 20704301110560.0, "grad_norm": 2.011686479020518, "language_loss": 0.73052996, "learning_rate": 2.5841557114669135e-06, "loss": 0.75249112, "num_input_tokens_seen": 75906055, "step": 3522, "time_per_iteration": 3.2806851863861084 }, { "auxiliary_loss_clip": 0.01193922, "auxiliary_loss_mlp": 0.01033332, "balance_loss_clip": 1.05562568, "balance_loss_mlp": 1.02448666, "epoch": 0.42361570372151747, "flos": 18585068155680.0, "grad_norm": 2.4521124085144548, "language_loss": 0.6733225, "learning_rate": 2.58341066294338e-06, "loss": 0.69559503, "num_input_tokens_seen": 75922720, "step": 3523, "time_per_iteration": 2.453655242919922 }, { "auxiliary_loss_clip": 0.01137488, "auxiliary_loss_mlp": 0.00763575, "balance_loss_clip": 1.05044222, "balance_loss_mlp": 1.00055254, "epoch": 0.4237359466121566, "flos": 20959482310080.0, "grad_norm": 1.985046255120476, "language_loss": 0.85424179, "learning_rate": 2.5826655259169124e-06, "loss": 0.87325239, "num_input_tokens_seen": 75941375, "step": 3524, "time_per_iteration": 3.395341396331787 }, { "auxiliary_loss_clip": 0.0118976, "auxiliary_loss_mlp": 0.01033335, "balance_loss_clip": 1.05755973, "balance_loss_mlp": 1.02574742, "epoch": 0.42385618950279563, "flos": 18038184596640.0, "grad_norm": 1.731792896502607, "language_loss": 0.90307879, "learning_rate": 2.5819203005005475e-06, "loss": 0.92530978, "num_input_tokens_seen": 75958710, "step": 3525, "time_per_iteration": 2.4759578704833984 }, { "auxiliary_loss_clip": 0.01156083, "auxiliary_loss_mlp": 0.01031352, "balance_loss_clip": 1.05404496, "balance_loss_mlp": 1.02383566, "epoch": 0.42397643239343474, "flos": 23769133359840.0, "grad_norm": 1.5031796789367535, "language_loss": 0.78565693, "learning_rate": 2.581174986807336e-06, "loss": 0.80753124, "num_input_tokens_seen": 75978945, "step": 3526, "time_per_iteration": 3.304152250289917 }, { "auxiliary_loss_clip": 0.01164162, "auxiliary_loss_mlp": 0.00762973, "balance_loss_clip": 1.05018973, "balance_loss_mlp": 1.00050902, "epoch": 0.42409667528407385, "flos": 16545091154880.0, "grad_norm": 2.0726246325371847, "language_loss": 0.90700084, "learning_rate": 2.580429584950341e-06, "loss": 0.92627215, "num_input_tokens_seen": 75994695, "step": 3527, "time_per_iteration": 2.4952869415283203 }, { "auxiliary_loss_clip": 0.01150087, "auxiliary_loss_mlp": 0.01029094, "balance_loss_clip": 1.05047822, "balance_loss_mlp": 1.02090168, "epoch": 0.4242169181747129, "flos": 16034190001440.0, "grad_norm": 2.051072005291282, "language_loss": 0.66700768, "learning_rate": 2.5796840950426397e-06, "loss": 0.6887995, "num_input_tokens_seen": 76011780, "step": 3528, "time_per_iteration": 2.5337939262390137 }, { "auxiliary_loss_clip": 0.01166118, "auxiliary_loss_mlp": 0.01031245, "balance_loss_clip": 1.05193281, "balance_loss_mlp": 1.0230968, "epoch": 0.424337161065352, "flos": 20084016061920.0, "grad_norm": 3.4593788304633866, "language_loss": 0.65414578, "learning_rate": 2.578938517197322e-06, "loss": 0.67611945, "num_input_tokens_seen": 76029875, "step": 3529, "time_per_iteration": 3.2990825176239014 }, { "auxiliary_loss_clip": 0.0114923, "auxiliary_loss_mlp": 0.01031489, "balance_loss_clip": 1.04906464, "balance_loss_mlp": 1.02346611, "epoch": 0.4244574039559911, "flos": 23878373587200.0, "grad_norm": 2.325795794700063, "language_loss": 0.62172419, "learning_rate": 2.5781928515274916e-06, "loss": 0.64353138, "num_input_tokens_seen": 76048595, "step": 3530, "time_per_iteration": 2.595426082611084 }, { "auxiliary_loss_clip": 0.01178762, "auxiliary_loss_mlp": 0.01030024, "balance_loss_clip": 1.05679691, "balance_loss_mlp": 1.02204895, "epoch": 0.4245776468466302, "flos": 17565923703840.0, "grad_norm": 1.944871127215842, "language_loss": 0.6806944, "learning_rate": 2.577447098146265e-06, "loss": 0.70278227, "num_input_tokens_seen": 76065770, "step": 3531, "time_per_iteration": 2.4913063049316406 }, { "auxiliary_loss_clip": 0.01147768, "auxiliary_loss_mlp": 0.0102785, "balance_loss_clip": 1.05114341, "balance_loss_mlp": 1.01935637, "epoch": 0.4246978897372693, "flos": 27776260543200.0, "grad_norm": 1.5868034496554382, "language_loss": 0.78955424, "learning_rate": 2.5767012571667724e-06, "loss": 0.81131035, "num_input_tokens_seen": 76085250, "step": 3532, "time_per_iteration": 2.6326496601104736 }, { "auxiliary_loss_clip": 0.01176371, "auxiliary_loss_mlp": 0.01032342, "balance_loss_clip": 1.05247617, "balance_loss_mlp": 1.02400303, "epoch": 0.42481813262790835, "flos": 15596618503680.0, "grad_norm": 1.898774049169108, "language_loss": 0.68286395, "learning_rate": 2.5759553287021587e-06, "loss": 0.70495105, "num_input_tokens_seen": 76103580, "step": 3533, "time_per_iteration": 2.487060308456421 }, { "auxiliary_loss_clip": 0.01158455, "auxiliary_loss_mlp": 0.01038454, "balance_loss_clip": 1.0525701, "balance_loss_mlp": 1.03006768, "epoch": 0.42493837551854746, "flos": 23951092654560.0, "grad_norm": 1.8833797745131953, "language_loss": 0.77751029, "learning_rate": 2.5752093128655786e-06, "loss": 0.79947937, "num_input_tokens_seen": 76121825, "step": 3534, "time_per_iteration": 2.57515549659729 }, { "auxiliary_loss_clip": 0.01152657, "auxiliary_loss_mlp": 0.01027696, "balance_loss_clip": 1.05074358, "balance_loss_mlp": 1.01969075, "epoch": 0.4250586184091866, "flos": 20813469504000.0, "grad_norm": 1.6814755365055902, "language_loss": 0.73821855, "learning_rate": 2.574463209770204e-06, "loss": 0.76002204, "num_input_tokens_seen": 76141140, "step": 3535, "time_per_iteration": 2.539832592010498 }, { "auxiliary_loss_clip": 0.01139297, "auxiliary_loss_mlp": 0.0103045, "balance_loss_clip": 1.04537773, "balance_loss_mlp": 1.02208781, "epoch": 0.42517886129982563, "flos": 30371022834240.0, "grad_norm": 1.591038594391176, "language_loss": 0.79662108, "learning_rate": 2.5737170195292165e-06, "loss": 0.81831855, "num_input_tokens_seen": 76164475, "step": 3536, "time_per_iteration": 2.667537212371826 }, { "auxiliary_loss_clip": 0.01147264, "auxiliary_loss_mlp": 0.01031164, "balance_loss_clip": 1.05023241, "balance_loss_mlp": 1.02320325, "epoch": 0.42529910419046474, "flos": 20080639867680.0, "grad_norm": 1.806203205556563, "language_loss": 0.77776575, "learning_rate": 2.572970742255814e-06, "loss": 0.79955006, "num_input_tokens_seen": 76182965, "step": 3537, "time_per_iteration": 2.5857999324798584 }, { "auxiliary_loss_clip": 0.01173639, "auxiliary_loss_mlp": 0.01027673, "balance_loss_clip": 1.05597425, "balance_loss_mlp": 1.01955175, "epoch": 0.42541934708110385, "flos": 22632451862880.0, "grad_norm": 1.6549207342353294, "language_loss": 0.81720334, "learning_rate": 2.5722243780632046e-06, "loss": 0.83921647, "num_input_tokens_seen": 76201230, "step": 3538, "time_per_iteration": 2.5550448894500732 }, { "auxiliary_loss_clip": 0.01038213, "auxiliary_loss_mlp": 0.01008429, "balance_loss_clip": 1.01423025, "balance_loss_mlp": 1.00711215, "epoch": 0.4255395899717429, "flos": 66200687230560.0, "grad_norm": 0.7658281121790409, "language_loss": 0.60471737, "learning_rate": 2.5714779270646125e-06, "loss": 0.62518382, "num_input_tokens_seen": 76262000, "step": 3539, "time_per_iteration": 3.2129111289978027 }, { "auxiliary_loss_clip": 0.01165349, "auxiliary_loss_mlp": 0.00763571, "balance_loss_clip": 1.05600035, "balance_loss_mlp": 1.00053728, "epoch": 0.425659832862382, "flos": 17931817726080.0, "grad_norm": 2.1546725592543465, "language_loss": 0.77816463, "learning_rate": 2.5707313893732735e-06, "loss": 0.79745376, "num_input_tokens_seen": 76280540, "step": 3540, "time_per_iteration": 2.6987481117248535 }, { "auxiliary_loss_clip": 0.01094682, "auxiliary_loss_mlp": 0.01023388, "balance_loss_clip": 1.04104733, "balance_loss_mlp": 1.01568758, "epoch": 0.4257800757530211, "flos": 24022554628320.0, "grad_norm": 1.7232981029896453, "language_loss": 0.7663821, "learning_rate": 2.5699847651024364e-06, "loss": 0.78756285, "num_input_tokens_seen": 76301180, "step": 3541, "time_per_iteration": 2.793555974960327 }, { "auxiliary_loss_clip": 0.01175789, "auxiliary_loss_mlp": 0.01035573, "balance_loss_clip": 1.05822158, "balance_loss_mlp": 1.02769876, "epoch": 0.4259003186436602, "flos": 23696019205920.0, "grad_norm": 2.0413650613970558, "language_loss": 0.76497316, "learning_rate": 2.5692380543653627e-06, "loss": 0.78708684, "num_input_tokens_seen": 76319335, "step": 3542, "time_per_iteration": 2.541436195373535 }, { "auxiliary_loss_clip": 0.01181433, "auxiliary_loss_mlp": 0.00763265, "balance_loss_clip": 1.05687118, "balance_loss_mlp": 1.00062037, "epoch": 0.4260205615342993, "flos": 15259774913760.0, "grad_norm": 1.8900069884968924, "language_loss": 0.69687271, "learning_rate": 2.5684912572753293e-06, "loss": 0.71631974, "num_input_tokens_seen": 76335010, "step": 3543, "time_per_iteration": 2.4912614822387695 }, { "auxiliary_loss_clip": 0.01186494, "auxiliary_loss_mlp": 0.01027499, "balance_loss_clip": 1.05473042, "balance_loss_mlp": 1.02009594, "epoch": 0.4261408044249384, "flos": 30665311214880.0, "grad_norm": 1.9440552649631149, "language_loss": 0.83855069, "learning_rate": 2.5677443739456245e-06, "loss": 0.86069059, "num_input_tokens_seen": 76356670, "step": 3544, "time_per_iteration": 2.5651867389678955 }, { "auxiliary_loss_clip": 0.01163948, "auxiliary_loss_mlp": 0.01037206, "balance_loss_clip": 1.05545413, "balance_loss_mlp": 1.02886152, "epoch": 0.42626104731557746, "flos": 23257908953760.0, "grad_norm": 3.0445068416770202, "language_loss": 0.79334271, "learning_rate": 2.5669974044895495e-06, "loss": 0.81535435, "num_input_tokens_seen": 76373065, "step": 3545, "time_per_iteration": 2.5595741271972656 }, { "auxiliary_loss_clip": 0.01152844, "auxiliary_loss_mlp": 0.01033057, "balance_loss_clip": 1.0503763, "balance_loss_mlp": 1.02475452, "epoch": 0.42638129020621657, "flos": 25884774618720.0, "grad_norm": 2.133777494686555, "language_loss": 0.79144764, "learning_rate": 2.5662503490204187e-06, "loss": 0.81330669, "num_input_tokens_seen": 76393230, "step": 3546, "time_per_iteration": 2.6182096004486084 }, { "auxiliary_loss_clip": 0.01160248, "auxiliary_loss_mlp": 0.01033427, "balance_loss_clip": 1.05286026, "balance_loss_mlp": 1.02540398, "epoch": 0.4265015330968556, "flos": 26502365895360.0, "grad_norm": 1.7974307606003426, "language_loss": 0.76086497, "learning_rate": 2.5655032076515603e-06, "loss": 0.78280175, "num_input_tokens_seen": 76412555, "step": 3547, "time_per_iteration": 2.5929758548736572 }, { "auxiliary_loss_clip": 0.01163657, "auxiliary_loss_mlp": 0.01029239, "balance_loss_clip": 1.05514741, "balance_loss_mlp": 1.02123427, "epoch": 0.42662177598749473, "flos": 24389526159360.0, "grad_norm": 1.921129183706382, "language_loss": 0.82062018, "learning_rate": 2.5647559804963155e-06, "loss": 0.8425492, "num_input_tokens_seen": 76432485, "step": 3548, "time_per_iteration": 3.3391289710998535 }, { "auxiliary_loss_clip": 0.01142785, "auxiliary_loss_mlp": 0.01025507, "balance_loss_clip": 1.05442619, "balance_loss_mlp": 1.01770437, "epoch": 0.42674201887813384, "flos": 23148632809440.0, "grad_norm": 1.976658841616467, "language_loss": 0.78544426, "learning_rate": 2.5640086676680364e-06, "loss": 0.80712712, "num_input_tokens_seen": 76453980, "step": 3549, "time_per_iteration": 2.6500539779663086 }, { "auxiliary_loss_clip": 0.01175998, "auxiliary_loss_mlp": 0.01028553, "balance_loss_clip": 1.05659366, "balance_loss_mlp": 1.02039933, "epoch": 0.4268622617687729, "flos": 21689618174400.0, "grad_norm": 2.6081811827283468, "language_loss": 0.80947161, "learning_rate": 2.5632612692800923e-06, "loss": 0.8315171, "num_input_tokens_seen": 76473045, "step": 3550, "time_per_iteration": 3.236919641494751 }, { "auxiliary_loss_clip": 0.01150422, "auxiliary_loss_mlp": 0.01031347, "balance_loss_clip": 1.05352926, "balance_loss_mlp": 1.02275753, "epoch": 0.426982504659412, "flos": 23440155584160.0, "grad_norm": 1.9497032499298637, "language_loss": 0.75752306, "learning_rate": 2.5625137854458603e-06, "loss": 0.77934068, "num_input_tokens_seen": 76492060, "step": 3551, "time_per_iteration": 2.5862698554992676 }, { "auxiliary_loss_clip": 0.01163784, "auxiliary_loss_mlp": 0.0102887, "balance_loss_clip": 1.05372322, "balance_loss_mlp": 1.02103209, "epoch": 0.4271027475500511, "flos": 18916560118560.0, "grad_norm": 1.856653491945687, "language_loss": 0.7955687, "learning_rate": 2.561766216278735e-06, "loss": 0.81749529, "num_input_tokens_seen": 76509655, "step": 3552, "time_per_iteration": 3.5990254878997803 }, { "auxiliary_loss_clip": 0.01131374, "auxiliary_loss_mlp": 0.01021851, "balance_loss_clip": 1.04985023, "balance_loss_mlp": 1.01403725, "epoch": 0.4272229904406902, "flos": 26870558603040.0, "grad_norm": 1.8328484812260186, "language_loss": 0.81167805, "learning_rate": 2.561018561892121e-06, "loss": 0.83321035, "num_input_tokens_seen": 76528795, "step": 3553, "time_per_iteration": 2.6384811401367188 }, { "auxiliary_loss_clip": 0.01161574, "auxiliary_loss_mlp": 0.01029669, "balance_loss_clip": 1.05221128, "balance_loss_mlp": 1.02194381, "epoch": 0.4273432333313293, "flos": 23951379990240.0, "grad_norm": 1.6607123421571488, "language_loss": 0.76868856, "learning_rate": 2.5602708223994363e-06, "loss": 0.79060102, "num_input_tokens_seen": 76550660, "step": 3554, "time_per_iteration": 2.5985655784606934 }, { "auxiliary_loss_clip": 0.01149966, "auxiliary_loss_mlp": 0.01028225, "balance_loss_clip": 1.04889822, "balance_loss_mlp": 1.0201602, "epoch": 0.4274634762219684, "flos": 29570358837120.0, "grad_norm": 1.9660912676918845, "language_loss": 0.67829466, "learning_rate": 2.559522997914115e-06, "loss": 0.70007658, "num_input_tokens_seen": 76570240, "step": 3555, "time_per_iteration": 3.3586862087249756 }, { "auxiliary_loss_clip": 0.01190017, "auxiliary_loss_mlp": 0.01025513, "balance_loss_clip": 1.05941582, "balance_loss_mlp": 1.01842642, "epoch": 0.42758371911260745, "flos": 21434149639200.0, "grad_norm": 2.1787302616759474, "language_loss": 0.8449136, "learning_rate": 2.558775088549599e-06, "loss": 0.86706889, "num_input_tokens_seen": 76589820, "step": 3556, "time_per_iteration": 2.5096428394317627 }, { "auxiliary_loss_clip": 0.0118116, "auxiliary_loss_mlp": 0.01027302, "balance_loss_clip": 1.05631673, "balance_loss_mlp": 1.01868963, "epoch": 0.42770396200324656, "flos": 14752824625920.0, "grad_norm": 2.2408990843352004, "language_loss": 0.66625488, "learning_rate": 2.5580270944193467e-06, "loss": 0.68833947, "num_input_tokens_seen": 76606640, "step": 3557, "time_per_iteration": 2.4701790809631348 }, { "auxiliary_loss_clip": 0.01081992, "auxiliary_loss_mlp": 0.01004917, "balance_loss_clip": 1.01732111, "balance_loss_mlp": 1.00355852, "epoch": 0.4278242048938857, "flos": 70654724321280.0, "grad_norm": 0.7752722358548954, "language_loss": 0.55541241, "learning_rate": 2.557279015636827e-06, "loss": 0.57628155, "num_input_tokens_seen": 76667050, "step": 3558, "time_per_iteration": 3.0836875438690186 }, { "auxiliary_loss_clip": 0.01067561, "auxiliary_loss_mlp": 0.01004515, "balance_loss_clip": 1.0167762, "balance_loss_mlp": 1.00316238, "epoch": 0.42794444778452473, "flos": 69366175553760.0, "grad_norm": 0.7694833532971362, "language_loss": 0.61259735, "learning_rate": 2.5565308523155245e-06, "loss": 0.63331813, "num_input_tokens_seen": 76726650, "step": 3559, "time_per_iteration": 3.045123338699341 }, { "auxiliary_loss_clip": 0.01128052, "auxiliary_loss_mlp": 0.01027405, "balance_loss_clip": 1.05084038, "balance_loss_mlp": 1.0194093, "epoch": 0.42806469067516384, "flos": 18215331018720.0, "grad_norm": 2.586431643482118, "language_loss": 0.81456101, "learning_rate": 2.5557826045689336e-06, "loss": 0.8361156, "num_input_tokens_seen": 76742890, "step": 3560, "time_per_iteration": 2.576131820678711 }, { "auxiliary_loss_clip": 0.01049268, "auxiliary_loss_mlp": 0.01003448, "balance_loss_clip": 1.02007341, "balance_loss_mlp": 1.00213718, "epoch": 0.4281849335658029, "flos": 54535825376160.0, "grad_norm": 0.8224523008115633, "language_loss": 0.58828449, "learning_rate": 2.5550342725105643e-06, "loss": 0.60881162, "num_input_tokens_seen": 76801055, "step": 3561, "time_per_iteration": 3.1092183589935303 }, { "auxiliary_loss_clip": 0.01179566, "auxiliary_loss_mlp": 0.01031415, "balance_loss_clip": 1.05989158, "balance_loss_mlp": 1.02309418, "epoch": 0.428305176456442, "flos": 17274831932640.0, "grad_norm": 1.4951279709201148, "language_loss": 0.80730218, "learning_rate": 2.554285856253937e-06, "loss": 0.82941198, "num_input_tokens_seen": 76819890, "step": 3562, "time_per_iteration": 2.5116238594055176 }, { "auxiliary_loss_clip": 0.01163586, "auxiliary_loss_mlp": 0.01027691, "balance_loss_clip": 1.05845523, "balance_loss_mlp": 1.02001429, "epoch": 0.4284254193470811, "flos": 26359513781760.0, "grad_norm": 1.7219819165135837, "language_loss": 0.77323306, "learning_rate": 2.5535373559125855e-06, "loss": 0.79514581, "num_input_tokens_seen": 76840255, "step": 3563, "time_per_iteration": 2.6025805473327637 }, { "auxiliary_loss_clip": 0.01106022, "auxiliary_loss_mlp": 0.01035223, "balance_loss_clip": 1.0478549, "balance_loss_mlp": 1.0265801, "epoch": 0.42854566223772017, "flos": 29714252542560.0, "grad_norm": 1.5127806970459843, "language_loss": 0.81753743, "learning_rate": 2.552788771600057e-06, "loss": 0.83894992, "num_input_tokens_seen": 76860565, "step": 3564, "time_per_iteration": 2.7535316944122314 }, { "auxiliary_loss_clip": 0.01155624, "auxiliary_loss_mlp": 0.01042421, "balance_loss_clip": 1.05849636, "balance_loss_mlp": 1.03469038, "epoch": 0.4286659051283593, "flos": 22018164946560.0, "grad_norm": 2.9214161440323094, "language_loss": 0.82254761, "learning_rate": 2.5520401034299118e-06, "loss": 0.84452808, "num_input_tokens_seen": 76878325, "step": 3565, "time_per_iteration": 2.6002683639526367 }, { "auxiliary_loss_clip": 0.0118241, "auxiliary_loss_mlp": 0.01026438, "balance_loss_clip": 1.05832005, "balance_loss_mlp": 1.01801562, "epoch": 0.4287861480189984, "flos": 13334425684320.0, "grad_norm": 2.0342894841538106, "language_loss": 0.87482655, "learning_rate": 2.551291351515722e-06, "loss": 0.89691496, "num_input_tokens_seen": 76895340, "step": 3566, "time_per_iteration": 2.5371901988983154 }, { "auxiliary_loss_clip": 0.01142206, "auxiliary_loss_mlp": 0.00763698, "balance_loss_clip": 1.04912138, "balance_loss_mlp": 1.00067139, "epoch": 0.42890639090963745, "flos": 26651539393920.0, "grad_norm": 1.6255991873308422, "language_loss": 0.85993552, "learning_rate": 2.5505425159710726e-06, "loss": 0.87899458, "num_input_tokens_seen": 76915150, "step": 3567, "time_per_iteration": 2.6456573009490967 }, { "auxiliary_loss_clip": 0.01167249, "auxiliary_loss_mlp": 0.00763698, "balance_loss_clip": 1.05212498, "balance_loss_mlp": 1.00072575, "epoch": 0.42902663380027656, "flos": 24055771428000.0, "grad_norm": 5.547137580776996, "language_loss": 0.82887965, "learning_rate": 2.549793596909561e-06, "loss": 0.84818912, "num_input_tokens_seen": 76933770, "step": 3568, "time_per_iteration": 2.591066598892212 }, { "auxiliary_loss_clip": 0.01159256, "auxiliary_loss_mlp": 0.01029595, "balance_loss_clip": 1.0549438, "balance_loss_mlp": 1.02175069, "epoch": 0.42914687669091567, "flos": 15632564992320.0, "grad_norm": 1.8960334573908126, "language_loss": 0.66031301, "learning_rate": 2.5490445944447976e-06, "loss": 0.68220162, "num_input_tokens_seen": 76952265, "step": 3569, "time_per_iteration": 2.5327811241149902 }, { "auxiliary_loss_clip": 0.01178373, "auxiliary_loss_mlp": 0.0103082, "balance_loss_clip": 1.05717301, "balance_loss_mlp": 1.02242184, "epoch": 0.4292671195815547, "flos": 31467806976960.0, "grad_norm": 2.172433754179168, "language_loss": 0.65635371, "learning_rate": 2.548295508690406e-06, "loss": 0.67844558, "num_input_tokens_seen": 76973560, "step": 3570, "time_per_iteration": 2.57865834236145 }, { "auxiliary_loss_clip": 0.01178041, "auxiliary_loss_mlp": 0.01027262, "balance_loss_clip": 1.05284095, "balance_loss_mlp": 1.01969779, "epoch": 0.42938736247219383, "flos": 30257759907360.0, "grad_norm": 1.7563474746926642, "language_loss": 0.76493883, "learning_rate": 2.5475463397600217e-06, "loss": 0.78699189, "num_input_tokens_seen": 76993640, "step": 3571, "time_per_iteration": 2.570840358734131 }, { "auxiliary_loss_clip": 0.01196285, "auxiliary_loss_mlp": 0.01025719, "balance_loss_clip": 1.06104219, "balance_loss_mlp": 1.01774943, "epoch": 0.42950760536283294, "flos": 29349687447840.0, "grad_norm": 2.4080726686389964, "language_loss": 0.77425843, "learning_rate": 2.546797087767293e-06, "loss": 0.79647845, "num_input_tokens_seen": 77013765, "step": 3572, "time_per_iteration": 2.522780179977417 }, { "auxiliary_loss_clip": 0.01131239, "auxiliary_loss_mlp": 0.0103102, "balance_loss_clip": 1.05236804, "balance_loss_mlp": 1.02255654, "epoch": 0.429627848253472, "flos": 26869948014720.0, "grad_norm": 1.8401986984662524, "language_loss": 0.87250662, "learning_rate": 2.546047752825881e-06, "loss": 0.89412916, "num_input_tokens_seen": 77034370, "step": 3573, "time_per_iteration": 2.62446665763855 }, { "auxiliary_loss_clip": 0.0114067, "auxiliary_loss_mlp": 0.01027476, "balance_loss_clip": 1.05258656, "balance_loss_mlp": 1.01990056, "epoch": 0.4297480911441111, "flos": 13881273326400.0, "grad_norm": 1.9006343727934283, "language_loss": 0.93143022, "learning_rate": 2.5452983350494595e-06, "loss": 0.95311171, "num_input_tokens_seen": 77049925, "step": 3574, "time_per_iteration": 3.3136065006256104 }, { "auxiliary_loss_clip": 0.01176914, "auxiliary_loss_mlp": 0.00763245, "balance_loss_clip": 1.05631089, "balance_loss_mlp": 1.00085187, "epoch": 0.4298683340347502, "flos": 20741145523200.0, "grad_norm": 2.1277736827601528, "language_loss": 0.65415037, "learning_rate": 2.544548834551713e-06, "loss": 0.67355198, "num_input_tokens_seen": 77068930, "step": 3575, "time_per_iteration": 2.529876232147217 }, { "auxiliary_loss_clip": 0.01141798, "auxiliary_loss_mlp": 0.00763259, "balance_loss_clip": 1.05041194, "balance_loss_mlp": 1.00065982, "epoch": 0.4299885769253893, "flos": 20882129954880.0, "grad_norm": 2.2962382244256165, "language_loss": 0.94268727, "learning_rate": 2.5437992514463424e-06, "loss": 0.96173787, "num_input_tokens_seen": 77082255, "step": 3576, "time_per_iteration": 3.499565362930298 }, { "auxiliary_loss_clip": 0.01176808, "auxiliary_loss_mlp": 0.01032317, "balance_loss_clip": 1.05586314, "balance_loss_mlp": 1.02296543, "epoch": 0.4301088198160284, "flos": 25484622204960.0, "grad_norm": 1.570789838829556, "language_loss": 0.87830031, "learning_rate": 2.5430495858470565e-06, "loss": 0.90039152, "num_input_tokens_seen": 77101725, "step": 3577, "time_per_iteration": 2.5418314933776855 }, { "auxiliary_loss_clip": 0.0117553, "auxiliary_loss_mlp": 0.01031968, "balance_loss_clip": 1.05730593, "balance_loss_mlp": 1.02442193, "epoch": 0.43022906270666744, "flos": 18259430657280.0, "grad_norm": 2.2435323404182244, "language_loss": 0.77231604, "learning_rate": 2.54229983786758e-06, "loss": 0.79439098, "num_input_tokens_seen": 77119670, "step": 3578, "time_per_iteration": 3.192286729812622 }, { "auxiliary_loss_clip": 0.01161052, "auxiliary_loss_mlp": 0.01023586, "balance_loss_clip": 1.05074596, "balance_loss_mlp": 1.01584983, "epoch": 0.43034930559730655, "flos": 23399539890720.0, "grad_norm": 1.8499070439809657, "language_loss": 0.84830141, "learning_rate": 2.541550007621651e-06, "loss": 0.87014782, "num_input_tokens_seen": 77138160, "step": 3579, "time_per_iteration": 2.5979177951812744 }, { "auxiliary_loss_clip": 0.01173771, "auxiliary_loss_mlp": 0.01027553, "balance_loss_clip": 1.05550814, "balance_loss_mlp": 1.01975644, "epoch": 0.43046954848794566, "flos": 28184386522080.0, "grad_norm": 1.7653536098071063, "language_loss": 0.80026567, "learning_rate": 2.5408000952230156e-06, "loss": 0.82227886, "num_input_tokens_seen": 77156950, "step": 3580, "time_per_iteration": 2.566287040710449 }, { "auxiliary_loss_clip": 0.0116046, "auxiliary_loss_mlp": 0.01030916, "balance_loss_clip": 1.05416584, "balance_loss_mlp": 1.02181423, "epoch": 0.4305897913785847, "flos": 28580480319360.0, "grad_norm": 1.9909692078675565, "language_loss": 0.90834105, "learning_rate": 2.5400501007854357e-06, "loss": 0.93025476, "num_input_tokens_seen": 77176395, "step": 3581, "time_per_iteration": 3.3471014499664307 }, { "auxiliary_loss_clip": 0.01129174, "auxiliary_loss_mlp": 0.01028584, "balance_loss_clip": 1.04653823, "balance_loss_mlp": 1.02099609, "epoch": 0.43071003426922383, "flos": 20448724824480.0, "grad_norm": 1.6987731863670257, "language_loss": 0.751562, "learning_rate": 2.539300024422685e-06, "loss": 0.7731396, "num_input_tokens_seen": 77194340, "step": 3582, "time_per_iteration": 2.597339630126953 }, { "auxiliary_loss_clip": 0.010422, "auxiliary_loss_mlp": 0.01005322, "balance_loss_clip": 1.01126647, "balance_loss_mlp": 1.00402892, "epoch": 0.43083027715986294, "flos": 51997978690080.0, "grad_norm": 0.7933577138563569, "language_loss": 0.60963452, "learning_rate": 2.538549866248549e-06, "loss": 0.63010973, "num_input_tokens_seen": 77249320, "step": 3583, "time_per_iteration": 3.0039420127868652 }, { "auxiliary_loss_clip": 0.01177651, "auxiliary_loss_mlp": 0.01034803, "balance_loss_clip": 1.05565906, "balance_loss_mlp": 1.0267446, "epoch": 0.430950520050502, "flos": 16690888459200.0, "grad_norm": 1.953972431175811, "language_loss": 0.81192523, "learning_rate": 2.5377996263768274e-06, "loss": 0.83404982, "num_input_tokens_seen": 77267400, "step": 3584, "time_per_iteration": 2.4821345806121826 }, { "auxiliary_loss_clip": 0.01172263, "auxiliary_loss_mlp": 0.01030182, "balance_loss_clip": 1.05349851, "balance_loss_mlp": 1.02209365, "epoch": 0.4310707629411411, "flos": 24608437617600.0, "grad_norm": 1.6319941296437368, "language_loss": 0.68635571, "learning_rate": 2.5370493049213293e-06, "loss": 0.70838016, "num_input_tokens_seen": 77287045, "step": 3585, "time_per_iteration": 2.6147069931030273 }, { "auxiliary_loss_clip": 0.01088012, "auxiliary_loss_mlp": 0.01029071, "balance_loss_clip": 1.04742515, "balance_loss_mlp": 1.02102447, "epoch": 0.4311910058317802, "flos": 26432987105280.0, "grad_norm": 2.0154129627207604, "language_loss": 0.79964793, "learning_rate": 2.536298901995878e-06, "loss": 0.82081878, "num_input_tokens_seen": 77306255, "step": 3586, "time_per_iteration": 2.810723304748535 }, { "auxiliary_loss_clip": 0.01164978, "auxiliary_loss_mlp": 0.01024979, "balance_loss_clip": 1.05433547, "balance_loss_mlp": 1.01692057, "epoch": 0.43131124872241927, "flos": 25155895848000.0, "grad_norm": 2.2057508086705795, "language_loss": 0.79966265, "learning_rate": 2.535548417714311e-06, "loss": 0.82156217, "num_input_tokens_seen": 77325555, "step": 3587, "time_per_iteration": 2.650636911392212 }, { "auxiliary_loss_clip": 0.01179348, "auxiliary_loss_mlp": 0.01028436, "balance_loss_clip": 1.05545366, "balance_loss_mlp": 1.02003765, "epoch": 0.4314314916130584, "flos": 21614815923360.0, "grad_norm": 1.5893167540593338, "language_loss": 0.87156284, "learning_rate": 2.534797852190474e-06, "loss": 0.89364064, "num_input_tokens_seen": 77345735, "step": 3588, "time_per_iteration": 2.507157564163208 }, { "auxiliary_loss_clip": 0.01172233, "auxiliary_loss_mlp": 0.0102746, "balance_loss_clip": 1.05265212, "balance_loss_mlp": 1.02005672, "epoch": 0.4315517345036975, "flos": 19275019330080.0, "grad_norm": 2.0962026777401706, "language_loss": 0.81779206, "learning_rate": 2.5340472055382283e-06, "loss": 0.83978903, "num_input_tokens_seen": 77361765, "step": 3589, "time_per_iteration": 2.5040135383605957 }, { "auxiliary_loss_clip": 0.01142833, "auxiliary_loss_mlp": 0.01028529, "balance_loss_clip": 1.04642487, "balance_loss_mlp": 1.02027369, "epoch": 0.43167197739433655, "flos": 24273856796160.0, "grad_norm": 2.078179959942327, "language_loss": 0.8137393, "learning_rate": 2.5332964778714468e-06, "loss": 0.83545291, "num_input_tokens_seen": 77378950, "step": 3590, "time_per_iteration": 2.571643590927124 }, { "auxiliary_loss_clip": 0.01145844, "auxiliary_loss_mlp": 0.01029936, "balance_loss_clip": 1.05421591, "balance_loss_mlp": 1.02153158, "epoch": 0.43179222028497566, "flos": 16867819379520.0, "grad_norm": 1.824062534688691, "language_loss": 0.66487324, "learning_rate": 2.5325456693040123e-06, "loss": 0.68663102, "num_input_tokens_seen": 77396145, "step": 3591, "time_per_iteration": 2.559283971786499 }, { "auxiliary_loss_clip": 0.01180355, "auxiliary_loss_mlp": 0.01030541, "balance_loss_clip": 1.05379868, "balance_loss_mlp": 1.02261996, "epoch": 0.43191246317561477, "flos": 17639217442560.0, "grad_norm": 1.8566863382888563, "language_loss": 0.74635452, "learning_rate": 2.531794779949824e-06, "loss": 0.76846349, "num_input_tokens_seen": 77414045, "step": 3592, "time_per_iteration": 2.474621057510376 }, { "auxiliary_loss_clip": 0.01135609, "auxiliary_loss_mlp": 0.01032169, "balance_loss_clip": 1.04889882, "balance_loss_mlp": 1.0240984, "epoch": 0.4320327060662538, "flos": 23878804590720.0, "grad_norm": 1.627771461911872, "language_loss": 0.88266325, "learning_rate": 2.5310438099227903e-06, "loss": 0.90434098, "num_input_tokens_seen": 77431310, "step": 3593, "time_per_iteration": 2.5780553817749023 }, { "auxiliary_loss_clip": 0.01066444, "auxiliary_loss_mlp": 0.01000193, "balance_loss_clip": 1.01110852, "balance_loss_mlp": 0.99892944, "epoch": 0.43215294895689293, "flos": 66394930125600.0, "grad_norm": 0.8024818555479736, "language_loss": 0.53397399, "learning_rate": 2.530292759336833e-06, "loss": 0.55464035, "num_input_tokens_seen": 77492045, "step": 3594, "time_per_iteration": 3.141489028930664 }, { "auxiliary_loss_clip": 0.01159649, "auxiliary_loss_mlp": 0.01034249, "balance_loss_clip": 1.05416346, "balance_loss_mlp": 1.0256238, "epoch": 0.432273191847532, "flos": 20594270710080.0, "grad_norm": 2.2269296697412604, "language_loss": 0.69478762, "learning_rate": 2.5295416283058855e-06, "loss": 0.71672654, "num_input_tokens_seen": 77510910, "step": 3595, "time_per_iteration": 2.5035879611968994 }, { "auxiliary_loss_clip": 0.0115566, "auxiliary_loss_mlp": 0.00763775, "balance_loss_clip": 1.05198824, "balance_loss_mlp": 1.00056982, "epoch": 0.4323934347381711, "flos": 19282130888160.0, "grad_norm": 1.6242079845445865, "language_loss": 0.65808034, "learning_rate": 2.5287904169438943e-06, "loss": 0.67727464, "num_input_tokens_seen": 77530115, "step": 3596, "time_per_iteration": 2.53336763381958 }, { "auxiliary_loss_clip": 0.01112509, "auxiliary_loss_mlp": 0.01035127, "balance_loss_clip": 1.04964662, "balance_loss_mlp": 1.02628803, "epoch": 0.4325136776288102, "flos": 21726318919200.0, "grad_norm": 2.9707593437601933, "language_loss": 0.63794708, "learning_rate": 2.528039125364817e-06, "loss": 0.65942347, "num_input_tokens_seen": 77548920, "step": 3597, "time_per_iteration": 2.657348871231079 }, { "auxiliary_loss_clip": 0.01150472, "auxiliary_loss_mlp": 0.01027424, "balance_loss_clip": 1.05282879, "balance_loss_mlp": 1.01944876, "epoch": 0.43263392051944927, "flos": 22340749503360.0, "grad_norm": 2.334684911696711, "language_loss": 0.76161516, "learning_rate": 2.5272877536826246e-06, "loss": 0.7833941, "num_input_tokens_seen": 77567715, "step": 3598, "time_per_iteration": 2.6027326583862305 }, { "auxiliary_loss_clip": 0.01136242, "auxiliary_loss_mlp": 0.01036377, "balance_loss_clip": 1.04829717, "balance_loss_mlp": 1.02731133, "epoch": 0.4327541634100884, "flos": 29168410575360.0, "grad_norm": 2.3172028044255235, "language_loss": 0.70519716, "learning_rate": 2.5265363020112986e-06, "loss": 0.72692329, "num_input_tokens_seen": 77588035, "step": 3599, "time_per_iteration": 3.4432971477508545 }, { "auxiliary_loss_clip": 0.01175164, "auxiliary_loss_mlp": 0.01024169, "balance_loss_clip": 1.05566657, "balance_loss_mlp": 1.01674807, "epoch": 0.4328744063007275, "flos": 26067452252640.0, "grad_norm": 1.8295592751301706, "language_loss": 0.83697236, "learning_rate": 2.5257847704648344e-06, "loss": 0.85896569, "num_input_tokens_seen": 77609265, "step": 3600, "time_per_iteration": 2.571394205093384 }, { "auxiliary_loss_clip": 0.01188095, "auxiliary_loss_mlp": 0.01034253, "balance_loss_clip": 1.05507922, "balance_loss_mlp": 1.02632856, "epoch": 0.43299464919136654, "flos": 16581360896160.0, "grad_norm": 2.068534500301549, "language_loss": 0.75282097, "learning_rate": 2.525033159157239e-06, "loss": 0.77504444, "num_input_tokens_seen": 77625580, "step": 3601, "time_per_iteration": 2.4423153400421143 }, { "auxiliary_loss_clip": 0.01171251, "auxiliary_loss_mlp": 0.01029392, "balance_loss_clip": 1.05260015, "balance_loss_mlp": 1.02074289, "epoch": 0.43311489208200565, "flos": 16107268238400.0, "grad_norm": 1.7249078971795604, "language_loss": 0.76847255, "learning_rate": 2.52428146820253e-06, "loss": 0.790479, "num_input_tokens_seen": 77643835, "step": 3602, "time_per_iteration": 3.184333324432373 }, { "auxiliary_loss_clip": 0.01151333, "auxiliary_loss_mlp": 0.01029344, "balance_loss_clip": 1.05418539, "balance_loss_mlp": 1.02121997, "epoch": 0.43323513497264476, "flos": 22930224188640.0, "grad_norm": 1.6848844845788844, "language_loss": 0.8186745, "learning_rate": 2.52352969771474e-06, "loss": 0.84048128, "num_input_tokens_seen": 77663060, "step": 3603, "time_per_iteration": 3.347663164138794 }, { "auxiliary_loss_clip": 0.01162107, "auxiliary_loss_mlp": 0.01034125, "balance_loss_clip": 1.05231094, "balance_loss_mlp": 1.02618563, "epoch": 0.4333553778632838, "flos": 25299035297280.0, "grad_norm": 1.958807030199374, "language_loss": 0.88472992, "learning_rate": 2.5227778478079106e-06, "loss": 0.90669221, "num_input_tokens_seen": 77682470, "step": 3604, "time_per_iteration": 2.583347797393799 }, { "auxiliary_loss_clip": 0.01169777, "auxiliary_loss_mlp": 0.01036975, "balance_loss_clip": 1.05123901, "balance_loss_mlp": 1.02912533, "epoch": 0.43347562075392293, "flos": 19387168831200.0, "grad_norm": 1.564238353390979, "language_loss": 0.7673198, "learning_rate": 2.522025918596098e-06, "loss": 0.78938729, "num_input_tokens_seen": 77700770, "step": 3605, "time_per_iteration": 2.511038303375244 }, { "auxiliary_loss_clip": 0.01171587, "auxiliary_loss_mlp": 0.01031476, "balance_loss_clip": 1.05316949, "balance_loss_mlp": 1.02402568, "epoch": 0.43359586364456204, "flos": 26325973729440.0, "grad_norm": 1.4777989742021918, "language_loss": 0.65351826, "learning_rate": 2.521273910193368e-06, "loss": 0.67554891, "num_input_tokens_seen": 77723950, "step": 3606, "time_per_iteration": 2.545258045196533 }, { "auxiliary_loss_clip": 0.01181713, "auxiliary_loss_mlp": 0.01028344, "balance_loss_clip": 1.05605698, "balance_loss_mlp": 1.01999915, "epoch": 0.4337161065352011, "flos": 15989264272800.0, "grad_norm": 2.091649194617168, "language_loss": 0.87189436, "learning_rate": 2.5205218227138006e-06, "loss": 0.89399487, "num_input_tokens_seen": 77736905, "step": 3607, "time_per_iteration": 3.220003366470337 }, { "auxiliary_loss_clip": 0.01190914, "auxiliary_loss_mlp": 0.01030514, "balance_loss_clip": 1.05642474, "balance_loss_mlp": 1.02264047, "epoch": 0.4338363494258402, "flos": 20224713157920.0, "grad_norm": 1.9792023696266723, "language_loss": 0.79168242, "learning_rate": 2.519769656271486e-06, "loss": 0.81389672, "num_input_tokens_seen": 77754325, "step": 3608, "time_per_iteration": 2.455916166305542 }, { "auxiliary_loss_clip": 0.01127297, "auxiliary_loss_mlp": 0.01031485, "balance_loss_clip": 1.05171096, "balance_loss_mlp": 1.02336144, "epoch": 0.43395659231647926, "flos": 20083908311040.0, "grad_norm": 2.017430875429252, "language_loss": 0.67365319, "learning_rate": 2.5190174109805285e-06, "loss": 0.69524097, "num_input_tokens_seen": 77774150, "step": 3609, "time_per_iteration": 2.5831527709960938 }, { "auxiliary_loss_clip": 0.01149424, "auxiliary_loss_mlp": 0.01028587, "balance_loss_clip": 1.04885936, "balance_loss_mlp": 1.02018237, "epoch": 0.43407683520711837, "flos": 19901805348480.0, "grad_norm": 1.740430844865569, "language_loss": 0.63956702, "learning_rate": 2.518265086955042e-06, "loss": 0.66134709, "num_input_tokens_seen": 77791870, "step": 3610, "time_per_iteration": 2.5011863708496094 }, { "auxiliary_loss_clip": 0.01188601, "auxiliary_loss_mlp": 0.01032086, "balance_loss_clip": 1.05428767, "balance_loss_mlp": 1.02391982, "epoch": 0.4341970780977575, "flos": 23108735455200.0, "grad_norm": 1.8905708689322056, "language_loss": 0.83277899, "learning_rate": 2.5175126843091534e-06, "loss": 0.85498583, "num_input_tokens_seen": 77811240, "step": 3611, "time_per_iteration": 2.478611469268799 }, { "auxiliary_loss_clip": 0.01161145, "auxiliary_loss_mlp": 0.01029055, "balance_loss_clip": 1.05346632, "balance_loss_mlp": 1.02103865, "epoch": 0.43431732098839654, "flos": 37408292871840.0, "grad_norm": 2.1725944839078153, "language_loss": 0.75460589, "learning_rate": 2.5167602031570034e-06, "loss": 0.77650785, "num_input_tokens_seen": 77831425, "step": 3612, "time_per_iteration": 2.6497292518615723 }, { "auxiliary_loss_clip": 0.01191705, "auxiliary_loss_mlp": 0.01025641, "balance_loss_clip": 1.05689287, "balance_loss_mlp": 1.01784492, "epoch": 0.43443756387903565, "flos": 31868210809440.0, "grad_norm": 1.6707877467528454, "language_loss": 0.73316067, "learning_rate": 2.51600764361274e-06, "loss": 0.75533414, "num_input_tokens_seen": 77852950, "step": 3613, "time_per_iteration": 2.5555269718170166 }, { "auxiliary_loss_clip": 0.01189952, "auxiliary_loss_mlp": 0.01028106, "balance_loss_clip": 1.05534863, "balance_loss_mlp": 1.01864052, "epoch": 0.43455780676967476, "flos": 23477143664640.0, "grad_norm": 4.1087426668768305, "language_loss": 0.78814054, "learning_rate": 2.5152550057905283e-06, "loss": 0.81032109, "num_input_tokens_seen": 77872840, "step": 3614, "time_per_iteration": 2.487348794937134 }, { "auxiliary_loss_clip": 0.01174598, "auxiliary_loss_mlp": 0.0076417, "balance_loss_clip": 1.05370164, "balance_loss_mlp": 1.00075614, "epoch": 0.4346780496603138, "flos": 24207064027200.0, "grad_norm": 1.9143233940926596, "language_loss": 0.76611614, "learning_rate": 2.5145022898045415e-06, "loss": 0.7855038, "num_input_tokens_seen": 77892025, "step": 3615, "time_per_iteration": 2.535407543182373 }, { "auxiliary_loss_clip": 0.01160417, "auxiliary_loss_mlp": 0.01032404, "balance_loss_clip": 1.04856062, "balance_loss_mlp": 1.02392864, "epoch": 0.4347982925509529, "flos": 17092369800480.0, "grad_norm": 1.9893469627221545, "language_loss": 0.89883244, "learning_rate": 2.5137494957689664e-06, "loss": 0.92076063, "num_input_tokens_seen": 77907635, "step": 3616, "time_per_iteration": 2.4865968227386475 }, { "auxiliary_loss_clip": 0.01053522, "auxiliary_loss_mlp": 0.01001224, "balance_loss_clip": 1.01032329, "balance_loss_mlp": 0.99995416, "epoch": 0.43491853544159204, "flos": 60945555139200.0, "grad_norm": 0.7693438370796853, "language_loss": 0.57368547, "learning_rate": 2.5129966237980016e-06, "loss": 0.59423292, "num_input_tokens_seen": 77970630, "step": 3617, "time_per_iteration": 3.1410789489746094 }, { "auxiliary_loss_clip": 0.01146328, "auxiliary_loss_mlp": 0.01030078, "balance_loss_clip": 1.04885125, "balance_loss_mlp": 1.02158439, "epoch": 0.4350387783322311, "flos": 21944655706080.0, "grad_norm": 1.74283778964856, "language_loss": 0.78194201, "learning_rate": 2.512243674005857e-06, "loss": 0.80370605, "num_input_tokens_seen": 77989995, "step": 3618, "time_per_iteration": 2.5673985481262207 }, { "auxiliary_loss_clip": 0.01112598, "auxiliary_loss_mlp": 0.0102624, "balance_loss_clip": 1.04542494, "balance_loss_mlp": 1.01871145, "epoch": 0.4351590212228702, "flos": 25082709860160.0, "grad_norm": 1.7365125403561645, "language_loss": 0.86230838, "learning_rate": 2.5114906465067537e-06, "loss": 0.88369673, "num_input_tokens_seen": 78010980, "step": 3619, "time_per_iteration": 2.6665852069854736 }, { "auxiliary_loss_clip": 0.01175015, "auxiliary_loss_mlp": 0.01023299, "balance_loss_clip": 1.05133271, "balance_loss_mlp": 1.01547885, "epoch": 0.4352792641135093, "flos": 21506545453920.0, "grad_norm": 2.203621922275279, "language_loss": 0.74910867, "learning_rate": 2.5107375414149264e-06, "loss": 0.77109182, "num_input_tokens_seen": 78030225, "step": 3620, "time_per_iteration": 2.5459823608398438 }, { "auxiliary_loss_clip": 0.01122407, "auxiliary_loss_mlp": 0.01026551, "balance_loss_clip": 1.04493332, "balance_loss_mlp": 1.01817656, "epoch": 0.43539950700414837, "flos": 16253460629280.0, "grad_norm": 3.749078791100014, "language_loss": 0.71263623, "learning_rate": 2.5099843588446197e-06, "loss": 0.73412585, "num_input_tokens_seen": 78048545, "step": 3621, "time_per_iteration": 2.603480100631714 }, { "auxiliary_loss_clip": 0.01134338, "auxiliary_loss_mlp": 0.01031247, "balance_loss_clip": 1.0496738, "balance_loss_mlp": 1.02327812, "epoch": 0.4355197498947875, "flos": 16691822300160.0, "grad_norm": 1.5365575353388345, "language_loss": 0.61383802, "learning_rate": 2.509231098910091e-06, "loss": 0.63549387, "num_input_tokens_seen": 78068415, "step": 3622, "time_per_iteration": 2.6088674068450928 }, { "auxiliary_loss_clip": 0.01155734, "auxiliary_loss_mlp": 0.01027974, "balance_loss_clip": 1.05201948, "balance_loss_mlp": 1.02001715, "epoch": 0.4356399927854266, "flos": 16362736773600.0, "grad_norm": 2.3719368402324137, "language_loss": 0.74564624, "learning_rate": 2.508477761725611e-06, "loss": 0.76748335, "num_input_tokens_seen": 78086690, "step": 3623, "time_per_iteration": 2.505060911178589 }, { "auxiliary_loss_clip": 0.01176235, "auxiliary_loss_mlp": 0.01030599, "balance_loss_clip": 1.05307412, "balance_loss_mlp": 1.02256465, "epoch": 0.43576023567606564, "flos": 17202041031360.0, "grad_norm": 2.008077168696537, "language_loss": 0.8043524, "learning_rate": 2.507724347405458e-06, "loss": 0.82642078, "num_input_tokens_seen": 78104640, "step": 3624, "time_per_iteration": 2.500481128692627 }, { "auxiliary_loss_clip": 0.01123917, "auxiliary_loss_mlp": 0.01025027, "balance_loss_clip": 1.04360509, "balance_loss_mlp": 1.01722443, "epoch": 0.43588047856670475, "flos": 15917658631200.0, "grad_norm": 2.0937003481956835, "language_loss": 0.82180703, "learning_rate": 2.5069708560639243e-06, "loss": 0.84329641, "num_input_tokens_seen": 78122550, "step": 3625, "time_per_iteration": 3.3222248554229736 }, { "auxiliary_loss_clip": 0.01147846, "auxiliary_loss_mlp": 0.01027731, "balance_loss_clip": 1.04969132, "balance_loss_mlp": 1.0194813, "epoch": 0.4360007214573438, "flos": 23659569879840.0, "grad_norm": 1.9777330986288477, "language_loss": 0.61375499, "learning_rate": 2.5062172878153158e-06, "loss": 0.6355108, "num_input_tokens_seen": 78141825, "step": 3626, "time_per_iteration": 2.5856871604919434 }, { "auxiliary_loss_clip": 0.01123272, "auxiliary_loss_mlp": 0.01027382, "balance_loss_clip": 1.04699552, "balance_loss_mlp": 1.01895356, "epoch": 0.4361209643479829, "flos": 21978770429760.0, "grad_norm": 1.7225440466517834, "language_loss": 0.87373126, "learning_rate": 2.505463642773947e-06, "loss": 0.89523768, "num_input_tokens_seen": 78161790, "step": 3627, "time_per_iteration": 2.615217685699463 }, { "auxiliary_loss_clip": 0.01141079, "auxiliary_loss_mlp": 0.00763174, "balance_loss_clip": 1.04735684, "balance_loss_mlp": 1.00075138, "epoch": 0.43624120723862203, "flos": 17420162316480.0, "grad_norm": 2.258214457071004, "language_loss": 0.75201172, "learning_rate": 2.504709921054146e-06, "loss": 0.77105421, "num_input_tokens_seen": 78178605, "step": 3628, "time_per_iteration": 3.244936227798462 }, { "auxiliary_loss_clip": 0.01137529, "auxiliary_loss_mlp": 0.01031026, "balance_loss_clip": 1.04395413, "balance_loss_mlp": 1.02293158, "epoch": 0.4363614501292611, "flos": 17895296566080.0, "grad_norm": 2.7522112550086715, "language_loss": 0.8402158, "learning_rate": 2.50395612277025e-06, "loss": 0.86190128, "num_input_tokens_seen": 78194460, "step": 3629, "time_per_iteration": 3.316763162612915 }, { "auxiliary_loss_clip": 0.01162221, "auxiliary_loss_mlp": 0.01025111, "balance_loss_clip": 1.04931426, "balance_loss_mlp": 1.01729107, "epoch": 0.4364816930199002, "flos": 20302891603200.0, "grad_norm": 1.9362399361310059, "language_loss": 0.72834504, "learning_rate": 2.503202248036612e-06, "loss": 0.75021833, "num_input_tokens_seen": 78213315, "step": 3630, "time_per_iteration": 2.5311436653137207 }, { "auxiliary_loss_clip": 0.01186053, "auxiliary_loss_mlp": 0.01029239, "balance_loss_clip": 1.05433023, "balance_loss_mlp": 1.02104926, "epoch": 0.4366019359105393, "flos": 24061338556800.0, "grad_norm": 1.749805129562833, "language_loss": 0.73585153, "learning_rate": 2.5024482969675927e-06, "loss": 0.75800443, "num_input_tokens_seen": 78233270, "step": 3631, "time_per_iteration": 2.516389846801758 }, { "auxiliary_loss_clip": 0.01132294, "auxiliary_loss_mlp": 0.01023212, "balance_loss_clip": 1.0470767, "balance_loss_mlp": 1.01583326, "epoch": 0.43672217880117836, "flos": 21754112257920.0, "grad_norm": 2.1123391950997785, "language_loss": 0.84826618, "learning_rate": 2.501694269677566e-06, "loss": 0.86982131, "num_input_tokens_seen": 78251040, "step": 3632, "time_per_iteration": 2.5802001953125 }, { "auxiliary_loss_clip": 0.01178323, "auxiliary_loss_mlp": 0.01028911, "balance_loss_clip": 1.05378389, "balance_loss_mlp": 1.02044678, "epoch": 0.4368424216918175, "flos": 18035203488960.0, "grad_norm": 2.6024884115519797, "language_loss": 0.80468011, "learning_rate": 2.500940166280918e-06, "loss": 0.82675248, "num_input_tokens_seen": 78269470, "step": 3633, "time_per_iteration": 3.2653732299804688 }, { "auxiliary_loss_clip": 0.01167835, "auxiliary_loss_mlp": 0.01032019, "balance_loss_clip": 1.04868567, "balance_loss_mlp": 1.02392435, "epoch": 0.4369626645824566, "flos": 25447131287040.0, "grad_norm": 1.8026190974208494, "language_loss": 0.78994805, "learning_rate": 2.500185986892045e-06, "loss": 0.81194669, "num_input_tokens_seen": 78288955, "step": 3634, "time_per_iteration": 2.5414018630981445 }, { "auxiliary_loss_clip": 0.01167777, "auxiliary_loss_mlp": 0.01035015, "balance_loss_clip": 1.05053496, "balance_loss_mlp": 1.02678943, "epoch": 0.43708290747309564, "flos": 25302698827200.0, "grad_norm": 1.9015452157476442, "language_loss": 0.77060699, "learning_rate": 2.499431731625355e-06, "loss": 0.79263496, "num_input_tokens_seen": 78307980, "step": 3635, "time_per_iteration": 2.5422751903533936 }, { "auxiliary_loss_clip": 0.01189979, "auxiliary_loss_mlp": 0.01029924, "balance_loss_clip": 1.0548954, "balance_loss_mlp": 1.02181768, "epoch": 0.43720315036373475, "flos": 31575107688480.0, "grad_norm": 1.779064420210211, "language_loss": 0.79025131, "learning_rate": 2.4986774005952686e-06, "loss": 0.81245035, "num_input_tokens_seen": 78330355, "step": 3636, "time_per_iteration": 2.5781283378601074 }, { "auxiliary_loss_clip": 0.01169815, "auxiliary_loss_mlp": 0.01032811, "balance_loss_clip": 1.05254161, "balance_loss_mlp": 1.0249908, "epoch": 0.43732339325437386, "flos": 23112003898560.0, "grad_norm": 2.324423572877207, "language_loss": 0.84826785, "learning_rate": 2.4979229939162166e-06, "loss": 0.87029409, "num_input_tokens_seen": 78349135, "step": 3637, "time_per_iteration": 2.5232903957366943 }, { "auxiliary_loss_clip": 0.01167989, "auxiliary_loss_mlp": 0.01025474, "balance_loss_clip": 1.05293775, "balance_loss_mlp": 1.01805949, "epoch": 0.4374436361450129, "flos": 27746276269920.0, "grad_norm": 1.529211804714098, "language_loss": 0.80425322, "learning_rate": 2.4971685117026433e-06, "loss": 0.82618785, "num_input_tokens_seen": 78368900, "step": 3638, "time_per_iteration": 2.556856155395508 }, { "auxiliary_loss_clip": 0.01174679, "auxiliary_loss_mlp": 0.01028026, "balance_loss_clip": 1.05373955, "balance_loss_mlp": 1.02011645, "epoch": 0.437563879035652, "flos": 24172374632160.0, "grad_norm": 1.623444231789209, "language_loss": 0.76654649, "learning_rate": 2.4964139540690018e-06, "loss": 0.7885735, "num_input_tokens_seen": 78392235, "step": 3639, "time_per_iteration": 2.5724222660064697 }, { "auxiliary_loss_clip": 0.01146066, "auxiliary_loss_mlp": 0.01033553, "balance_loss_clip": 1.05055785, "balance_loss_mlp": 1.02471399, "epoch": 0.4376841219262911, "flos": 23477215498560.0, "grad_norm": 1.969332981789695, "language_loss": 0.72824013, "learning_rate": 2.495659321129758e-06, "loss": 0.75003636, "num_input_tokens_seen": 78409980, "step": 3640, "time_per_iteration": 2.58358097076416 }, { "auxiliary_loss_clip": 0.01169084, "auxiliary_loss_mlp": 0.01029367, "balance_loss_clip": 1.05016243, "balance_loss_mlp": 1.0215888, "epoch": 0.4378043648169302, "flos": 25447813709280.0, "grad_norm": 1.7643382155430256, "language_loss": 0.75348055, "learning_rate": 2.494904612999389e-06, "loss": 0.77546507, "num_input_tokens_seen": 78428690, "step": 3641, "time_per_iteration": 2.559030055999756 }, { "auxiliary_loss_clip": 0.01064861, "auxiliary_loss_mlp": 0.01004605, "balance_loss_clip": 1.01375353, "balance_loss_mlp": 1.00322819, "epoch": 0.4379246077075693, "flos": 53914067732160.0, "grad_norm": 0.751764372410622, "language_loss": 0.56534594, "learning_rate": 2.4941498297923843e-06, "loss": 0.58604062, "num_input_tokens_seen": 78489260, "step": 3642, "time_per_iteration": 3.077134132385254 }, { "auxiliary_loss_clip": 0.01172211, "auxiliary_loss_mlp": 0.01026832, "balance_loss_clip": 1.05432653, "balance_loss_mlp": 1.0193634, "epoch": 0.43804485059820836, "flos": 20588308494720.0, "grad_norm": 1.6252393363635094, "language_loss": 0.69797736, "learning_rate": 2.4933949716232424e-06, "loss": 0.71996778, "num_input_tokens_seen": 78506785, "step": 3643, "time_per_iteration": 2.5119354724884033 }, { "auxiliary_loss_clip": 0.01144349, "auxiliary_loss_mlp": 0.01025562, "balance_loss_clip": 1.05283689, "balance_loss_mlp": 1.01761055, "epoch": 0.43816509348884747, "flos": 23876218569600.0, "grad_norm": 2.0705740014687426, "language_loss": 0.73692626, "learning_rate": 2.492640038606476e-06, "loss": 0.75862533, "num_input_tokens_seen": 78525150, "step": 3644, "time_per_iteration": 2.5700764656066895 }, { "auxiliary_loss_clip": 0.01171511, "auxiliary_loss_mlp": 0.01031706, "balance_loss_clip": 1.05070817, "balance_loss_mlp": 1.02367759, "epoch": 0.4382853363794866, "flos": 14684451510720.0, "grad_norm": 2.5228619118306748, "language_loss": 0.78597265, "learning_rate": 2.491885030856608e-06, "loss": 0.80800486, "num_input_tokens_seen": 78543245, "step": 3645, "time_per_iteration": 2.474543809890747 }, { "auxiliary_loss_clip": 0.01159738, "auxiliary_loss_mlp": 0.01026794, "balance_loss_clip": 1.05134892, "balance_loss_mlp": 1.01913476, "epoch": 0.43840557927012563, "flos": 17165304369600.0, "grad_norm": 2.0565564586886653, "language_loss": 0.82796019, "learning_rate": 2.4911299484881713e-06, "loss": 0.8498255, "num_input_tokens_seen": 78560775, "step": 3646, "time_per_iteration": 2.4882421493530273 }, { "auxiliary_loss_clip": 0.01151727, "auxiliary_loss_mlp": 0.01027212, "balance_loss_clip": 1.04901624, "balance_loss_mlp": 1.01949906, "epoch": 0.43852582216076474, "flos": 19390688693280.0, "grad_norm": 1.5670901963696116, "language_loss": 0.81203336, "learning_rate": 2.490374791615712e-06, "loss": 0.83382273, "num_input_tokens_seen": 78580800, "step": 3647, "time_per_iteration": 2.5405147075653076 }, { "auxiliary_loss_clip": 0.01194071, "auxiliary_loss_mlp": 0.00763293, "balance_loss_clip": 1.05773473, "balance_loss_mlp": 1.0008018, "epoch": 0.43864606505140386, "flos": 18075136760160.0, "grad_norm": 2.4128555942794705, "language_loss": 0.77823496, "learning_rate": 2.4896195603537867e-06, "loss": 0.79780853, "num_input_tokens_seen": 78595410, "step": 3648, "time_per_iteration": 2.425966262817383 }, { "auxiliary_loss_clip": 0.01125896, "auxiliary_loss_mlp": 0.0103193, "balance_loss_clip": 1.05252504, "balance_loss_mlp": 1.02446425, "epoch": 0.4387663079420429, "flos": 19644900134880.0, "grad_norm": 1.8056527716522215, "language_loss": 0.73981774, "learning_rate": 2.488864254816964e-06, "loss": 0.76139605, "num_input_tokens_seen": 78614100, "step": 3649, "time_per_iteration": 2.586134910583496 }, { "auxiliary_loss_clip": 0.01176966, "auxiliary_loss_mlp": 0.01032056, "balance_loss_clip": 1.05529928, "balance_loss_mlp": 1.02359235, "epoch": 0.438886550832682, "flos": 19719343216320.0, "grad_norm": 2.152168651758217, "language_loss": 0.68537724, "learning_rate": 2.4881088751198218e-06, "loss": 0.7074675, "num_input_tokens_seen": 78632260, "step": 3650, "time_per_iteration": 2.479606866836548 }, { "auxiliary_loss_clip": 0.01161858, "auxiliary_loss_mlp": 0.01033061, "balance_loss_clip": 1.04994106, "balance_loss_mlp": 1.02495503, "epoch": 0.43900679372332113, "flos": 14536678773600.0, "grad_norm": 2.3388967463841595, "language_loss": 0.64626878, "learning_rate": 2.4873534213769517e-06, "loss": 0.66821796, "num_input_tokens_seen": 78647490, "step": 3651, "time_per_iteration": 3.2744462490081787 }, { "auxiliary_loss_clip": 0.01141652, "auxiliary_loss_mlp": 0.01029067, "balance_loss_clip": 1.05354202, "balance_loss_mlp": 1.0211035, "epoch": 0.4391270366139602, "flos": 24056238348480.0, "grad_norm": 1.7503870150606882, "language_loss": 0.72100055, "learning_rate": 2.4865978937029547e-06, "loss": 0.74270779, "num_input_tokens_seen": 78666470, "step": 3652, "time_per_iteration": 2.588153839111328 }, { "auxiliary_loss_clip": 0.01122333, "auxiliary_loss_mlp": 0.0103434, "balance_loss_clip": 1.04947901, "balance_loss_mlp": 1.02570295, "epoch": 0.4392472795045993, "flos": 31538514694560.0, "grad_norm": 1.5146643838890517, "language_loss": 0.66454184, "learning_rate": 2.485842292212445e-06, "loss": 0.68610859, "num_input_tokens_seen": 78687685, "step": 3653, "time_per_iteration": 2.659123420715332 }, { "auxiliary_loss_clip": 0.01191142, "auxiliary_loss_mlp": 0.01030296, "balance_loss_clip": 1.05792117, "balance_loss_mlp": 1.02280927, "epoch": 0.4393675223952384, "flos": 14866303054560.0, "grad_norm": 1.9958326225870047, "language_loss": 0.80128688, "learning_rate": 2.485086617020045e-06, "loss": 0.82350123, "num_input_tokens_seen": 78706180, "step": 3654, "time_per_iteration": 3.209076166152954 }, { "auxiliary_loss_clip": 0.01151162, "auxiliary_loss_mlp": 0.01030999, "balance_loss_clip": 1.04837465, "balance_loss_mlp": 1.02226067, "epoch": 0.43948776528587746, "flos": 14825902862880.0, "grad_norm": 2.120730859298518, "language_loss": 0.81788754, "learning_rate": 2.4843308682403903e-06, "loss": 0.83970904, "num_input_tokens_seen": 78723095, "step": 3655, "time_per_iteration": 3.285090208053589 }, { "auxiliary_loss_clip": 0.0118941, "auxiliary_loss_mlp": 0.01030838, "balance_loss_clip": 1.05538893, "balance_loss_mlp": 1.02348888, "epoch": 0.4396080081765166, "flos": 13914526043040.0, "grad_norm": 1.632075268348236, "language_loss": 0.82461905, "learning_rate": 2.4835750459881294e-06, "loss": 0.84682155, "num_input_tokens_seen": 78739720, "step": 3656, "time_per_iteration": 2.443427085876465 }, { "auxiliary_loss_clip": 0.01152638, "auxiliary_loss_mlp": 0.01029787, "balance_loss_clip": 1.04942369, "balance_loss_mlp": 1.02208018, "epoch": 0.43972825106715563, "flos": 18222981331200.0, "grad_norm": 1.9794024667174361, "language_loss": 0.81812918, "learning_rate": 2.4828191503779177e-06, "loss": 0.83995348, "num_input_tokens_seen": 78757820, "step": 3657, "time_per_iteration": 2.499920129776001 }, { "auxiliary_loss_clip": 0.01146141, "auxiliary_loss_mlp": 0.01026742, "balance_loss_clip": 1.05174685, "balance_loss_mlp": 1.01839757, "epoch": 0.43984849395779474, "flos": 16873242840480.0, "grad_norm": 2.120187314000231, "language_loss": 0.89773345, "learning_rate": 2.482063181524425e-06, "loss": 0.91946232, "num_input_tokens_seen": 78773720, "step": 3658, "time_per_iteration": 2.5309195518493652 }, { "auxiliary_loss_clip": 0.01193484, "auxiliary_loss_mlp": 0.01033949, "balance_loss_clip": 1.05779934, "balance_loss_mlp": 1.02552652, "epoch": 0.43996873684843385, "flos": 18691506860160.0, "grad_norm": 3.007750979855205, "language_loss": 0.81251252, "learning_rate": 2.4813071395423307e-06, "loss": 0.83478689, "num_input_tokens_seen": 78791285, "step": 3659, "time_per_iteration": 3.223951578140259 }, { "auxiliary_loss_clip": 0.01179267, "auxiliary_loss_mlp": 0.01028135, "balance_loss_clip": 1.05608797, "balance_loss_mlp": 1.02013588, "epoch": 0.4400889797390729, "flos": 23653463996640.0, "grad_norm": 1.7226313548571737, "language_loss": 0.64219582, "learning_rate": 2.4805510245463263e-06, "loss": 0.66426986, "num_input_tokens_seen": 78811440, "step": 3660, "time_per_iteration": 2.527672529220581 }, { "auxiliary_loss_clip": 0.01171576, "auxiliary_loss_mlp": 0.010294, "balance_loss_clip": 1.05171943, "balance_loss_mlp": 1.0212996, "epoch": 0.440209222629712, "flos": 23149207480800.0, "grad_norm": 2.4079781669297398, "language_loss": 0.6008364, "learning_rate": 2.4797948366511137e-06, "loss": 0.62284613, "num_input_tokens_seen": 78831150, "step": 3661, "time_per_iteration": 2.5552639961242676 }, { "auxiliary_loss_clip": 0.01143546, "auxiliary_loss_mlp": 0.01036462, "balance_loss_clip": 1.04765761, "balance_loss_mlp": 1.02866912, "epoch": 0.4403294655203511, "flos": 24823398210240.0, "grad_norm": 1.9586986775356026, "language_loss": 0.75835741, "learning_rate": 2.4790385759714055e-06, "loss": 0.78015757, "num_input_tokens_seen": 78850215, "step": 3662, "time_per_iteration": 2.591853618621826 }, { "auxiliary_loss_clip": 0.0117368, "auxiliary_loss_mlp": 0.01027132, "balance_loss_clip": 1.05590045, "balance_loss_mlp": 1.01911473, "epoch": 0.4404497084109902, "flos": 22565084422560.0, "grad_norm": 1.9497446092415798, "language_loss": 0.71005553, "learning_rate": 2.478282242621926e-06, "loss": 0.73206359, "num_input_tokens_seen": 78870675, "step": 3663, "time_per_iteration": 2.526078462600708 }, { "auxiliary_loss_clip": 0.01046542, "auxiliary_loss_mlp": 0.0100379, "balance_loss_clip": 1.01345611, "balance_loss_mlp": 1.0024308, "epoch": 0.4405699513016293, "flos": 64967085023520.0, "grad_norm": 0.8467061251658201, "language_loss": 0.59646487, "learning_rate": 2.477525836717411e-06, "loss": 0.61696815, "num_input_tokens_seen": 78938440, "step": 3664, "time_per_iteration": 3.2790606021881104 }, { "auxiliary_loss_clip": 0.0117571, "auxiliary_loss_mlp": 0.01034283, "balance_loss_clip": 1.05437815, "balance_loss_mlp": 1.02633214, "epoch": 0.4406901941922684, "flos": 35661957746400.0, "grad_norm": 3.3458968693976394, "language_loss": 0.79649544, "learning_rate": 2.476769358372606e-06, "loss": 0.81859541, "num_input_tokens_seen": 78960090, "step": 3665, "time_per_iteration": 2.6395468711853027 }, { "auxiliary_loss_clip": 0.0114406, "auxiliary_loss_mlp": 0.01030451, "balance_loss_clip": 1.05335617, "balance_loss_mlp": 1.02257705, "epoch": 0.44081043708290746, "flos": 18040770617760.0, "grad_norm": 1.995613289142143, "language_loss": 0.74639034, "learning_rate": 2.4760128077022683e-06, "loss": 0.76813543, "num_input_tokens_seen": 78978225, "step": 3666, "time_per_iteration": 2.5253570079803467 }, { "auxiliary_loss_clip": 0.01120113, "auxiliary_loss_mlp": 0.01030892, "balance_loss_clip": 1.04824412, "balance_loss_mlp": 1.02320313, "epoch": 0.44093067997354657, "flos": 30153512137440.0, "grad_norm": 1.4632674233983092, "language_loss": 0.68485379, "learning_rate": 2.4752561848211672e-06, "loss": 0.70636392, "num_input_tokens_seen": 79000625, "step": 3667, "time_per_iteration": 2.6534993648529053 }, { "auxiliary_loss_clip": 0.01170991, "auxiliary_loss_mlp": 0.01029377, "balance_loss_clip": 1.05615151, "balance_loss_mlp": 1.02216208, "epoch": 0.4410509228641857, "flos": 23255071513920.0, "grad_norm": 1.8152251415142906, "language_loss": 0.70981979, "learning_rate": 2.4744994898440797e-06, "loss": 0.73182344, "num_input_tokens_seen": 79019415, "step": 3668, "time_per_iteration": 2.5095298290252686 }, { "auxiliary_loss_clip": 0.01151822, "auxiliary_loss_mlp": 0.01029703, "balance_loss_clip": 1.05148804, "balance_loss_mlp": 1.02216911, "epoch": 0.44117116575482473, "flos": 19500575425920.0, "grad_norm": 2.1637886889456825, "language_loss": 0.83435118, "learning_rate": 2.473742722885797e-06, "loss": 0.85616642, "num_input_tokens_seen": 79038435, "step": 3669, "time_per_iteration": 2.5604560375213623 }, { "auxiliary_loss_clip": 0.01176424, "auxiliary_loss_mlp": 0.00763923, "balance_loss_clip": 1.05678618, "balance_loss_mlp": 1.00069082, "epoch": 0.44129140864546385, "flos": 27053128486080.0, "grad_norm": 1.9795048422641865, "language_loss": 0.64894629, "learning_rate": 2.4729858840611197e-06, "loss": 0.6683498, "num_input_tokens_seen": 79057345, "step": 3670, "time_per_iteration": 2.537641763687134 }, { "auxiliary_loss_clip": 0.01188656, "auxiliary_loss_mlp": 0.0102806, "balance_loss_clip": 1.05694354, "balance_loss_mlp": 1.02085352, "epoch": 0.4414116515361029, "flos": 26102105730720.0, "grad_norm": 1.946665149772741, "language_loss": 0.72703862, "learning_rate": 2.4722289734848605e-06, "loss": 0.74920577, "num_input_tokens_seen": 79077810, "step": 3671, "time_per_iteration": 2.512186288833618 }, { "auxiliary_loss_clip": 0.01143899, "auxiliary_loss_mlp": 0.01031647, "balance_loss_clip": 1.0526582, "balance_loss_mlp": 1.02395773, "epoch": 0.441531894426742, "flos": 21906087279360.0, "grad_norm": 1.9125125501957863, "language_loss": 0.77719676, "learning_rate": 2.471471991271841e-06, "loss": 0.79895222, "num_input_tokens_seen": 79094935, "step": 3672, "time_per_iteration": 2.549072027206421 }, { "auxiliary_loss_clip": 0.01163683, "auxiliary_loss_mlp": 0.01023189, "balance_loss_clip": 1.05004287, "balance_loss_mlp": 1.01542258, "epoch": 0.4416521373173811, "flos": 23437102642560.0, "grad_norm": 1.9120781932324404, "language_loss": 0.79145193, "learning_rate": 2.470714937536896e-06, "loss": 0.8133207, "num_input_tokens_seen": 79113660, "step": 3673, "time_per_iteration": 2.5145795345306396 }, { "auxiliary_loss_clip": 0.01132564, "auxiliary_loss_mlp": 0.0103213, "balance_loss_clip": 1.05091608, "balance_loss_mlp": 1.02382755, "epoch": 0.4417723802080202, "flos": 20334348471840.0, "grad_norm": 2.039964312464061, "language_loss": 0.70381308, "learning_rate": 2.469957812394868e-06, "loss": 0.72546005, "num_input_tokens_seen": 79132470, "step": 3674, "time_per_iteration": 2.577228307723999 }, { "auxiliary_loss_clip": 0.01189142, "auxiliary_loss_mlp": 0.01027059, "balance_loss_clip": 1.05808973, "balance_loss_mlp": 1.01908398, "epoch": 0.4418926230986593, "flos": 18880685463840.0, "grad_norm": 1.7947628497187973, "language_loss": 0.76026815, "learning_rate": 2.4692006159606148e-06, "loss": 0.78243017, "num_input_tokens_seen": 79150000, "step": 3675, "time_per_iteration": 2.467978000640869 }, { "auxiliary_loss_clip": 0.01187391, "auxiliary_loss_mlp": 0.01026515, "balance_loss_clip": 1.05508351, "balance_loss_mlp": 1.01880527, "epoch": 0.4420128659892984, "flos": 19464413435520.0, "grad_norm": 1.6251621329178019, "language_loss": 0.78420997, "learning_rate": 2.468443348349e-06, "loss": 0.80634898, "num_input_tokens_seen": 79167875, "step": 3676, "time_per_iteration": 2.4522743225097656 }, { "auxiliary_loss_clip": 0.0112934, "auxiliary_loss_mlp": 0.01030113, "balance_loss_clip": 1.04911816, "balance_loss_mlp": 1.02142906, "epoch": 0.44213310887993745, "flos": 17894362725120.0, "grad_norm": 2.564793606131374, "language_loss": 0.82684886, "learning_rate": 2.467686009674902e-06, "loss": 0.84844339, "num_input_tokens_seen": 79182325, "step": 3677, "time_per_iteration": 3.3811607360839844 }, { "auxiliary_loss_clip": 0.01170238, "auxiliary_loss_mlp": 0.01037905, "balance_loss_clip": 1.05263638, "balance_loss_mlp": 1.02977467, "epoch": 0.44225335177057656, "flos": 19204670782080.0, "grad_norm": 2.2284042373228385, "language_loss": 0.84977794, "learning_rate": 2.466928600053209e-06, "loss": 0.87185931, "num_input_tokens_seen": 79197630, "step": 3678, "time_per_iteration": 2.4864985942840576 }, { "auxiliary_loss_clip": 0.01154507, "auxiliary_loss_mlp": 0.01030978, "balance_loss_clip": 1.04804206, "balance_loss_mlp": 1.02275836, "epoch": 0.4423735946612157, "flos": 23471325117120.0, "grad_norm": 1.689694264955993, "language_loss": 0.71257782, "learning_rate": 2.466171119598818e-06, "loss": 0.7344327, "num_input_tokens_seen": 79217600, "step": 3679, "time_per_iteration": 2.559741497039795 }, { "auxiliary_loss_clip": 0.01181985, "auxiliary_loss_mlp": 0.01029751, "balance_loss_clip": 1.0537796, "balance_loss_mlp": 1.02073872, "epoch": 0.44249383755185473, "flos": 26685402698880.0, "grad_norm": 1.8025242769921213, "language_loss": 0.77815533, "learning_rate": 2.465413568426639e-06, "loss": 0.8002727, "num_input_tokens_seen": 79238550, "step": 3680, "time_per_iteration": 3.328409194946289 }, { "auxiliary_loss_clip": 0.01164481, "auxiliary_loss_mlp": 0.0102784, "balance_loss_clip": 1.04917705, "balance_loss_mlp": 1.02074087, "epoch": 0.44261408044249384, "flos": 23147663051520.0, "grad_norm": 1.6229302726567971, "language_loss": 0.81147641, "learning_rate": 2.464655946651591e-06, "loss": 0.83339959, "num_input_tokens_seen": 79257555, "step": 3681, "time_per_iteration": 3.23703932762146 }, { "auxiliary_loss_clip": 0.01175231, "auxiliary_loss_mlp": 0.01038251, "balance_loss_clip": 1.05475414, "balance_loss_mlp": 1.02958488, "epoch": 0.44273432333313295, "flos": 24462568479360.0, "grad_norm": 1.9491430951283453, "language_loss": 0.8024165, "learning_rate": 2.4638982543886065e-06, "loss": 0.82455134, "num_input_tokens_seen": 79277595, "step": 3682, "time_per_iteration": 2.53302001953125 }, { "auxiliary_loss_clip": 0.01181905, "auxiliary_loss_mlp": 0.0103832, "balance_loss_clip": 1.05807519, "balance_loss_mlp": 1.03052342, "epoch": 0.442854566223772, "flos": 17528935623360.0, "grad_norm": 2.1002959897482434, "language_loss": 0.87150127, "learning_rate": 2.4631404917526254e-06, "loss": 0.89370346, "num_input_tokens_seen": 79294550, "step": 3683, "time_per_iteration": 2.4840338230133057 }, { "auxiliary_loss_clip": 0.01166795, "auxiliary_loss_mlp": 0.01033716, "balance_loss_clip": 1.05211067, "balance_loss_mlp": 1.02589571, "epoch": 0.4429748091144111, "flos": 24896296862400.0, "grad_norm": 1.5079842215812302, "language_loss": 0.79129004, "learning_rate": 2.4623826588586e-06, "loss": 0.81329513, "num_input_tokens_seen": 79314820, "step": 3684, "time_per_iteration": 2.5340681076049805 }, { "auxiliary_loss_clip": 0.01152086, "auxiliary_loss_mlp": 0.01033289, "balance_loss_clip": 1.04730415, "balance_loss_mlp": 1.02532899, "epoch": 0.4430950520050502, "flos": 21614708172480.0, "grad_norm": 2.293073217924611, "language_loss": 0.82780915, "learning_rate": 2.461624755821492e-06, "loss": 0.8496629, "num_input_tokens_seen": 79334300, "step": 3685, "time_per_iteration": 3.245516777038574 }, { "auxiliary_loss_clip": 0.01145467, "auxiliary_loss_mlp": 0.010316, "balance_loss_clip": 1.0514344, "balance_loss_mlp": 1.02332127, "epoch": 0.4432152948956893, "flos": 24572275627200.0, "grad_norm": 2.066391954595895, "language_loss": 0.76311469, "learning_rate": 2.4608667827562763e-06, "loss": 0.78488535, "num_input_tokens_seen": 79353630, "step": 3686, "time_per_iteration": 2.5776379108428955 }, { "auxiliary_loss_clip": 0.01180712, "auxiliary_loss_mlp": 0.01036976, "balance_loss_clip": 1.05636215, "balance_loss_mlp": 1.02865517, "epoch": 0.4433355377863284, "flos": 21762265407840.0, "grad_norm": 2.112863581670289, "language_loss": 0.89883524, "learning_rate": 2.460108739777936e-06, "loss": 0.9210121, "num_input_tokens_seen": 79372765, "step": 3687, "time_per_iteration": 2.530841112136841 }, { "auxiliary_loss_clip": 0.0115851, "auxiliary_loss_mlp": 0.01025391, "balance_loss_clip": 1.05404902, "balance_loss_mlp": 1.01783884, "epoch": 0.44345578067696745, "flos": 20084482982400.0, "grad_norm": 1.4727852854478323, "language_loss": 0.76183498, "learning_rate": 2.4593506270014656e-06, "loss": 0.783674, "num_input_tokens_seen": 79391735, "step": 3688, "time_per_iteration": 2.5158164501190186 }, { "auxiliary_loss_clip": 0.01161126, "auxiliary_loss_mlp": 0.01025851, "balance_loss_clip": 1.05104923, "balance_loss_mlp": 1.01835918, "epoch": 0.44357602356760656, "flos": 24169501275360.0, "grad_norm": 1.4939798286080883, "language_loss": 0.81867623, "learning_rate": 2.45859244454187e-06, "loss": 0.84054601, "num_input_tokens_seen": 79411525, "step": 3689, "time_per_iteration": 2.5776779651641846 }, { "auxiliary_loss_clip": 0.01171487, "auxiliary_loss_mlp": 0.01029175, "balance_loss_clip": 1.05455518, "balance_loss_mlp": 1.02156901, "epoch": 0.44369626645824567, "flos": 22707721034400.0, "grad_norm": 3.3395311631778317, "language_loss": 0.66261995, "learning_rate": 2.4578341925141655e-06, "loss": 0.68462658, "num_input_tokens_seen": 79430740, "step": 3690, "time_per_iteration": 2.499191999435425 }, { "auxiliary_loss_clip": 0.0117971, "auxiliary_loss_mlp": 0.01033412, "balance_loss_clip": 1.05316436, "balance_loss_mlp": 1.02459633, "epoch": 0.4438165093488847, "flos": 38030230100640.0, "grad_norm": 1.9927632113893434, "language_loss": 0.72398388, "learning_rate": 2.457075871033378e-06, "loss": 0.74611509, "num_input_tokens_seen": 79452615, "step": 3691, "time_per_iteration": 2.66440486907959 }, { "auxiliary_loss_clip": 0.01144278, "auxiliary_loss_mlp": 0.01032564, "balance_loss_clip": 1.05149269, "balance_loss_mlp": 1.02448785, "epoch": 0.44393675223952384, "flos": 15523217014080.0, "grad_norm": 1.9570245541917555, "language_loss": 0.88241565, "learning_rate": 2.4563174802145445e-06, "loss": 0.9041841, "num_input_tokens_seen": 79469865, "step": 3692, "time_per_iteration": 2.543978452682495 }, { "auxiliary_loss_clip": 0.01056941, "auxiliary_loss_mlp": 0.01008184, "balance_loss_clip": 1.01534045, "balance_loss_mlp": 1.00665796, "epoch": 0.44405699513016295, "flos": 64574906174880.0, "grad_norm": 0.6371645517017185, "language_loss": 0.48585647, "learning_rate": 2.455559020172712e-06, "loss": 0.50650775, "num_input_tokens_seen": 79537220, "step": 3693, "time_per_iteration": 3.219404935836792 }, { "auxiliary_loss_clip": 0.01136443, "auxiliary_loss_mlp": 0.01028029, "balance_loss_clip": 1.05516505, "balance_loss_mlp": 1.0203011, "epoch": 0.444177238020802, "flos": 23987398312800.0, "grad_norm": 2.142419149532821, "language_loss": 0.90073287, "learning_rate": 2.4548004910229385e-06, "loss": 0.92237765, "num_input_tokens_seen": 79554795, "step": 3694, "time_per_iteration": 2.620805025100708 }, { "auxiliary_loss_clip": 0.01174368, "auxiliary_loss_mlp": 0.00762559, "balance_loss_clip": 1.05376959, "balance_loss_mlp": 1.00044608, "epoch": 0.4442974809114411, "flos": 22563073072800.0, "grad_norm": 2.1883474003891554, "language_loss": 0.87137198, "learning_rate": 2.4540418928802913e-06, "loss": 0.89074117, "num_input_tokens_seen": 79573530, "step": 3695, "time_per_iteration": 2.6620378494262695 }, { "auxiliary_loss_clip": 0.01159222, "auxiliary_loss_mlp": 0.01036921, "balance_loss_clip": 1.052531, "balance_loss_mlp": 1.02862966, "epoch": 0.4444177238020802, "flos": 17675702685600.0, "grad_norm": 1.9615787355346692, "language_loss": 0.6569097, "learning_rate": 2.4532832258598506e-06, "loss": 0.67887115, "num_input_tokens_seen": 79591360, "step": 3696, "time_per_iteration": 2.513509511947632 }, { "auxiliary_loss_clip": 0.01183599, "auxiliary_loss_mlp": 0.01032999, "balance_loss_clip": 1.05386829, "balance_loss_mlp": 1.02532816, "epoch": 0.4445379666927193, "flos": 28621598850240.0, "grad_norm": 8.14532841467808, "language_loss": 0.80499768, "learning_rate": 2.4525244900767047e-06, "loss": 0.8271637, "num_input_tokens_seen": 79612175, "step": 3697, "time_per_iteration": 2.550191640853882 }, { "auxiliary_loss_clip": 0.01063408, "auxiliary_loss_mlp": 0.01000764, "balance_loss_clip": 1.01533151, "balance_loss_mlp": 0.99940509, "epoch": 0.4446582095833584, "flos": 70487383229280.0, "grad_norm": 0.795765658398874, "language_loss": 0.60494429, "learning_rate": 2.4517656856459536e-06, "loss": 0.62558597, "num_input_tokens_seen": 79678020, "step": 3698, "time_per_iteration": 3.209567070007324 }, { "auxiliary_loss_clip": 0.01173619, "auxiliary_loss_mlp": 0.01032015, "balance_loss_clip": 1.05273736, "balance_loss_mlp": 1.02407885, "epoch": 0.4447784524739975, "flos": 26505203335200.0, "grad_norm": 1.714171812574928, "language_loss": 0.67933679, "learning_rate": 2.4510068126827073e-06, "loss": 0.70139319, "num_input_tokens_seen": 79699020, "step": 3699, "time_per_iteration": 2.566225290298462 }, { "auxiliary_loss_clip": 0.01154922, "auxiliary_loss_mlp": 0.01029295, "balance_loss_clip": 1.05075991, "balance_loss_mlp": 1.02111721, "epoch": 0.44489869536463655, "flos": 11656212255360.0, "grad_norm": 2.0453190759325004, "language_loss": 0.82533735, "learning_rate": 2.450247871302086e-06, "loss": 0.84717953, "num_input_tokens_seen": 79716795, "step": 3700, "time_per_iteration": 2.518721342086792 }, { "auxiliary_loss_clip": 0.01172314, "auxiliary_loss_mlp": 0.01029059, "balance_loss_clip": 1.05155706, "balance_loss_mlp": 1.02150667, "epoch": 0.44501893825527566, "flos": 20448473405760.0, "grad_norm": 2.0278186011777013, "language_loss": 0.83247089, "learning_rate": 2.44948886161922e-06, "loss": 0.85448456, "num_input_tokens_seen": 79735810, "step": 3701, "time_per_iteration": 2.4822821617126465 }, { "auxiliary_loss_clip": 0.01176097, "auxiliary_loss_mlp": 0.01032062, "balance_loss_clip": 1.05382955, "balance_loss_mlp": 1.02429008, "epoch": 0.4451391811459148, "flos": 18261082837440.0, "grad_norm": 1.5221351393079714, "language_loss": 0.84652936, "learning_rate": 2.4487297837492524e-06, "loss": 0.86861104, "num_input_tokens_seen": 79754975, "step": 3702, "time_per_iteration": 2.5224294662475586 }, { "auxiliary_loss_clip": 0.01143297, "auxiliary_loss_mlp": 0.01031471, "balance_loss_clip": 1.04991269, "balance_loss_mlp": 1.02372885, "epoch": 0.44525942403655383, "flos": 16910159087040.0, "grad_norm": 3.0180216721828588, "language_loss": 0.62174356, "learning_rate": 2.4479706378073323e-06, "loss": 0.64349121, "num_input_tokens_seen": 79773515, "step": 3703, "time_per_iteration": 3.3388113975524902 }, { "auxiliary_loss_clip": 0.01132107, "auxiliary_loss_mlp": 0.01040547, "balance_loss_clip": 1.04430091, "balance_loss_mlp": 1.03280449, "epoch": 0.44537966692719294, "flos": 23258842794720.0, "grad_norm": 2.124062427868885, "language_loss": 0.83537787, "learning_rate": 2.447211423908623e-06, "loss": 0.85710442, "num_input_tokens_seen": 79793560, "step": 3704, "time_per_iteration": 2.5912721157073975 }, { "auxiliary_loss_clip": 0.01176066, "auxiliary_loss_mlp": 0.01029083, "balance_loss_clip": 1.05362153, "balance_loss_mlp": 1.02126896, "epoch": 0.445499909817832, "flos": 21724163901600.0, "grad_norm": 1.9618727641290867, "language_loss": 0.75273275, "learning_rate": 2.4464521421682966e-06, "loss": 0.77478421, "num_input_tokens_seen": 79811150, "step": 3705, "time_per_iteration": 2.4806723594665527 }, { "auxiliary_loss_clip": 0.01167921, "auxiliary_loss_mlp": 0.01029232, "balance_loss_clip": 1.05408549, "balance_loss_mlp": 1.02158189, "epoch": 0.4456201527084711, "flos": 23987972984160.0, "grad_norm": 1.382325661876957, "language_loss": 0.87734389, "learning_rate": 2.4456927927015345e-06, "loss": 0.89931536, "num_input_tokens_seen": 79832190, "step": 3706, "time_per_iteration": 3.307880163192749 }, { "auxiliary_loss_clip": 0.01161117, "auxiliary_loss_mlp": 0.01029075, "balance_loss_clip": 1.05121827, "balance_loss_mlp": 1.02059364, "epoch": 0.4457403955991102, "flos": 18807068472480.0, "grad_norm": 3.089021837148152, "language_loss": 0.7608133, "learning_rate": 2.4449333756235307e-06, "loss": 0.7827152, "num_input_tokens_seen": 79848905, "step": 3707, "time_per_iteration": 3.191588878631592 }, { "auxiliary_loss_clip": 0.01173272, "auxiliary_loss_mlp": 0.0103338, "balance_loss_clip": 1.05336726, "balance_loss_mlp": 1.02547026, "epoch": 0.4458606384897493, "flos": 19207759640640.0, "grad_norm": 2.2084585025095125, "language_loss": 0.78381371, "learning_rate": 2.4441738910494876e-06, "loss": 0.80588019, "num_input_tokens_seen": 79863640, "step": 3708, "time_per_iteration": 2.482483386993408 }, { "auxiliary_loss_clip": 0.01163943, "auxiliary_loss_mlp": 0.01028105, "balance_loss_clip": 1.05116808, "balance_loss_mlp": 1.01999605, "epoch": 0.4459808813803884, "flos": 21361286904000.0, "grad_norm": 3.6841451951716016, "language_loss": 0.82392919, "learning_rate": 2.4434143390946176e-06, "loss": 0.84584963, "num_input_tokens_seen": 79882450, "step": 3709, "time_per_iteration": 2.5286355018615723 }, { "auxiliary_loss_clip": 0.01141322, "auxiliary_loss_mlp": 0.01025758, "balance_loss_clip": 1.05089378, "balance_loss_mlp": 1.01869226, "epoch": 0.4461011242710275, "flos": 23288970735840.0, "grad_norm": 2.1924771787717514, "language_loss": 0.85357445, "learning_rate": 2.4426547198741457e-06, "loss": 0.87524521, "num_input_tokens_seen": 79900655, "step": 3710, "time_per_iteration": 2.614628791809082 }, { "auxiliary_loss_clip": 0.01129745, "auxiliary_loss_mlp": 0.01033192, "balance_loss_clip": 1.0529182, "balance_loss_mlp": 1.02506816, "epoch": 0.44622136716166655, "flos": 20193004870560.0, "grad_norm": 2.237558519763843, "language_loss": 0.75071454, "learning_rate": 2.441895033503305e-06, "loss": 0.77234387, "num_input_tokens_seen": 79918575, "step": 3711, "time_per_iteration": 3.352025270462036 }, { "auxiliary_loss_clip": 0.01172296, "auxiliary_loss_mlp": 0.01030794, "balance_loss_clip": 1.0535965, "balance_loss_mlp": 1.02231205, "epoch": 0.44634161005230566, "flos": 21283036624800.0, "grad_norm": 1.7139117657282579, "language_loss": 0.82130444, "learning_rate": 2.4411352800973375e-06, "loss": 0.84333533, "num_input_tokens_seen": 79937010, "step": 3712, "time_per_iteration": 2.493093967437744 }, { "auxiliary_loss_clip": 0.01135296, "auxiliary_loss_mlp": 0.01028962, "balance_loss_clip": 1.0465672, "balance_loss_mlp": 1.02067089, "epoch": 0.44646185294294477, "flos": 22929362181600.0, "grad_norm": 2.4600447884795162, "language_loss": 0.75274241, "learning_rate": 2.4403754597715005e-06, "loss": 0.77438498, "num_input_tokens_seen": 79956455, "step": 3713, "time_per_iteration": 2.589416742324829 }, { "auxiliary_loss_clip": 0.01161356, "auxiliary_loss_mlp": 0.01033539, "balance_loss_clip": 1.04862881, "balance_loss_mlp": 1.02444625, "epoch": 0.4465820958335838, "flos": 22637695739040.0, "grad_norm": 2.3492904907643455, "language_loss": 0.92812908, "learning_rate": 2.4396155726410553e-06, "loss": 0.95007801, "num_input_tokens_seen": 79975065, "step": 3714, "time_per_iteration": 2.548879384994507 }, { "auxiliary_loss_clip": 0.01175331, "auxiliary_loss_mlp": 0.01031005, "balance_loss_clip": 1.0530386, "balance_loss_mlp": 1.02328587, "epoch": 0.44670233872422294, "flos": 22672528801920.0, "grad_norm": 2.698325434901191, "language_loss": 0.90597618, "learning_rate": 2.438855618821278e-06, "loss": 0.92803955, "num_input_tokens_seen": 79990865, "step": 3715, "time_per_iteration": 2.5139968395233154 }, { "auxiliary_loss_clip": 0.01160125, "auxiliary_loss_mlp": 0.0102484, "balance_loss_clip": 1.04718876, "balance_loss_mlp": 1.01718068, "epoch": 0.44682258161486205, "flos": 23582181607680.0, "grad_norm": 1.5476233960192993, "language_loss": 0.6743924, "learning_rate": 2.4380955984274517e-06, "loss": 0.69624203, "num_input_tokens_seen": 80009520, "step": 3716, "time_per_iteration": 2.4927561283111572 }, { "auxiliary_loss_clip": 0.01170951, "auxiliary_loss_mlp": 0.01029861, "balance_loss_clip": 1.0511744, "balance_loss_mlp": 1.02200508, "epoch": 0.4469428245055011, "flos": 26501360220480.0, "grad_norm": 3.087464539260666, "language_loss": 0.76850104, "learning_rate": 2.4373355115748716e-06, "loss": 0.7905091, "num_input_tokens_seen": 80030350, "step": 3717, "time_per_iteration": 2.6014065742492676 }, { "auxiliary_loss_clip": 0.01150308, "auxiliary_loss_mlp": 0.01032916, "balance_loss_clip": 1.04998016, "balance_loss_mlp": 1.02500033, "epoch": 0.4470630673961402, "flos": 21504929190720.0, "grad_norm": 1.726745086373698, "language_loss": 0.71921968, "learning_rate": 2.436575358378842e-06, "loss": 0.74105191, "num_input_tokens_seen": 80049840, "step": 3718, "time_per_iteration": 2.544985055923462 }, { "auxiliary_loss_clip": 0.01165595, "auxiliary_loss_mlp": 0.01031606, "balance_loss_clip": 1.05383694, "balance_loss_mlp": 1.02369702, "epoch": 0.44718331028677927, "flos": 16173989173440.0, "grad_norm": 3.3092312820730765, "language_loss": 0.83100164, "learning_rate": 2.4358151389546782e-06, "loss": 0.85297358, "num_input_tokens_seen": 80066525, "step": 3719, "time_per_iteration": 2.544793128967285 }, { "auxiliary_loss_clip": 0.01184099, "auxiliary_loss_mlp": 0.01027973, "balance_loss_clip": 1.05229223, "balance_loss_mlp": 1.0207901, "epoch": 0.4473035531774184, "flos": 19681241710080.0, "grad_norm": 2.105668932995885, "language_loss": 0.75960302, "learning_rate": 2.4350548534177035e-06, "loss": 0.78172374, "num_input_tokens_seen": 80083355, "step": 3720, "time_per_iteration": 2.4415619373321533 }, { "auxiliary_loss_clip": 0.01142832, "auxiliary_loss_mlp": 0.01029389, "balance_loss_clip": 1.04799485, "balance_loss_mlp": 1.02144933, "epoch": 0.4474237960680575, "flos": 41427595904640.0, "grad_norm": 1.6419923133000542, "language_loss": 0.66532522, "learning_rate": 2.434294501883254e-06, "loss": 0.68704748, "num_input_tokens_seen": 80106450, "step": 3721, "time_per_iteration": 2.747124195098877 }, { "auxiliary_loss_clip": 0.011464, "auxiliary_loss_mlp": 0.01028854, "balance_loss_clip": 1.04714632, "balance_loss_mlp": 1.02086723, "epoch": 0.44754403895869654, "flos": 22891332509280.0, "grad_norm": 1.7617277432026262, "language_loss": 0.65684402, "learning_rate": 2.433534084466674e-06, "loss": 0.67859656, "num_input_tokens_seen": 80125670, "step": 3722, "time_per_iteration": 2.5305118560791016 }, { "auxiliary_loss_clip": 0.01183249, "auxiliary_loss_mlp": 0.01033862, "balance_loss_clip": 1.05381465, "balance_loss_mlp": 1.02630758, "epoch": 0.44766428184933565, "flos": 25630275841440.0, "grad_norm": 1.6715722326899134, "language_loss": 0.70536399, "learning_rate": 2.4327736012833178e-06, "loss": 0.72753513, "num_input_tokens_seen": 80147390, "step": 3723, "time_per_iteration": 2.5403635501861572 }, { "auxiliary_loss_clip": 0.01169192, "auxiliary_loss_mlp": 0.01026563, "balance_loss_clip": 1.05240273, "balance_loss_mlp": 1.01850486, "epoch": 0.44778452473997477, "flos": 20448976243200.0, "grad_norm": 2.662023028137487, "language_loss": 0.76115298, "learning_rate": 2.4320130524485506e-06, "loss": 0.78311056, "num_input_tokens_seen": 80166185, "step": 3724, "time_per_iteration": 2.4828040599823 }, { "auxiliary_loss_clip": 0.01153287, "auxiliary_loss_mlp": 0.01027249, "balance_loss_clip": 1.05482507, "balance_loss_mlp": 1.01921153, "epoch": 0.4479047676306138, "flos": 21975466069440.0, "grad_norm": 2.620896087401512, "language_loss": 0.79803735, "learning_rate": 2.431252438077746e-06, "loss": 0.8198427, "num_input_tokens_seen": 80185685, "step": 3725, "time_per_iteration": 2.5524110794067383 }, { "auxiliary_loss_clip": 0.01175535, "auxiliary_loss_mlp": 0.00762847, "balance_loss_clip": 1.0526613, "balance_loss_mlp": 1.00060439, "epoch": 0.44802501052125293, "flos": 21467222771040.0, "grad_norm": 2.2151471888675953, "language_loss": 0.7675401, "learning_rate": 2.4304917582862906e-06, "loss": 0.78692389, "num_input_tokens_seen": 80204865, "step": 3726, "time_per_iteration": 2.5059566497802734 }, { "auxiliary_loss_clip": 0.01186554, "auxiliary_loss_mlp": 0.01031539, "balance_loss_clip": 1.05539823, "balance_loss_mlp": 1.02379978, "epoch": 0.44814525341189204, "flos": 22126974170400.0, "grad_norm": 1.8282409724074222, "language_loss": 0.8772862, "learning_rate": 2.4297310131895774e-06, "loss": 0.89946717, "num_input_tokens_seen": 80223410, "step": 3727, "time_per_iteration": 2.47912335395813 }, { "auxiliary_loss_clip": 0.01169683, "auxiliary_loss_mlp": 0.01027058, "balance_loss_clip": 1.05161226, "balance_loss_mlp": 1.01870131, "epoch": 0.4482654963025311, "flos": 16653936295680.0, "grad_norm": 2.0208101594387227, "language_loss": 0.7427938, "learning_rate": 2.4289702029030113e-06, "loss": 0.76476121, "num_input_tokens_seen": 80240880, "step": 3728, "time_per_iteration": 2.4588027000427246 }, { "auxiliary_loss_clip": 0.01174112, "auxiliary_loss_mlp": 0.01025505, "balance_loss_clip": 1.05514407, "balance_loss_mlp": 1.01806033, "epoch": 0.4483857391931702, "flos": 18841255030080.0, "grad_norm": 1.7774355338377914, "language_loss": 0.83143342, "learning_rate": 2.4282093275420057e-06, "loss": 0.85342962, "num_input_tokens_seen": 80259910, "step": 3729, "time_per_iteration": 3.313410997390747 }, { "auxiliary_loss_clip": 0.01175048, "auxiliary_loss_mlp": 0.01026746, "balance_loss_clip": 1.05398536, "balance_loss_mlp": 1.01911056, "epoch": 0.4485059820838093, "flos": 20372593645920.0, "grad_norm": 1.9993529455102153, "language_loss": 0.70590979, "learning_rate": 2.4274483872219863e-06, "loss": 0.72792774, "num_input_tokens_seen": 80277270, "step": 3730, "time_per_iteration": 2.479677677154541 }, { "auxiliary_loss_clip": 0.0116848, "auxiliary_loss_mlp": 0.0103537, "balance_loss_clip": 1.05169296, "balance_loss_mlp": 1.02758539, "epoch": 0.4486262249744484, "flos": 20047746320640.0, "grad_norm": 1.8129648507005474, "language_loss": 0.93982613, "learning_rate": 2.426687382058386e-06, "loss": 0.96186459, "num_input_tokens_seen": 80295550, "step": 3731, "time_per_iteration": 2.511077642440796 }, { "auxiliary_loss_clip": 0.01066335, "auxiliary_loss_mlp": 0.01007244, "balance_loss_clip": 1.01749229, "balance_loss_mlp": 1.00580776, "epoch": 0.4487464678650875, "flos": 64595695706400.0, "grad_norm": 0.8693094810946088, "language_loss": 0.59862721, "learning_rate": 2.425926312166649e-06, "loss": 0.61936301, "num_input_tokens_seen": 80348425, "step": 3732, "time_per_iteration": 3.7435808181762695 }, { "auxiliary_loss_clip": 0.01163002, "auxiliary_loss_mlp": 0.01033992, "balance_loss_clip": 1.0534296, "balance_loss_mlp": 1.02516484, "epoch": 0.4488667107557266, "flos": 20769800868960.0, "grad_norm": 2.3933888934965526, "language_loss": 0.73101652, "learning_rate": 2.42516517766223e-06, "loss": 0.75298643, "num_input_tokens_seen": 80366505, "step": 3733, "time_per_iteration": 3.238868474960327 }, { "auxiliary_loss_clip": 0.01187254, "auxiliary_loss_mlp": 0.01024223, "balance_loss_clip": 1.05604017, "balance_loss_mlp": 1.01661181, "epoch": 0.44898695364636565, "flos": 23951739159840.0, "grad_norm": 1.8800713331650845, "language_loss": 0.67938709, "learning_rate": 2.4244039786605907e-06, "loss": 0.70150185, "num_input_tokens_seen": 80387510, "step": 3734, "time_per_iteration": 2.5062644481658936 }, { "auxiliary_loss_clip": 0.01129695, "auxiliary_loss_mlp": 0.01032757, "balance_loss_clip": 1.04552865, "balance_loss_mlp": 1.02507949, "epoch": 0.44910719653700476, "flos": 18624354921600.0, "grad_norm": 2.176211010647258, "language_loss": 0.82329345, "learning_rate": 2.4236427152772055e-06, "loss": 0.84491789, "num_input_tokens_seen": 80405915, "step": 3735, "time_per_iteration": 2.5674006938934326 }, { "auxiliary_loss_clip": 0.01035742, "auxiliary_loss_mlp": 0.01004213, "balance_loss_clip": 1.0158999, "balance_loss_mlp": 1.00275314, "epoch": 0.4492274394276438, "flos": 57033445067040.0, "grad_norm": 0.8215328762464144, "language_loss": 0.57375312, "learning_rate": 2.422881387627557e-06, "loss": 0.59415263, "num_input_tokens_seen": 80458365, "step": 3736, "time_per_iteration": 2.8883748054504395 }, { "auxiliary_loss_clip": 0.01159512, "auxiliary_loss_mlp": 0.0102866, "balance_loss_clip": 1.05447149, "balance_loss_mlp": 1.02108431, "epoch": 0.4493476823182829, "flos": 23254891929120.0, "grad_norm": 1.6606284309441541, "language_loss": 0.77422738, "learning_rate": 2.422119995827139e-06, "loss": 0.79610908, "num_input_tokens_seen": 80478490, "step": 3737, "time_per_iteration": 3.304921865463257 }, { "auxiliary_loss_clip": 0.01174384, "auxiliary_loss_mlp": 0.01032679, "balance_loss_clip": 1.05460978, "balance_loss_mlp": 1.02487051, "epoch": 0.44946792520892204, "flos": 15815134875360.0, "grad_norm": 2.488419712063803, "language_loss": 0.73655713, "learning_rate": 2.4213585399914528e-06, "loss": 0.75862777, "num_input_tokens_seen": 80495695, "step": 3738, "time_per_iteration": 2.4578258991241455 }, { "auxiliary_loss_clip": 0.01171153, "auxiliary_loss_mlp": 0.01029314, "balance_loss_clip": 1.05444193, "balance_loss_mlp": 1.02182436, "epoch": 0.4495881680995611, "flos": 19610067072000.0, "grad_norm": 1.9309070851124162, "language_loss": 0.85545218, "learning_rate": 2.4205970202360113e-06, "loss": 0.87745678, "num_input_tokens_seen": 80515260, "step": 3739, "time_per_iteration": 2.501046895980835 }, { "auxiliary_loss_clip": 0.01121545, "auxiliary_loss_mlp": 0.01029839, "balance_loss_clip": 1.04874897, "balance_loss_mlp": 1.02205443, "epoch": 0.4497084109902002, "flos": 26031469847040.0, "grad_norm": 1.8759261643254066, "language_loss": 0.78415382, "learning_rate": 2.4198354366763354e-06, "loss": 0.80566764, "num_input_tokens_seen": 80533900, "step": 3740, "time_per_iteration": 2.6223671436309814 }, { "auxiliary_loss_clip": 0.01160764, "auxiliary_loss_mlp": 0.0102761, "balance_loss_clip": 1.05217099, "balance_loss_mlp": 1.01970673, "epoch": 0.4498286538808393, "flos": 14793691738080.0, "grad_norm": 1.8894812388967983, "language_loss": 0.78955197, "learning_rate": 2.4190737894279587e-06, "loss": 0.8114357, "num_input_tokens_seen": 80551270, "step": 3741, "time_per_iteration": 2.526520252227783 }, { "auxiliary_loss_clip": 0.01129664, "auxiliary_loss_mlp": 0.01033074, "balance_loss_clip": 1.04343259, "balance_loss_mlp": 1.02530193, "epoch": 0.44994889677147837, "flos": 15450174694080.0, "grad_norm": 2.03074600448943, "language_loss": 0.80329102, "learning_rate": 2.4183120786064203e-06, "loss": 0.82491839, "num_input_tokens_seen": 80568145, "step": 3742, "time_per_iteration": 2.535494804382324 }, { "auxiliary_loss_clip": 0.01175069, "auxiliary_loss_mlp": 0.00762697, "balance_loss_clip": 1.05508566, "balance_loss_mlp": 1.00050735, "epoch": 0.4500691396621175, "flos": 21798319647360.0, "grad_norm": 2.1021486911202683, "language_loss": 0.85160255, "learning_rate": 2.417550304327273e-06, "loss": 0.87098014, "num_input_tokens_seen": 80586185, "step": 3743, "time_per_iteration": 2.5359039306640625 }, { "auxiliary_loss_clip": 0.0119084, "auxiliary_loss_mlp": 0.01031672, "balance_loss_clip": 1.05651593, "balance_loss_mlp": 1.02332139, "epoch": 0.4501893825527566, "flos": 32382021236640.0, "grad_norm": 1.6521102461373809, "language_loss": 0.75827485, "learning_rate": 2.4167884667060763e-06, "loss": 0.78049999, "num_input_tokens_seen": 80608895, "step": 3744, "time_per_iteration": 2.5840706825256348 }, { "auxiliary_loss_clip": 0.01158295, "auxiliary_loss_mlp": 0.01033108, "balance_loss_clip": 1.05043972, "balance_loss_mlp": 1.02463818, "epoch": 0.45030962544339564, "flos": 16544947487040.0, "grad_norm": 2.0358933065518903, "language_loss": 0.87249529, "learning_rate": 2.4160265658584e-06, "loss": 0.8944093, "num_input_tokens_seen": 80623785, "step": 3745, "time_per_iteration": 2.525803327560425 }, { "auxiliary_loss_clip": 0.011766, "auxiliary_loss_mlp": 0.01034299, "balance_loss_clip": 1.05337441, "balance_loss_mlp": 1.0264405, "epoch": 0.45042986833403476, "flos": 19573043074560.0, "grad_norm": 2.0824131365699663, "language_loss": 0.68418193, "learning_rate": 2.4152646018998253e-06, "loss": 0.70629096, "num_input_tokens_seen": 80642735, "step": 3746, "time_per_iteration": 2.4825761318206787 }, { "auxiliary_loss_clip": 0.01170044, "auxiliary_loss_mlp": 0.01031043, "balance_loss_clip": 1.05333567, "balance_loss_mlp": 1.02267122, "epoch": 0.45055011122467387, "flos": 23112470819040.0, "grad_norm": 3.8699141764551745, "language_loss": 0.71947026, "learning_rate": 2.4145025749459403e-06, "loss": 0.74148113, "num_input_tokens_seen": 80663760, "step": 3747, "time_per_iteration": 2.5538973808288574 }, { "auxiliary_loss_clip": 0.01101686, "auxiliary_loss_mlp": 0.01033268, "balance_loss_clip": 1.04660988, "balance_loss_mlp": 1.02509046, "epoch": 0.4506703541153129, "flos": 19934627061600.0, "grad_norm": 2.030827431595004, "language_loss": 0.70123821, "learning_rate": 2.413740485112344e-06, "loss": 0.72258782, "num_input_tokens_seen": 80682100, "step": 3748, "time_per_iteration": 2.622309923171997 }, { "auxiliary_loss_clip": 0.01152749, "auxiliary_loss_mlp": 0.01030565, "balance_loss_clip": 1.05224073, "balance_loss_mlp": 1.02333212, "epoch": 0.45079059700595203, "flos": 19499533834080.0, "grad_norm": 1.5266455698070631, "language_loss": 0.82000911, "learning_rate": 2.412978332514646e-06, "loss": 0.84184229, "num_input_tokens_seen": 80700880, "step": 3749, "time_per_iteration": 2.5428216457366943 }, { "auxiliary_loss_clip": 0.01161357, "auxiliary_loss_mlp": 0.01034686, "balance_loss_clip": 1.05214524, "balance_loss_mlp": 1.02607322, "epoch": 0.4509108398965911, "flos": 27636317703360.0, "grad_norm": 1.8841914229146632, "language_loss": 0.7211684, "learning_rate": 2.4122161172684623e-06, "loss": 0.7431289, "num_input_tokens_seen": 80721675, "step": 3750, "time_per_iteration": 2.575498580932617 }, { "auxiliary_loss_clip": 0.01157433, "auxiliary_loss_mlp": 0.01027083, "balance_loss_clip": 1.05069244, "balance_loss_mlp": 1.01940572, "epoch": 0.4510310827872302, "flos": 20995716134400.0, "grad_norm": 2.008935982460064, "language_loss": 0.83840042, "learning_rate": 2.4114538394894216e-06, "loss": 0.86024559, "num_input_tokens_seen": 80739315, "step": 3751, "time_per_iteration": 2.5329716205596924 }, { "auxiliary_loss_clip": 0.01155013, "auxiliary_loss_mlp": 0.01024906, "balance_loss_clip": 1.04828334, "balance_loss_mlp": 1.01778352, "epoch": 0.4511513256778693, "flos": 16216688050560.0, "grad_norm": 2.1418352016589934, "language_loss": 0.83177066, "learning_rate": 2.410691499293161e-06, "loss": 0.85356981, "num_input_tokens_seen": 80757470, "step": 3752, "time_per_iteration": 2.5173444747924805 }, { "auxiliary_loss_clip": 0.01170086, "auxiliary_loss_mlp": 0.01026767, "balance_loss_clip": 1.05262375, "balance_loss_mlp": 1.01924503, "epoch": 0.45127156856850836, "flos": 25186706211360.0, "grad_norm": 1.5460683372873605, "language_loss": 0.74171376, "learning_rate": 2.409929096795326e-06, "loss": 0.76368237, "num_input_tokens_seen": 80777840, "step": 3753, "time_per_iteration": 2.567206621170044 }, { "auxiliary_loss_clip": 0.01172213, "auxiliary_loss_mlp": 0.01030939, "balance_loss_clip": 1.05196166, "balance_loss_mlp": 1.02290988, "epoch": 0.4513918114591475, "flos": 20412526917120.0, "grad_norm": 1.8380023737521043, "language_loss": 0.7907753, "learning_rate": 2.409166632111573e-06, "loss": 0.81280684, "num_input_tokens_seen": 80795975, "step": 3754, "time_per_iteration": 2.4905598163604736 }, { "auxiliary_loss_clip": 0.01179636, "auxiliary_loss_mlp": 0.01030115, "balance_loss_clip": 1.05303717, "balance_loss_mlp": 1.021487, "epoch": 0.4515120543497866, "flos": 26648486452320.0, "grad_norm": 1.9811209473620368, "language_loss": 0.80214977, "learning_rate": 2.4084041053575674e-06, "loss": 0.82424736, "num_input_tokens_seen": 80815395, "step": 3755, "time_per_iteration": 3.389118194580078 }, { "auxiliary_loss_clip": 0.01162826, "auxiliary_loss_mlp": 0.01034688, "balance_loss_clip": 1.05440974, "balance_loss_mlp": 1.02645063, "epoch": 0.45163229724042564, "flos": 20595096800160.0, "grad_norm": 1.8911917979024646, "language_loss": 0.72515154, "learning_rate": 2.4076415166489834e-06, "loss": 0.74712664, "num_input_tokens_seen": 80834805, "step": 3756, "time_per_iteration": 2.523493528366089 }, { "auxiliary_loss_clip": 0.01133074, "auxiliary_loss_mlp": 0.0102122, "balance_loss_clip": 1.04858387, "balance_loss_mlp": 1.01376355, "epoch": 0.45175254013106475, "flos": 21689007586080.0, "grad_norm": 1.5925253394113665, "language_loss": 0.78929532, "learning_rate": 2.406878866101506e-06, "loss": 0.81083822, "num_input_tokens_seen": 80853770, "step": 3757, "time_per_iteration": 2.589841604232788 }, { "auxiliary_loss_clip": 0.0118843, "auxiliary_loss_mlp": 0.01037197, "balance_loss_clip": 1.05761623, "balance_loss_mlp": 1.02933478, "epoch": 0.45187278302170386, "flos": 18878853698880.0, "grad_norm": 1.9216324712094939, "language_loss": 0.78052211, "learning_rate": 2.4061161538308273e-06, "loss": 0.80277836, "num_input_tokens_seen": 80870615, "step": 3758, "time_per_iteration": 3.19594407081604 }, { "auxiliary_loss_clip": 0.0117063, "auxiliary_loss_mlp": 0.01024327, "balance_loss_clip": 1.05271626, "balance_loss_mlp": 1.01707935, "epoch": 0.4519930259123429, "flos": 18582482134560.0, "grad_norm": 2.3332098089813833, "language_loss": 0.89184535, "learning_rate": 2.4053533799526523e-06, "loss": 0.91379488, "num_input_tokens_seen": 80886335, "step": 3759, "time_per_iteration": 3.2047319412231445 }, { "auxiliary_loss_clip": 0.01152238, "auxiliary_loss_mlp": 0.01026771, "balance_loss_clip": 1.05188179, "balance_loss_mlp": 1.0193119, "epoch": 0.452113268802982, "flos": 25192381091040.0, "grad_norm": 1.607267508632131, "language_loss": 0.86197817, "learning_rate": 2.404590544582691e-06, "loss": 0.8837682, "num_input_tokens_seen": 80904570, "step": 3760, "time_per_iteration": 2.575523614883423 }, { "auxiliary_loss_clip": 0.01131637, "auxiliary_loss_mlp": 0.01024631, "balance_loss_clip": 1.04362178, "balance_loss_mlp": 1.01655436, "epoch": 0.45223351169362114, "flos": 39378926999520.0, "grad_norm": 1.56709364559215, "language_loss": 0.81377435, "learning_rate": 2.403827647836666e-06, "loss": 0.83533704, "num_input_tokens_seen": 80925125, "step": 3761, "time_per_iteration": 2.747333526611328 }, { "auxiliary_loss_clip": 0.01186494, "auxiliary_loss_mlp": 0.01031248, "balance_loss_clip": 1.05312288, "balance_loss_mlp": 1.02382374, "epoch": 0.4523537545842602, "flos": 21582173795040.0, "grad_norm": 1.797135660855374, "language_loss": 0.6963582, "learning_rate": 2.4030646898303075e-06, "loss": 0.7185356, "num_input_tokens_seen": 80946615, "step": 3762, "time_per_iteration": 2.5019943714141846 }, { "auxiliary_loss_clip": 0.01162028, "auxiliary_loss_mlp": 0.01029625, "balance_loss_clip": 1.05101216, "balance_loss_mlp": 1.02115512, "epoch": 0.4524739974748993, "flos": 28439531804640.0, "grad_norm": 2.358332321744699, "language_loss": 0.81899148, "learning_rate": 2.4023016706793566e-06, "loss": 0.84090805, "num_input_tokens_seen": 80966410, "step": 3763, "time_per_iteration": 3.316052198410034 }, { "auxiliary_loss_clip": 0.0104609, "auxiliary_loss_mlp": 0.01005313, "balance_loss_clip": 1.00994337, "balance_loss_mlp": 1.00401926, "epoch": 0.4525942403655384, "flos": 61556501778240.0, "grad_norm": 0.7671314987240639, "language_loss": 0.56889009, "learning_rate": 2.401538590499561e-06, "loss": 0.58940411, "num_input_tokens_seen": 81026865, "step": 3764, "time_per_iteration": 3.208867073059082 }, { "auxiliary_loss_clip": 0.01178294, "auxiliary_loss_mlp": 0.00762671, "balance_loss_clip": 1.05612993, "balance_loss_mlp": 1.0004797, "epoch": 0.45271448325617747, "flos": 27529843081920.0, "grad_norm": 1.7565546146380897, "language_loss": 0.71651918, "learning_rate": 2.400775449406682e-06, "loss": 0.73592877, "num_input_tokens_seen": 81050060, "step": 3765, "time_per_iteration": 2.598442316055298 }, { "auxiliary_loss_clip": 0.01168788, "auxiliary_loss_mlp": 0.01030155, "balance_loss_clip": 1.04905224, "balance_loss_mlp": 1.02237701, "epoch": 0.4528347261468166, "flos": 22452611668800.0, "grad_norm": 1.6711777921548536, "language_loss": 0.72810292, "learning_rate": 2.400012247516485e-06, "loss": 0.75009239, "num_input_tokens_seen": 81070625, "step": 3766, "time_per_iteration": 2.5437891483306885 }, { "auxiliary_loss_clip": 0.01145069, "auxiliary_loss_mlp": 0.01026849, "balance_loss_clip": 1.04773164, "balance_loss_mlp": 1.01945186, "epoch": 0.45295496903745563, "flos": 21103878852960.0, "grad_norm": 1.6259168795465488, "language_loss": 0.89928913, "learning_rate": 2.3992489849447484e-06, "loss": 0.92100823, "num_input_tokens_seen": 81089080, "step": 3767, "time_per_iteration": 2.5904529094696045 }, { "auxiliary_loss_clip": 0.01148073, "auxiliary_loss_mlp": 0.01025574, "balance_loss_clip": 1.05048418, "balance_loss_mlp": 1.01799512, "epoch": 0.45307521192809475, "flos": 23221172292000.0, "grad_norm": 1.6947190327070363, "language_loss": 0.79090548, "learning_rate": 2.3984856618072584e-06, "loss": 0.81264192, "num_input_tokens_seen": 81109115, "step": 3768, "time_per_iteration": 2.574721097946167 }, { "auxiliary_loss_clip": 0.01146725, "auxiliary_loss_mlp": 0.01032185, "balance_loss_clip": 1.04968023, "balance_loss_mlp": 1.0249548, "epoch": 0.45319545481873386, "flos": 15560097343680.0, "grad_norm": 1.9757741781312468, "language_loss": 0.73951042, "learning_rate": 2.3977222782198098e-06, "loss": 0.76129949, "num_input_tokens_seen": 81127750, "step": 3769, "time_per_iteration": 2.5608205795288086 }, { "auxiliary_loss_clip": 0.0113258, "auxiliary_loss_mlp": 0.0103204, "balance_loss_clip": 1.04509711, "balance_loss_mlp": 1.02367163, "epoch": 0.4533156977093729, "flos": 21944763456960.0, "grad_norm": 1.62185919602432, "language_loss": 0.75288588, "learning_rate": 2.3969588342982077e-06, "loss": 0.77453208, "num_input_tokens_seen": 81147125, "step": 3770, "time_per_iteration": 2.574751615524292 }, { "auxiliary_loss_clip": 0.01171504, "auxiliary_loss_mlp": 0.01028752, "balance_loss_clip": 1.05534482, "balance_loss_mlp": 1.01950717, "epoch": 0.453435940600012, "flos": 24242184425760.0, "grad_norm": 1.6809654087842887, "language_loss": 0.7299546, "learning_rate": 2.396195330158267e-06, "loss": 0.75195718, "num_input_tokens_seen": 81167015, "step": 3771, "time_per_iteration": 2.5452585220336914 }, { "auxiliary_loss_clip": 0.01187615, "auxiliary_loss_mlp": 0.0102923, "balance_loss_clip": 1.05481446, "balance_loss_mlp": 1.02130842, "epoch": 0.45355618349065113, "flos": 23440371085920.0, "grad_norm": 1.6965046105289565, "language_loss": 0.79693568, "learning_rate": 2.3954317659158094e-06, "loss": 0.81910408, "num_input_tokens_seen": 81187350, "step": 3772, "time_per_iteration": 2.50622296333313 }, { "auxiliary_loss_clip": 0.01072772, "auxiliary_loss_mlp": 0.01003276, "balance_loss_clip": 1.01109946, "balance_loss_mlp": 1.00197077, "epoch": 0.4536764263812902, "flos": 66903173424000.0, "grad_norm": 0.9072356920035037, "language_loss": 0.56960654, "learning_rate": 2.394668141686667e-06, "loss": 0.59036696, "num_input_tokens_seen": 81249315, "step": 3773, "time_per_iteration": 3.1303138732910156 }, { "auxiliary_loss_clip": 0.01166034, "auxiliary_loss_mlp": 0.01025786, "balance_loss_clip": 1.04990852, "balance_loss_mlp": 1.01820135, "epoch": 0.4537966692719293, "flos": 42739771643520.0, "grad_norm": 1.932636912504011, "language_loss": 0.6967541, "learning_rate": 2.3939044575866813e-06, "loss": 0.71867228, "num_input_tokens_seen": 81272065, "step": 3774, "time_per_iteration": 2.7050347328186035 }, { "auxiliary_loss_clip": 0.01151245, "auxiliary_loss_mlp": 0.00762848, "balance_loss_clip": 1.04939175, "balance_loss_mlp": 1.00054574, "epoch": 0.4539169121625684, "flos": 35549484992640.0, "grad_norm": 2.231037615884809, "language_loss": 0.75423646, "learning_rate": 2.3931407137317024e-06, "loss": 0.77337736, "num_input_tokens_seen": 81292220, "step": 3775, "time_per_iteration": 2.655336380004883 }, { "auxiliary_loss_clip": 0.01135738, "auxiliary_loss_mlp": 0.01026363, "balance_loss_clip": 1.04518938, "balance_loss_mlp": 1.01824486, "epoch": 0.45403715505320746, "flos": 18514719607680.0, "grad_norm": 1.6585262824079308, "language_loss": 0.84939003, "learning_rate": 2.3923769102375907e-06, "loss": 0.87101102, "num_input_tokens_seen": 81311085, "step": 3776, "time_per_iteration": 2.5603578090667725 }, { "auxiliary_loss_clip": 0.01139955, "auxiliary_loss_mlp": 0.01029401, "balance_loss_clip": 1.04621089, "balance_loss_mlp": 1.02105045, "epoch": 0.4541573979438466, "flos": 25045829530560.0, "grad_norm": 2.0505459834214723, "language_loss": 0.7862606, "learning_rate": 2.391613047220213e-06, "loss": 0.80795413, "num_input_tokens_seen": 81330985, "step": 3777, "time_per_iteration": 2.598341464996338 }, { "auxiliary_loss_clip": 0.01136097, "auxiliary_loss_mlp": 0.01026448, "balance_loss_clip": 1.04990721, "balance_loss_mlp": 1.01802564, "epoch": 0.4542776408344857, "flos": 18332401143360.0, "grad_norm": 2.0731420895015376, "language_loss": 0.78902686, "learning_rate": 2.390849124795447e-06, "loss": 0.81065226, "num_input_tokens_seen": 81346985, "step": 3778, "time_per_iteration": 2.5540335178375244 }, { "auxiliary_loss_clip": 0.01187102, "auxiliary_loss_mlp": 0.0103197, "balance_loss_clip": 1.05499196, "balance_loss_mlp": 1.02428114, "epoch": 0.45439788372512474, "flos": 20701176335040.0, "grad_norm": 1.7358200109874709, "language_loss": 0.83908051, "learning_rate": 2.3900851430791804e-06, "loss": 0.86127126, "num_input_tokens_seen": 81365005, "step": 3779, "time_per_iteration": 2.467888116836548 }, { "auxiliary_loss_clip": 0.01188714, "auxiliary_loss_mlp": 0.01034664, "balance_loss_clip": 1.05339992, "balance_loss_mlp": 1.02609277, "epoch": 0.45451812661576385, "flos": 22309436302560.0, "grad_norm": 1.9301518621471807, "language_loss": 0.84388351, "learning_rate": 2.389321102187307e-06, "loss": 0.86611724, "num_input_tokens_seen": 81383785, "step": 3780, "time_per_iteration": 2.467348098754883 }, { "auxiliary_loss_clip": 0.01157872, "auxiliary_loss_mlp": 0.00763488, "balance_loss_clip": 1.05310202, "balance_loss_mlp": 1.0005455, "epoch": 0.4546383695064029, "flos": 21763306999680.0, "grad_norm": 1.9060056022009169, "language_loss": 0.81797743, "learning_rate": 2.3885570022357326e-06, "loss": 0.83719099, "num_input_tokens_seen": 81402915, "step": 3781, "time_per_iteration": 2.557807683944702 }, { "auxiliary_loss_clip": 0.01041939, "auxiliary_loss_mlp": 0.0100167, "balance_loss_clip": 1.00994778, "balance_loss_mlp": 1.00028694, "epoch": 0.454758612397042, "flos": 64242767706720.0, "grad_norm": 0.8041947710462127, "language_loss": 0.60865569, "learning_rate": 2.38779284334037e-06, "loss": 0.62909174, "num_input_tokens_seen": 81467890, "step": 3782, "time_per_iteration": 3.9375107288360596 }, { "auxiliary_loss_clip": 0.01111829, "auxiliary_loss_mlp": 0.0102941, "balance_loss_clip": 1.04383159, "balance_loss_mlp": 1.02107692, "epoch": 0.4548788552876811, "flos": 27304179235200.0, "grad_norm": 1.9123314170178185, "language_loss": 0.78644514, "learning_rate": 2.387028625617141e-06, "loss": 0.80785751, "num_input_tokens_seen": 81487105, "step": 3783, "time_per_iteration": 3.3854610919952393 }, { "auxiliary_loss_clip": 0.01142739, "auxiliary_loss_mlp": 0.01024822, "balance_loss_clip": 1.04650784, "balance_loss_mlp": 1.01691294, "epoch": 0.4549990981783202, "flos": 22857145951680.0, "grad_norm": 1.8589131914551347, "language_loss": 0.84790754, "learning_rate": 2.3862643491819766e-06, "loss": 0.86958319, "num_input_tokens_seen": 81505670, "step": 3784, "time_per_iteration": 2.537557363510132 }, { "auxiliary_loss_clip": 0.01168014, "auxiliary_loss_mlp": 0.01025695, "balance_loss_clip": 1.04925334, "balance_loss_mlp": 1.01776731, "epoch": 0.4551193410689593, "flos": 23258591376000.0, "grad_norm": 2.19369732957101, "language_loss": 0.84709191, "learning_rate": 2.3855000141508186e-06, "loss": 0.86902899, "num_input_tokens_seen": 81525825, "step": 3785, "time_per_iteration": 3.2896032333374023 }, { "auxiliary_loss_clip": 0.01160342, "auxiliary_loss_mlp": 0.01031641, "balance_loss_clip": 1.0524478, "balance_loss_mlp": 1.02338576, "epoch": 0.4552395839595984, "flos": 20777523015360.0, "grad_norm": 5.0845269496113765, "language_loss": 0.83821976, "learning_rate": 2.3847356206396143e-06, "loss": 0.86013961, "num_input_tokens_seen": 81543135, "step": 3786, "time_per_iteration": 2.530865430831909 }, { "auxiliary_loss_clip": 0.01185065, "auxiliary_loss_mlp": 0.01029612, "balance_loss_clip": 1.05378878, "balance_loss_mlp": 1.02187586, "epoch": 0.45535982685023746, "flos": 23257513867200.0, "grad_norm": 1.4179961695864345, "language_loss": 0.78372461, "learning_rate": 2.3839711687643227e-06, "loss": 0.80587143, "num_input_tokens_seen": 81564360, "step": 3787, "time_per_iteration": 2.5233547687530518 }, { "auxiliary_loss_clip": 0.01175146, "auxiliary_loss_mlp": 0.01029661, "balance_loss_clip": 1.05607772, "balance_loss_mlp": 1.02025545, "epoch": 0.45548006974087657, "flos": 19646121311520.0, "grad_norm": 2.035555919333769, "language_loss": 0.74055767, "learning_rate": 2.38320665864091e-06, "loss": 0.76260579, "num_input_tokens_seen": 81583710, "step": 3788, "time_per_iteration": 2.4904184341430664 }, { "auxiliary_loss_clip": 0.01116281, "auxiliary_loss_mlp": 0.01031756, "balance_loss_clip": 1.04381156, "balance_loss_mlp": 1.02317286, "epoch": 0.4556003126315157, "flos": 20047782237600.0, "grad_norm": 1.6967939764400968, "language_loss": 0.82045037, "learning_rate": 2.3824420903853516e-06, "loss": 0.84193075, "num_input_tokens_seen": 81602175, "step": 3789, "time_per_iteration": 3.413271427154541 }, { "auxiliary_loss_clip": 0.01172117, "auxiliary_loss_mlp": 0.01033893, "balance_loss_clip": 1.05386114, "balance_loss_mlp": 1.0258168, "epoch": 0.45572055552215474, "flos": 22959741541440.0, "grad_norm": 2.087869047512641, "language_loss": 0.81600988, "learning_rate": 2.3816774641136324e-06, "loss": 0.83807003, "num_input_tokens_seen": 81619430, "step": 3790, "time_per_iteration": 2.495236396789551 }, { "auxiliary_loss_clip": 0.01169586, "auxiliary_loss_mlp": 0.00762729, "balance_loss_clip": 1.05364895, "balance_loss_mlp": 1.00063562, "epoch": 0.45584079841279385, "flos": 33109930249440.0, "grad_norm": 2.6361425286004767, "language_loss": 0.71220118, "learning_rate": 2.380912779941745e-06, "loss": 0.73152429, "num_input_tokens_seen": 81642550, "step": 3791, "time_per_iteration": 2.640665292739868 }, { "auxiliary_loss_clip": 0.01172653, "auxiliary_loss_mlp": 0.01028887, "balance_loss_clip": 1.04935789, "balance_loss_mlp": 1.01961279, "epoch": 0.45596104130343296, "flos": 27272183612160.0, "grad_norm": 2.0799707915626042, "language_loss": 0.82925284, "learning_rate": 2.3801480379856918e-06, "loss": 0.85126829, "num_input_tokens_seen": 81664260, "step": 3792, "time_per_iteration": 2.5525054931640625 }, { "auxiliary_loss_clip": 0.01159306, "auxiliary_loss_mlp": 0.01030281, "balance_loss_clip": 1.05337524, "balance_loss_mlp": 1.02203727, "epoch": 0.456081284194072, "flos": 21579803275680.0, "grad_norm": 1.6500163228287839, "language_loss": 0.83289444, "learning_rate": 2.379383238361484e-06, "loss": 0.85479021, "num_input_tokens_seen": 81683620, "step": 3793, "time_per_iteration": 2.554624080657959 }, { "auxiliary_loss_clip": 0.01168876, "auxiliary_loss_mlp": 0.01029438, "balance_loss_clip": 1.05312681, "balance_loss_mlp": 1.02117634, "epoch": 0.4562015270847111, "flos": 35918827043040.0, "grad_norm": 1.7961512230634817, "language_loss": 0.79349548, "learning_rate": 2.3786183811851407e-06, "loss": 0.81547856, "num_input_tokens_seen": 81704325, "step": 3794, "time_per_iteration": 2.620952844619751 }, { "auxiliary_loss_clip": 0.01188788, "auxiliary_loss_mlp": 0.0102968, "balance_loss_clip": 1.05581403, "balance_loss_mlp": 1.02135336, "epoch": 0.45632176997535023, "flos": 13589786468640.0, "grad_norm": 1.9177006033343171, "language_loss": 0.79979086, "learning_rate": 2.3778534665726892e-06, "loss": 0.82197559, "num_input_tokens_seen": 81721155, "step": 3795, "time_per_iteration": 2.4648468494415283 }, { "auxiliary_loss_clip": 0.01158993, "auxiliary_loss_mlp": 0.01035022, "balance_loss_clip": 1.0492034, "balance_loss_mlp": 1.02722561, "epoch": 0.4564420128659893, "flos": 32635406588160.0, "grad_norm": 1.6783198230560294, "language_loss": 0.72579467, "learning_rate": 2.377088494640168e-06, "loss": 0.7477349, "num_input_tokens_seen": 81742905, "step": 3796, "time_per_iteration": 2.5815773010253906 }, { "auxiliary_loss_clip": 0.01171513, "auxiliary_loss_mlp": 0.01028905, "balance_loss_clip": 1.0567832, "balance_loss_mlp": 1.02103138, "epoch": 0.4565622557566284, "flos": 20377693854240.0, "grad_norm": 1.759062804739426, "language_loss": 0.78606313, "learning_rate": 2.3763234655036216e-06, "loss": 0.80806732, "num_input_tokens_seen": 81762105, "step": 3797, "time_per_iteration": 2.5221080780029297 }, { "auxiliary_loss_clip": 0.01139012, "auxiliary_loss_mlp": 0.01032498, "balance_loss_clip": 1.04400778, "balance_loss_mlp": 1.02452946, "epoch": 0.45668249864726745, "flos": 25374376302720.0, "grad_norm": 2.4216249047016833, "language_loss": 0.87081951, "learning_rate": 2.3755583792791046e-06, "loss": 0.89253461, "num_input_tokens_seen": 81781975, "step": 3798, "time_per_iteration": 2.5850179195404053 }, { "auxiliary_loss_clip": 0.01174297, "auxiliary_loss_mlp": 0.01033475, "balance_loss_clip": 1.05376303, "balance_loss_mlp": 1.02532732, "epoch": 0.45680274153790656, "flos": 15559810008000.0, "grad_norm": 1.8987657206905664, "language_loss": 0.74533325, "learning_rate": 2.3747932360826803e-06, "loss": 0.76741099, "num_input_tokens_seen": 81798905, "step": 3799, "time_per_iteration": 2.4870400428771973 }, { "auxiliary_loss_clip": 0.01170519, "auxiliary_loss_mlp": 0.01033788, "balance_loss_clip": 1.05250537, "balance_loss_mlp": 1.02566719, "epoch": 0.4569229844285457, "flos": 19792888373760.0, "grad_norm": 1.766750274343394, "language_loss": 0.82172793, "learning_rate": 2.3740280360304205e-06, "loss": 0.84377104, "num_input_tokens_seen": 81816630, "step": 3800, "time_per_iteration": 2.4711594581604004 }, { "auxiliary_loss_clip": 0.01144909, "auxiliary_loss_mlp": 0.0102491, "balance_loss_clip": 1.05326295, "balance_loss_mlp": 1.01698554, "epoch": 0.45704322731918473, "flos": 24093944768160.0, "grad_norm": 1.712887655094274, "language_loss": 0.68014985, "learning_rate": 2.3732627792384038e-06, "loss": 0.70184803, "num_input_tokens_seen": 81837700, "step": 3801, "time_per_iteration": 2.588070869445801 }, { "auxiliary_loss_clip": 0.01186775, "auxiliary_loss_mlp": 0.01024066, "balance_loss_clip": 1.05495024, "balance_loss_mlp": 1.01607883, "epoch": 0.45716347020982384, "flos": 31317807388320.0, "grad_norm": 2.1324076576276907, "language_loss": 0.75294507, "learning_rate": 2.3724974658227207e-06, "loss": 0.7750535, "num_input_tokens_seen": 81858490, "step": 3802, "time_per_iteration": 2.547968864440918 }, { "auxiliary_loss_clip": 0.01155226, "auxiliary_loss_mlp": 0.00763394, "balance_loss_clip": 1.05162191, "balance_loss_mlp": 1.00048137, "epoch": 0.45728371310046295, "flos": 26501396137440.0, "grad_norm": 1.9493334980089003, "language_loss": 0.71055603, "learning_rate": 2.3717320958994687e-06, "loss": 0.72974223, "num_input_tokens_seen": 81876050, "step": 3803, "time_per_iteration": 2.579871892929077 }, { "auxiliary_loss_clip": 0.0114033, "auxiliary_loss_mlp": 0.01028096, "balance_loss_clip": 1.043751, "balance_loss_mlp": 1.02033544, "epoch": 0.457403955991102, "flos": 17929411289760.0, "grad_norm": 1.9448354340492873, "language_loss": 0.70119017, "learning_rate": 2.3709666695847534e-06, "loss": 0.7228744, "num_input_tokens_seen": 81894230, "step": 3804, "time_per_iteration": 2.5495784282684326 }, { "auxiliary_loss_clip": 0.01119504, "auxiliary_loss_mlp": 0.0102257, "balance_loss_clip": 1.0445559, "balance_loss_mlp": 1.015131, "epoch": 0.4575241988817411, "flos": 42230666338080.0, "grad_norm": 1.816100353072236, "language_loss": 0.70213401, "learning_rate": 2.370201186994689e-06, "loss": 0.72355479, "num_input_tokens_seen": 81917915, "step": 3805, "time_per_iteration": 2.795596122741699 }, { "auxiliary_loss_clip": 0.01147758, "auxiliary_loss_mlp": 0.01036293, "balance_loss_clip": 1.05011714, "balance_loss_mlp": 1.0278585, "epoch": 0.45764444177238023, "flos": 30117314230080.0, "grad_norm": 2.318146638511257, "language_loss": 0.69755858, "learning_rate": 2.369435648245399e-06, "loss": 0.71939909, "num_input_tokens_seen": 81938130, "step": 3806, "time_per_iteration": 2.626537799835205 }, { "auxiliary_loss_clip": 0.01155404, "auxiliary_loss_mlp": 0.01031367, "balance_loss_clip": 1.05012894, "balance_loss_mlp": 1.02316594, "epoch": 0.4577646846630193, "flos": 24060296964960.0, "grad_norm": 1.6679303448795657, "language_loss": 0.85195702, "learning_rate": 2.368670053453015e-06, "loss": 0.87382478, "num_input_tokens_seen": 81959820, "step": 3807, "time_per_iteration": 2.5682871341705322 }, { "auxiliary_loss_clip": 0.01179031, "auxiliary_loss_mlp": 0.0103412, "balance_loss_clip": 1.05509698, "balance_loss_mlp": 1.0259124, "epoch": 0.4578849275536584, "flos": 17418581970240.0, "grad_norm": 2.0364245660534688, "language_loss": 0.74148029, "learning_rate": 2.3679044027336757e-06, "loss": 0.76361179, "num_input_tokens_seen": 81975710, "step": 3808, "time_per_iteration": 3.240359306335449 }, { "auxiliary_loss_clip": 0.0118717, "auxiliary_loss_mlp": 0.01028364, "balance_loss_clip": 1.05449784, "balance_loss_mlp": 1.02010858, "epoch": 0.4580051704442975, "flos": 13510171344960.0, "grad_norm": 5.332270872059965, "language_loss": 0.68781149, "learning_rate": 2.3671386962035326e-06, "loss": 0.70996684, "num_input_tokens_seen": 81993180, "step": 3809, "time_per_iteration": 3.198880672454834 }, { "auxiliary_loss_clip": 0.01172611, "auxiliary_loss_mlp": 0.0102972, "balance_loss_clip": 1.05256832, "balance_loss_mlp": 1.02102387, "epoch": 0.45812541333493656, "flos": 18037609925280.0, "grad_norm": 2.0931516267445347, "language_loss": 0.68929803, "learning_rate": 2.3663729339787405e-06, "loss": 0.71132129, "num_input_tokens_seen": 82010115, "step": 3810, "time_per_iteration": 2.4951894283294678 }, { "auxiliary_loss_clip": 0.0118766, "auxiliary_loss_mlp": 0.01037015, "balance_loss_clip": 1.05461073, "balance_loss_mlp": 1.02824759, "epoch": 0.45824565622557567, "flos": 20222198970720.0, "grad_norm": 3.3858287188457323, "language_loss": 0.73507321, "learning_rate": 2.365607116175466e-06, "loss": 0.75731993, "num_input_tokens_seen": 82025540, "step": 3811, "time_per_iteration": 3.2310118675231934 }, { "auxiliary_loss_clip": 0.01183145, "auxiliary_loss_mlp": 0.01029465, "balance_loss_clip": 1.05256462, "balance_loss_mlp": 1.02119195, "epoch": 0.4583658991162148, "flos": 19864889101920.0, "grad_norm": 2.1591229919853956, "language_loss": 0.66656816, "learning_rate": 2.3648412429098825e-06, "loss": 0.68869424, "num_input_tokens_seen": 82043890, "step": 3812, "time_per_iteration": 2.473491668701172 }, { "auxiliary_loss_clip": 0.01139269, "auxiliary_loss_mlp": 0.01033979, "balance_loss_clip": 1.05001485, "balance_loss_mlp": 1.02566457, "epoch": 0.45848614200685384, "flos": 21029938608960.0, "grad_norm": 2.889827767697468, "language_loss": 0.81975484, "learning_rate": 2.364075314298172e-06, "loss": 0.84148735, "num_input_tokens_seen": 82061345, "step": 3813, "time_per_iteration": 2.559264659881592 }, { "auxiliary_loss_clip": 0.01173622, "auxiliary_loss_mlp": 0.00763599, "balance_loss_clip": 1.05233681, "balance_loss_mlp": 1.00044489, "epoch": 0.45860638489749295, "flos": 21069297208800.0, "grad_norm": 1.8470672082624953, "language_loss": 0.69867581, "learning_rate": 2.3633093304565267e-06, "loss": 0.71804798, "num_input_tokens_seen": 82080400, "step": 3814, "time_per_iteration": 2.5200679302215576 }, { "auxiliary_loss_clip": 0.01191737, "auxiliary_loss_mlp": 0.01036655, "balance_loss_clip": 1.05738163, "balance_loss_mlp": 1.02863789, "epoch": 0.458726627788132, "flos": 26833893775200.0, "grad_norm": 1.890092080002743, "language_loss": 0.62561893, "learning_rate": 2.3625432915011443e-06, "loss": 0.64790291, "num_input_tokens_seen": 82102310, "step": 3815, "time_per_iteration": 3.255408525466919 }, { "auxiliary_loss_clip": 0.01148067, "auxiliary_loss_mlp": 0.01031856, "balance_loss_clip": 1.0486896, "balance_loss_mlp": 1.0233202, "epoch": 0.4588468706787711, "flos": 24097931550720.0, "grad_norm": 1.56145920313865, "language_loss": 0.64977115, "learning_rate": 2.3617771975482334e-06, "loss": 0.67157042, "num_input_tokens_seen": 82121140, "step": 3816, "time_per_iteration": 2.5407934188842773 }, { "auxiliary_loss_clip": 0.01124543, "auxiliary_loss_mlp": 0.01030802, "balance_loss_clip": 1.04816055, "balance_loss_mlp": 1.02313113, "epoch": 0.4589671135694102, "flos": 17889334350720.0, "grad_norm": 1.5197698479868476, "language_loss": 0.74600911, "learning_rate": 2.3610110487140083e-06, "loss": 0.76756263, "num_input_tokens_seen": 82139575, "step": 3817, "time_per_iteration": 2.587376594543457 }, { "auxiliary_loss_clip": 0.01154445, "auxiliary_loss_mlp": 0.01030293, "balance_loss_clip": 1.05036163, "balance_loss_mlp": 1.02235925, "epoch": 0.4590873564600493, "flos": 25626971481120.0, "grad_norm": 1.6421176960218162, "language_loss": 0.80477947, "learning_rate": 2.360244845114695e-06, "loss": 0.82662684, "num_input_tokens_seen": 82159195, "step": 3818, "time_per_iteration": 2.5663719177246094 }, { "auxiliary_loss_clip": 0.01150759, "auxiliary_loss_mlp": 0.0103094, "balance_loss_clip": 1.05143452, "balance_loss_mlp": 1.02280974, "epoch": 0.4592075993506884, "flos": 18514791441600.0, "grad_norm": 2.1010407547783903, "language_loss": 0.68616611, "learning_rate": 2.3594785868665245e-06, "loss": 0.70798308, "num_input_tokens_seen": 82175500, "step": 3819, "time_per_iteration": 2.526407480239868 }, { "auxiliary_loss_clip": 0.01141123, "auxiliary_loss_mlp": 0.00763189, "balance_loss_clip": 1.04795969, "balance_loss_mlp": 1.00043583, "epoch": 0.4593278422413275, "flos": 20631115122720.0, "grad_norm": 2.5457937424517594, "language_loss": 0.80543387, "learning_rate": 2.3587122740857386e-06, "loss": 0.82447696, "num_input_tokens_seen": 82192600, "step": 3820, "time_per_iteration": 2.560260057449341 }, { "auxiliary_loss_clip": 0.01172295, "auxiliary_loss_mlp": 0.01022521, "balance_loss_clip": 1.05387878, "balance_loss_mlp": 1.01503432, "epoch": 0.45944808513196655, "flos": 21358521298080.0, "grad_norm": 1.5016260477651875, "language_loss": 0.77977437, "learning_rate": 2.357945906888586e-06, "loss": 0.80172253, "num_input_tokens_seen": 82212040, "step": 3821, "time_per_iteration": 2.5158703327178955 }, { "auxiliary_loss_clip": 0.01171728, "auxiliary_loss_mlp": 0.01036098, "balance_loss_clip": 1.05353999, "balance_loss_mlp": 1.027753, "epoch": 0.45956832802260567, "flos": 21427792337280.0, "grad_norm": 2.375610351956015, "language_loss": 0.80060184, "learning_rate": 2.357179485391324e-06, "loss": 0.82268006, "num_input_tokens_seen": 82229895, "step": 3822, "time_per_iteration": 2.497495651245117 }, { "auxiliary_loss_clip": 0.01183006, "auxiliary_loss_mlp": 0.01033691, "balance_loss_clip": 1.05396867, "balance_loss_mlp": 1.02612734, "epoch": 0.4596885709132448, "flos": 22382658207360.0, "grad_norm": 1.9342845413765266, "language_loss": 0.86064208, "learning_rate": 2.3564130097102173e-06, "loss": 0.8828091, "num_input_tokens_seen": 82249550, "step": 3823, "time_per_iteration": 2.4882020950317383 }, { "auxiliary_loss_clip": 0.01146118, "auxiliary_loss_mlp": 0.01035484, "balance_loss_clip": 1.05039823, "balance_loss_mlp": 1.0272944, "epoch": 0.45980881380388383, "flos": 28981961660640.0, "grad_norm": 1.7990626878981932, "language_loss": 0.75282323, "learning_rate": 2.355646479961541e-06, "loss": 0.77463931, "num_input_tokens_seen": 82268860, "step": 3824, "time_per_iteration": 2.577047348022461 }, { "auxiliary_loss_clip": 0.01184922, "auxiliary_loss_mlp": 0.01031308, "balance_loss_clip": 1.05485737, "balance_loss_mlp": 1.02312982, "epoch": 0.45992905669452294, "flos": 33396604234560.0, "grad_norm": 1.7234419358534472, "language_loss": 0.71861839, "learning_rate": 2.354879896261576e-06, "loss": 0.74078071, "num_input_tokens_seen": 82289070, "step": 3825, "time_per_iteration": 2.5861666202545166 }, { "auxiliary_loss_clip": 0.01139149, "auxiliary_loss_mlp": 0.01034276, "balance_loss_clip": 1.05183339, "balance_loss_mlp": 1.02647376, "epoch": 0.46004929958516205, "flos": 36318189283680.0, "grad_norm": 2.36355304677327, "language_loss": 0.56550086, "learning_rate": 2.3541132587266133e-06, "loss": 0.58723509, "num_input_tokens_seen": 82311790, "step": 3826, "time_per_iteration": 2.688615322113037 }, { "auxiliary_loss_clip": 0.01146088, "auxiliary_loss_mlp": 0.0103066, "balance_loss_clip": 1.04950798, "balance_loss_mlp": 1.02267885, "epoch": 0.4601695424758011, "flos": 17238454440480.0, "grad_norm": 1.9548455286572344, "language_loss": 0.69175899, "learning_rate": 2.3533465674729515e-06, "loss": 0.71352649, "num_input_tokens_seen": 82329020, "step": 3827, "time_per_iteration": 2.5908892154693604 }, { "auxiliary_loss_clip": 0.0118767, "auxiliary_loss_mlp": 0.01039639, "balance_loss_clip": 1.0563159, "balance_loss_mlp": 1.03075814, "epoch": 0.4602897853664402, "flos": 15888428614080.0, "grad_norm": 1.9877664147661631, "language_loss": 0.7328369, "learning_rate": 2.352579822616895e-06, "loss": 0.75510997, "num_input_tokens_seen": 82346455, "step": 3828, "time_per_iteration": 2.4538941383361816 }, { "auxiliary_loss_clip": 0.01154166, "auxiliary_loss_mlp": 0.01024006, "balance_loss_clip": 1.05057669, "balance_loss_mlp": 1.01551199, "epoch": 0.4604100282570793, "flos": 25412621476800.0, "grad_norm": 1.6804961068965043, "language_loss": 0.77897227, "learning_rate": 2.351813024274761e-06, "loss": 0.80075395, "num_input_tokens_seen": 82367810, "step": 3829, "time_per_iteration": 2.595526933670044 }, { "auxiliary_loss_clip": 0.01145042, "auxiliary_loss_mlp": 0.01031195, "balance_loss_clip": 1.04997849, "balance_loss_mlp": 1.02353573, "epoch": 0.4605302711477184, "flos": 27630714657600.0, "grad_norm": 1.7908321546559274, "language_loss": 0.73449475, "learning_rate": 2.3510461725628693e-06, "loss": 0.75625706, "num_input_tokens_seen": 82388275, "step": 3830, "time_per_iteration": 2.666142702102661 }, { "auxiliary_loss_clip": 0.01145274, "auxiliary_loss_mlp": 0.01025403, "balance_loss_clip": 1.05025136, "balance_loss_mlp": 1.01772571, "epoch": 0.4606505140383575, "flos": 23839661492640.0, "grad_norm": 1.9717894861475653, "language_loss": 0.71161449, "learning_rate": 2.350279267597554e-06, "loss": 0.73332125, "num_input_tokens_seen": 82408915, "step": 3831, "time_per_iteration": 2.6131768226623535 }, { "auxiliary_loss_clip": 0.01170189, "auxiliary_loss_mlp": 0.01031955, "balance_loss_clip": 1.05290413, "balance_loss_mlp": 1.02344966, "epoch": 0.46077075692899655, "flos": 16107016819680.0, "grad_norm": 2.3728361419794624, "language_loss": 0.82553893, "learning_rate": 2.3495123094951515e-06, "loss": 0.84756041, "num_input_tokens_seen": 82427260, "step": 3832, "time_per_iteration": 2.498640775680542 }, { "auxiliary_loss_clip": 0.01146529, "auxiliary_loss_mlp": 0.01033016, "balance_loss_clip": 1.04885304, "balance_loss_mlp": 1.02493954, "epoch": 0.46089099981963566, "flos": 48798153753120.0, "grad_norm": 2.4182110646611834, "language_loss": 0.75972819, "learning_rate": 2.34874529837201e-06, "loss": 0.78152364, "num_input_tokens_seen": 82450805, "step": 3833, "time_per_iteration": 2.767179012298584 }, { "auxiliary_loss_clip": 0.01103671, "auxiliary_loss_mlp": 0.01033338, "balance_loss_clip": 1.04219353, "balance_loss_mlp": 1.0257771, "epoch": 0.46101124271027477, "flos": 19099237752480.0, "grad_norm": 1.7886296343254597, "language_loss": 0.7873376, "learning_rate": 2.347978234344483e-06, "loss": 0.80870771, "num_input_tokens_seen": 82467010, "step": 3834, "time_per_iteration": 3.3803176879882812 }, { "auxiliary_loss_clip": 0.01175124, "auxiliary_loss_mlp": 0.01030106, "balance_loss_clip": 1.05338597, "balance_loss_mlp": 1.02199936, "epoch": 0.4611314856009138, "flos": 39347937051360.0, "grad_norm": 2.643775787037039, "language_loss": 0.68955177, "learning_rate": 2.347211117528935e-06, "loss": 0.71160406, "num_input_tokens_seen": 82489310, "step": 3835, "time_per_iteration": 3.3412294387817383 }, { "auxiliary_loss_clip": 0.01153748, "auxiliary_loss_mlp": 0.01031565, "balance_loss_clip": 1.05417347, "balance_loss_mlp": 1.02412081, "epoch": 0.46125172849155294, "flos": 20810775732000.0, "grad_norm": 1.5509360717578744, "language_loss": 0.71717954, "learning_rate": 2.3464439480417374e-06, "loss": 0.73903275, "num_input_tokens_seen": 82508830, "step": 3836, "time_per_iteration": 2.5776655673980713 }, { "auxiliary_loss_clip": 0.01173485, "auxiliary_loss_mlp": 0.01033883, "balance_loss_clip": 1.05288482, "balance_loss_mlp": 1.02516246, "epoch": 0.46137197138219205, "flos": 17930812051200.0, "grad_norm": 2.4252024671420096, "language_loss": 0.77287006, "learning_rate": 2.3456767259992676e-06, "loss": 0.79494375, "num_input_tokens_seen": 82526475, "step": 3837, "time_per_iteration": 3.2008302211761475 }, { "auxiliary_loss_clip": 0.01184599, "auxiliary_loss_mlp": 0.00762819, "balance_loss_clip": 1.05156469, "balance_loss_mlp": 1.00038242, "epoch": 0.4614922142728311, "flos": 16836613929600.0, "grad_norm": 1.9596594575867057, "language_loss": 0.88457346, "learning_rate": 2.3449094515179135e-06, "loss": 0.90404773, "num_input_tokens_seen": 82543935, "step": 3838, "time_per_iteration": 2.4701993465423584 }, { "auxiliary_loss_clip": 0.0115692, "auxiliary_loss_mlp": 0.01029192, "balance_loss_clip": 1.04700148, "balance_loss_mlp": 1.02153254, "epoch": 0.4616124571634702, "flos": 26614910483040.0, "grad_norm": 1.5597928242158532, "language_loss": 0.81770962, "learning_rate": 2.34414212471407e-06, "loss": 0.83957076, "num_input_tokens_seen": 82563730, "step": 3839, "time_per_iteration": 2.5695223808288574 }, { "auxiliary_loss_clip": 0.01175944, "auxiliary_loss_mlp": 0.01030914, "balance_loss_clip": 1.05159974, "balance_loss_mlp": 1.02256942, "epoch": 0.4617327000541093, "flos": 20340131102400.0, "grad_norm": 1.868147240193238, "language_loss": 0.72935152, "learning_rate": 2.3433747457041394e-06, "loss": 0.75142008, "num_input_tokens_seen": 82582435, "step": 3840, "time_per_iteration": 2.480217456817627 }, { "auxiliary_loss_clip": 0.01141445, "auxiliary_loss_mlp": 0.01028669, "balance_loss_clip": 1.04902434, "balance_loss_mlp": 1.02021658, "epoch": 0.4618529429447484, "flos": 29570753923680.0, "grad_norm": 1.657571476614973, "language_loss": 0.84444249, "learning_rate": 2.342607314604533e-06, "loss": 0.8661437, "num_input_tokens_seen": 82602185, "step": 3841, "time_per_iteration": 3.3491525650024414 }, { "auxiliary_loss_clip": 0.01167798, "auxiliary_loss_mlp": 0.01027645, "balance_loss_clip": 1.05218112, "balance_loss_mlp": 1.01935959, "epoch": 0.4619731858353875, "flos": 19787033909280.0, "grad_norm": 1.758324521387079, "language_loss": 0.84172344, "learning_rate": 2.3418398315316694e-06, "loss": 0.86367786, "num_input_tokens_seen": 82620005, "step": 3842, "time_per_iteration": 2.4798614978790283 }, { "auxiliary_loss_clip": 0.01183118, "auxiliary_loss_mlp": 0.0103423, "balance_loss_clip": 1.0537976, "balance_loss_mlp": 1.02648735, "epoch": 0.4620934287260266, "flos": 18951141762720.0, "grad_norm": 2.3671115491897523, "language_loss": 0.78097028, "learning_rate": 2.3410722966019755e-06, "loss": 0.80314374, "num_input_tokens_seen": 82635120, "step": 3843, "time_per_iteration": 2.4646716117858887 }, { "auxiliary_loss_clip": 0.01168522, "auxiliary_loss_mlp": 0.01028589, "balance_loss_clip": 1.05251586, "balance_loss_mlp": 1.02041698, "epoch": 0.46221367161666566, "flos": 37341679687680.0, "grad_norm": 2.0816596441756783, "language_loss": 0.65794885, "learning_rate": 2.3403047099318848e-06, "loss": 0.67991996, "num_input_tokens_seen": 82659190, "step": 3844, "time_per_iteration": 2.6237728595733643 }, { "auxiliary_loss_clip": 0.01119197, "auxiliary_loss_mlp": 0.01026149, "balance_loss_clip": 1.04446888, "balance_loss_mlp": 1.0187403, "epoch": 0.46233391450730477, "flos": 14428551972000.0, "grad_norm": 2.1886758010923226, "language_loss": 0.74562198, "learning_rate": 2.3395370716378405e-06, "loss": 0.76707542, "num_input_tokens_seen": 82676635, "step": 3845, "time_per_iteration": 2.5757336616516113 }, { "auxiliary_loss_clip": 0.01170358, "auxiliary_loss_mlp": 0.0102404, "balance_loss_clip": 1.0501709, "balance_loss_mlp": 1.01641679, "epoch": 0.4624541573979438, "flos": 22493047777440.0, "grad_norm": 2.3627735744429375, "language_loss": 0.72378218, "learning_rate": 2.338769381836292e-06, "loss": 0.74572617, "num_input_tokens_seen": 82696245, "step": 3846, "time_per_iteration": 2.4877402782440186 }, { "auxiliary_loss_clip": 0.01138809, "auxiliary_loss_mlp": 0.01028793, "balance_loss_clip": 1.04938579, "balance_loss_mlp": 1.0213964, "epoch": 0.46257440028858293, "flos": 14465073132000.0, "grad_norm": 2.0814821965580124, "language_loss": 0.73404068, "learning_rate": 2.3380016406436984e-06, "loss": 0.75571668, "num_input_tokens_seen": 82713725, "step": 3847, "time_per_iteration": 2.543248176574707 }, { "auxiliary_loss_clip": 0.01122527, "auxiliary_loss_mlp": 0.01029856, "balance_loss_clip": 1.04625964, "balance_loss_mlp": 1.02130294, "epoch": 0.46269464317922204, "flos": 23332208367360.0, "grad_norm": 1.9449552757530861, "language_loss": 0.81685674, "learning_rate": 2.337233848176524e-06, "loss": 0.83838058, "num_input_tokens_seen": 82731495, "step": 3848, "time_per_iteration": 2.5917885303497314 }, { "auxiliary_loss_clip": 0.01118844, "auxiliary_loss_mlp": 0.01027292, "balance_loss_clip": 1.04540324, "balance_loss_mlp": 1.01862526, "epoch": 0.4628148860698611, "flos": 18552030940800.0, "grad_norm": 1.8683979824040635, "language_loss": 0.83533877, "learning_rate": 2.3364660045512435e-06, "loss": 0.85680014, "num_input_tokens_seen": 82750255, "step": 3849, "time_per_iteration": 2.6054112911224365 }, { "auxiliary_loss_clip": 0.01050562, "auxiliary_loss_mlp": 0.01004875, "balance_loss_clip": 1.0130682, "balance_loss_mlp": 1.00361753, "epoch": 0.4629351289605002, "flos": 70667582592960.0, "grad_norm": 0.7457329378532498, "language_loss": 0.58266509, "learning_rate": 2.335698109884337e-06, "loss": 0.60321939, "num_input_tokens_seen": 82815460, "step": 3850, "time_per_iteration": 3.267056465148926 }, { "auxiliary_loss_clip": 0.01033546, "auxiliary_loss_mlp": 0.00999796, "balance_loss_clip": 1.01729691, "balance_loss_mlp": 0.99853194, "epoch": 0.4630553718511393, "flos": 59687212535040.0, "grad_norm": 0.7921983941492202, "language_loss": 0.59903556, "learning_rate": 2.334930164292294e-06, "loss": 0.61936897, "num_input_tokens_seen": 82878010, "step": 3851, "time_per_iteration": 3.317397117614746 }, { "auxiliary_loss_clip": 0.01117778, "auxiliary_loss_mlp": 0.01033828, "balance_loss_clip": 1.04297805, "balance_loss_mlp": 1.02596068, "epoch": 0.4631756147417784, "flos": 15960608927040.0, "grad_norm": 2.1538282125580097, "language_loss": 0.80151522, "learning_rate": 2.334162167891612e-06, "loss": 0.82303131, "num_input_tokens_seen": 82895275, "step": 3852, "time_per_iteration": 2.60632061958313 }, { "auxiliary_loss_clip": 0.01157335, "auxiliary_loss_mlp": 0.01028295, "balance_loss_clip": 1.04871464, "balance_loss_mlp": 1.02004552, "epoch": 0.4632958576324175, "flos": 16472910841920.0, "grad_norm": 1.8966386153186017, "language_loss": 0.75057554, "learning_rate": 2.333394120798795e-06, "loss": 0.77243179, "num_input_tokens_seen": 82914010, "step": 3853, "time_per_iteration": 2.536588430404663 }, { "auxiliary_loss_clip": 0.01156467, "auxiliary_loss_mlp": 0.01031334, "balance_loss_clip": 1.04924846, "balance_loss_mlp": 1.02375865, "epoch": 0.4634161005230566, "flos": 22346496216960.0, "grad_norm": 2.708596121859616, "language_loss": 0.71742058, "learning_rate": 2.3326260231303545e-06, "loss": 0.73929858, "num_input_tokens_seen": 82932610, "step": 3854, "time_per_iteration": 2.536740779876709 }, { "auxiliary_loss_clip": 0.01183348, "auxiliary_loss_mlp": 0.01023448, "balance_loss_clip": 1.05420041, "balance_loss_mlp": 1.01559806, "epoch": 0.46353634341369565, "flos": 15742236223200.0, "grad_norm": 1.6260254115720947, "language_loss": 0.86787188, "learning_rate": 2.331857875002811e-06, "loss": 0.88993979, "num_input_tokens_seen": 82951210, "step": 3855, "time_per_iteration": 2.467982530593872 }, { "auxiliary_loss_clip": 0.01157262, "auxiliary_loss_mlp": 0.01032906, "balance_loss_clip": 1.05291581, "balance_loss_mlp": 1.02534223, "epoch": 0.46365658630433476, "flos": 28329824656800.0, "grad_norm": 2.0182557995279087, "language_loss": 0.76182389, "learning_rate": 2.3310896765326916e-06, "loss": 0.78372562, "num_input_tokens_seen": 82972210, "step": 3856, "time_per_iteration": 2.617849111557007 }, { "auxiliary_loss_clip": 0.01135538, "auxiliary_loss_mlp": 0.0103433, "balance_loss_clip": 1.04715943, "balance_loss_mlp": 1.02628326, "epoch": 0.46377682919497387, "flos": 24608078448000.0, "grad_norm": 1.4979336628734379, "language_loss": 0.84149837, "learning_rate": 2.330321427836531e-06, "loss": 0.86319703, "num_input_tokens_seen": 82994080, "step": 3857, "time_per_iteration": 2.600079298019409 }, { "auxiliary_loss_clip": 0.01165661, "auxiliary_loss_mlp": 0.01031148, "balance_loss_clip": 1.05036628, "balance_loss_mlp": 1.02339983, "epoch": 0.4638970720856129, "flos": 19060956661440.0, "grad_norm": 1.5849195321286789, "language_loss": 0.82829571, "learning_rate": 2.3295531290308733e-06, "loss": 0.85026383, "num_input_tokens_seen": 83012230, "step": 3858, "time_per_iteration": 2.5179290771484375 }, { "auxiliary_loss_clip": 0.01187091, "auxiliary_loss_mlp": 0.00763208, "balance_loss_clip": 1.05551171, "balance_loss_mlp": 1.00046921, "epoch": 0.46401731497625204, "flos": 18471015055680.0, "grad_norm": 2.5224464852166806, "language_loss": 0.75506341, "learning_rate": 2.3287847802322678e-06, "loss": 0.77456641, "num_input_tokens_seen": 83027800, "step": 3859, "time_per_iteration": 2.433326482772827 }, { "auxiliary_loss_clip": 0.01163059, "auxiliary_loss_mlp": 0.01034948, "balance_loss_clip": 1.05387807, "balance_loss_mlp": 1.02618027, "epoch": 0.4641375578668911, "flos": 26067057166080.0, "grad_norm": 2.215521059776194, "language_loss": 0.84042263, "learning_rate": 2.3280163815572723e-06, "loss": 0.86240268, "num_input_tokens_seen": 83048395, "step": 3860, "time_per_iteration": 3.346388101577759 }, { "auxiliary_loss_clip": 0.01147238, "auxiliary_loss_mlp": 0.01032745, "balance_loss_clip": 1.04857528, "balance_loss_mlp": 1.0249722, "epoch": 0.4642578007575302, "flos": 19570385219520.0, "grad_norm": 1.9294135675174509, "language_loss": 0.77397752, "learning_rate": 2.3272479331224522e-06, "loss": 0.79577732, "num_input_tokens_seen": 83065825, "step": 3861, "time_per_iteration": 3.2632482051849365 }, { "auxiliary_loss_clip": 0.01182938, "auxiliary_loss_mlp": 0.01031445, "balance_loss_clip": 1.05145717, "balance_loss_mlp": 1.02353525, "epoch": 0.4643780436481693, "flos": 28186254204000.0, "grad_norm": 1.5532687421778202, "language_loss": 0.77928311, "learning_rate": 2.3264794350443817e-06, "loss": 0.80142689, "num_input_tokens_seen": 83087920, "step": 3862, "time_per_iteration": 2.5401058197021484 }, { "auxiliary_loss_clip": 0.01170485, "auxiliary_loss_mlp": 0.01026033, "balance_loss_clip": 1.04998541, "balance_loss_mlp": 1.01839757, "epoch": 0.46449828653880837, "flos": 25375274226720.0, "grad_norm": 1.912638574276096, "language_loss": 0.78266382, "learning_rate": 2.3257108874396396e-06, "loss": 0.80462897, "num_input_tokens_seen": 83109015, "step": 3863, "time_per_iteration": 2.5376229286193848 }, { "auxiliary_loss_clip": 0.01151904, "auxiliary_loss_mlp": 0.0102999, "balance_loss_clip": 1.04622459, "balance_loss_mlp": 1.02226281, "epoch": 0.4646185294294475, "flos": 16034333669280.0, "grad_norm": 2.078798608551786, "language_loss": 0.73506236, "learning_rate": 2.3249422904248152e-06, "loss": 0.7568813, "num_input_tokens_seen": 83127450, "step": 3864, "time_per_iteration": 3.233184576034546 }, { "auxiliary_loss_clip": 0.01172067, "auxiliary_loss_mlp": 0.01032168, "balance_loss_clip": 1.05095768, "balance_loss_mlp": 1.02337027, "epoch": 0.4647387723200866, "flos": 26363105477760.0, "grad_norm": 1.3807205766547994, "language_loss": 0.86889923, "learning_rate": 2.324173644116504e-06, "loss": 0.89094156, "num_input_tokens_seen": 83150300, "step": 3865, "time_per_iteration": 2.593144178390503 }, { "auxiliary_loss_clip": 0.01168719, "auxiliary_loss_mlp": 0.01031901, "balance_loss_clip": 1.05427766, "balance_loss_mlp": 1.02389026, "epoch": 0.46485901521072565, "flos": 27160213695840.0, "grad_norm": 1.6848785867254004, "language_loss": 0.81986529, "learning_rate": 2.3234049486313087e-06, "loss": 0.8418715, "num_input_tokens_seen": 83171750, "step": 3866, "time_per_iteration": 2.560277223587036 }, { "auxiliary_loss_clip": 0.01166761, "auxiliary_loss_mlp": 0.01027782, "balance_loss_clip": 1.05076337, "balance_loss_mlp": 1.02039671, "epoch": 0.46497925810136476, "flos": 24279854928480.0, "grad_norm": 1.7000129488260807, "language_loss": 0.75715935, "learning_rate": 2.322636204085839e-06, "loss": 0.77910483, "num_input_tokens_seen": 83191820, "step": 3867, "time_per_iteration": 3.2991504669189453 }, { "auxiliary_loss_clip": 0.01145621, "auxiliary_loss_mlp": 0.01034901, "balance_loss_clip": 1.04533172, "balance_loss_mlp": 1.02672279, "epoch": 0.46509950099200387, "flos": 16253137376640.0, "grad_norm": 2.0382966406971086, "language_loss": 0.78948909, "learning_rate": 2.3218674105967143e-06, "loss": 0.81129432, "num_input_tokens_seen": 83210085, "step": 3868, "time_per_iteration": 2.525920867919922 }, { "auxiliary_loss_clip": 0.01146628, "auxiliary_loss_mlp": 0.01029165, "balance_loss_clip": 1.0471307, "balance_loss_mlp": 1.02133894, "epoch": 0.4652197438826429, "flos": 23442274684800.0, "grad_norm": 1.5740043611948331, "language_loss": 0.83539915, "learning_rate": 2.3210985682805593e-06, "loss": 0.85715711, "num_input_tokens_seen": 83231865, "step": 3869, "time_per_iteration": 2.5927326679229736 }, { "auxiliary_loss_clip": 0.01185746, "auxiliary_loss_mlp": 0.01026959, "balance_loss_clip": 1.0554142, "balance_loss_mlp": 1.01915121, "epoch": 0.46533998677328203, "flos": 16216400714880.0, "grad_norm": 2.3351429459155484, "language_loss": 0.684084, "learning_rate": 2.320329677254007e-06, "loss": 0.70621109, "num_input_tokens_seen": 83249195, "step": 3870, "time_per_iteration": 2.439913272857666 }, { "auxiliary_loss_clip": 0.01185138, "auxiliary_loss_mlp": 0.01039289, "balance_loss_clip": 1.05424786, "balance_loss_mlp": 1.03085566, "epoch": 0.46546022966392114, "flos": 21141872608320.0, "grad_norm": 3.178756918629336, "language_loss": 0.7281884, "learning_rate": 2.319560737633697e-06, "loss": 0.75043267, "num_input_tokens_seen": 83267915, "step": 3871, "time_per_iteration": 2.4798331260681152 }, { "auxiliary_loss_clip": 0.01145935, "auxiliary_loss_mlp": 0.01028189, "balance_loss_clip": 1.04720736, "balance_loss_mlp": 1.01998448, "epoch": 0.4655804725545602, "flos": 41171947784640.0, "grad_norm": 1.5637609873006484, "language_loss": 0.67819345, "learning_rate": 2.3187917495362775e-06, "loss": 0.69993466, "num_input_tokens_seen": 83292325, "step": 3872, "time_per_iteration": 2.7216079235076904 }, { "auxiliary_loss_clip": 0.01123535, "auxiliary_loss_mlp": 0.01030072, "balance_loss_clip": 1.04698873, "balance_loss_mlp": 1.02207327, "epoch": 0.4657007154451993, "flos": 19570959890880.0, "grad_norm": 2.163952764702036, "language_loss": 0.76742744, "learning_rate": 2.318022713078403e-06, "loss": 0.78896356, "num_input_tokens_seen": 83306905, "step": 3873, "time_per_iteration": 2.5871567726135254 }, { "auxiliary_loss_clip": 0.01152763, "auxiliary_loss_mlp": 0.01029406, "balance_loss_clip": 1.04935646, "balance_loss_mlp": 1.02153254, "epoch": 0.4658209583358384, "flos": 15517829470080.0, "grad_norm": 2.157549782689641, "language_loss": 0.84822315, "learning_rate": 2.3172536283767354e-06, "loss": 0.87004483, "num_input_tokens_seen": 83320665, "step": 3874, "time_per_iteration": 2.47894024848938 }, { "auxiliary_loss_clip": 0.0114023, "auxiliary_loss_mlp": 0.01035203, "balance_loss_clip": 1.05175436, "balance_loss_mlp": 1.02639318, "epoch": 0.4659412012264775, "flos": 14903183384160.0, "grad_norm": 1.8187127921546218, "language_loss": 0.80660051, "learning_rate": 2.3164844955479447e-06, "loss": 0.82835484, "num_input_tokens_seen": 83336475, "step": 3875, "time_per_iteration": 2.5301437377929688 }, { "auxiliary_loss_clip": 0.01131543, "auxiliary_loss_mlp": 0.01026353, "balance_loss_clip": 1.04685521, "balance_loss_mlp": 1.01846695, "epoch": 0.4660614441171166, "flos": 24425616315840.0, "grad_norm": 1.674381915322148, "language_loss": 0.70697093, "learning_rate": 2.3157153147087082e-06, "loss": 0.72854984, "num_input_tokens_seen": 83358365, "step": 3876, "time_per_iteration": 2.62365984916687 }, { "auxiliary_loss_clip": 0.01133672, "auxiliary_loss_mlp": 0.01032356, "balance_loss_clip": 1.05073285, "balance_loss_mlp": 1.02390075, "epoch": 0.46618168700775564, "flos": 22091099515680.0, "grad_norm": 1.6970450898492704, "language_loss": 0.83123577, "learning_rate": 2.314946085975709e-06, "loss": 0.85289603, "num_input_tokens_seen": 83377345, "step": 3877, "time_per_iteration": 2.599400281906128 }, { "auxiliary_loss_clip": 0.01130195, "auxiliary_loss_mlp": 0.01028935, "balance_loss_clip": 1.04904878, "balance_loss_mlp": 1.02106988, "epoch": 0.46630192989839475, "flos": 26176979815680.0, "grad_norm": 1.7362022541551363, "language_loss": 0.82527941, "learning_rate": 2.3141768094656393e-06, "loss": 0.84687066, "num_input_tokens_seen": 83395920, "step": 3878, "time_per_iteration": 2.6072933673858643 }, { "auxiliary_loss_clip": 0.0110518, "auxiliary_loss_mlp": 0.01022244, "balance_loss_clip": 1.04262578, "balance_loss_mlp": 1.01426005, "epoch": 0.46642217278903386, "flos": 11509624777920.0, "grad_norm": 2.272625800883596, "language_loss": 0.83162922, "learning_rate": 2.3134074852951966e-06, "loss": 0.85290349, "num_input_tokens_seen": 83412510, "step": 3879, "time_per_iteration": 2.64689564704895 }, { "auxiliary_loss_clip": 0.01121153, "auxiliary_loss_mlp": 0.01030983, "balance_loss_clip": 1.04443908, "balance_loss_mlp": 1.02322865, "epoch": 0.4665424156796729, "flos": 32306823899040.0, "grad_norm": 1.6665071110773317, "language_loss": 0.77780032, "learning_rate": 2.312638113581088e-06, "loss": 0.79932171, "num_input_tokens_seen": 83432995, "step": 3880, "time_per_iteration": 2.6816117763519287 }, { "auxiliary_loss_clip": 0.01165283, "auxiliary_loss_mlp": 0.01030618, "balance_loss_clip": 1.04669023, "balance_loss_mlp": 1.02312541, "epoch": 0.46666265857031203, "flos": 18436181992800.0, "grad_norm": 2.6544144240322116, "language_loss": 0.78514999, "learning_rate": 2.311868694440027e-06, "loss": 0.807109, "num_input_tokens_seen": 83447415, "step": 3881, "time_per_iteration": 2.482848644256592 }, { "auxiliary_loss_clip": 0.01073422, "auxiliary_loss_mlp": 0.01006773, "balance_loss_clip": 1.01142991, "balance_loss_mlp": 1.00549161, "epoch": 0.46678290146095114, "flos": 68439001659840.0, "grad_norm": 0.7366090352075032, "language_loss": 0.62517428, "learning_rate": 2.3110992279887323e-06, "loss": 0.64597619, "num_input_tokens_seen": 83519340, "step": 3882, "time_per_iteration": 3.2253568172454834 }, { "auxiliary_loss_clip": 0.01147047, "auxiliary_loss_mlp": 0.01028674, "balance_loss_clip": 1.05064011, "balance_loss_mlp": 1.01984012, "epoch": 0.4669031443515902, "flos": 17712511181280.0, "grad_norm": 2.193050216839619, "language_loss": 0.84865236, "learning_rate": 2.310329714343932e-06, "loss": 0.87040961, "num_input_tokens_seen": 83535490, "step": 3883, "time_per_iteration": 2.5490922927856445 }, { "auxiliary_loss_clip": 0.011493, "auxiliary_loss_mlp": 0.01031678, "balance_loss_clip": 1.04910815, "balance_loss_mlp": 1.02404869, "epoch": 0.4670233872422293, "flos": 23947752377280.0, "grad_norm": 1.8342583764492988, "language_loss": 0.81941754, "learning_rate": 2.309560153622361e-06, "loss": 0.84122729, "num_input_tokens_seen": 83552400, "step": 3884, "time_per_iteration": 2.554975748062134 }, { "auxiliary_loss_clip": 0.01140933, "auxiliary_loss_mlp": 0.01031726, "balance_loss_clip": 1.050354, "balance_loss_mlp": 1.02304173, "epoch": 0.4671436301328684, "flos": 28111272368160.0, "grad_norm": 2.1326890909841776, "language_loss": 0.74130106, "learning_rate": 2.3087905459407602e-06, "loss": 0.76302767, "num_input_tokens_seen": 83571340, "step": 3885, "time_per_iteration": 2.6150963306427 }, { "auxiliary_loss_clip": 0.01060722, "auxiliary_loss_mlp": 0.01004644, "balance_loss_clip": 1.01030707, "balance_loss_mlp": 1.00338602, "epoch": 0.46726387302350747, "flos": 69369695415840.0, "grad_norm": 0.7903284411985536, "language_loss": 0.6296578, "learning_rate": 2.3080208914158795e-06, "loss": 0.65031147, "num_input_tokens_seen": 83634340, "step": 3886, "time_per_iteration": 3.9041240215301514 }, { "auxiliary_loss_clip": 0.01153882, "auxiliary_loss_mlp": 0.01029263, "balance_loss_clip": 1.05249274, "balance_loss_mlp": 1.02126098, "epoch": 0.4673841159141466, "flos": 25519670769600.0, "grad_norm": 2.425879230110787, "language_loss": 0.72847652, "learning_rate": 2.3072511901644753e-06, "loss": 0.75030804, "num_input_tokens_seen": 83653410, "step": 3887, "time_per_iteration": 3.3133552074432373 }, { "auxiliary_loss_clip": 0.01184873, "auxiliary_loss_mlp": 0.01023881, "balance_loss_clip": 1.05562222, "balance_loss_mlp": 1.01685977, "epoch": 0.4675043588047857, "flos": 24499269224160.0, "grad_norm": 1.9371062583882446, "language_loss": 0.80428791, "learning_rate": 2.306481442303309e-06, "loss": 0.82637548, "num_input_tokens_seen": 83672985, "step": 3888, "time_per_iteration": 2.4981191158294678 }, { "auxiliary_loss_clip": 0.01170648, "auxiliary_loss_mlp": 0.01026216, "balance_loss_clip": 1.0509851, "balance_loss_mlp": 1.01763904, "epoch": 0.46762460169542475, "flos": 20960775320640.0, "grad_norm": 1.8738985727247555, "language_loss": 0.73514056, "learning_rate": 2.3057116479491515e-06, "loss": 0.75710922, "num_input_tokens_seen": 83692395, "step": 3889, "time_per_iteration": 3.218602418899536 }, { "auxiliary_loss_clip": 0.01165686, "auxiliary_loss_mlp": 0.0102825, "balance_loss_clip": 1.04881549, "balance_loss_mlp": 1.02043283, "epoch": 0.46774484458606386, "flos": 19171669484160.0, "grad_norm": 1.8659577277920887, "language_loss": 0.75747335, "learning_rate": 2.30494180721878e-06, "loss": 0.77941275, "num_input_tokens_seen": 83709735, "step": 3890, "time_per_iteration": 2.4780216217041016 }, { "auxiliary_loss_clip": 0.01167309, "auxiliary_loss_mlp": 0.01030922, "balance_loss_clip": 1.05083275, "balance_loss_mlp": 1.02372789, "epoch": 0.4678650874767029, "flos": 17967692380800.0, "grad_norm": 1.8166928938747475, "language_loss": 0.8993476, "learning_rate": 2.3041719202289794e-06, "loss": 0.92132986, "num_input_tokens_seen": 83725910, "step": 3891, "time_per_iteration": 2.4812917709350586 }, { "auxiliary_loss_clip": 0.01172356, "auxiliary_loss_mlp": 0.0102534, "balance_loss_clip": 1.05391836, "balance_loss_mlp": 1.01755536, "epoch": 0.467985330367342, "flos": 21360819983520.0, "grad_norm": 1.5906475187205122, "language_loss": 0.8008734, "learning_rate": 2.30340198709654e-06, "loss": 0.82285035, "num_input_tokens_seen": 83745745, "step": 3892, "time_per_iteration": 2.4898364543914795 }, { "auxiliary_loss_clip": 0.01158842, "auxiliary_loss_mlp": 0.01035693, "balance_loss_clip": 1.04734528, "balance_loss_mlp": 1.02771175, "epoch": 0.46810557325798113, "flos": 20521838978400.0, "grad_norm": 2.371089959329799, "language_loss": 0.74066651, "learning_rate": 2.3026320079382605e-06, "loss": 0.76261187, "num_input_tokens_seen": 83762680, "step": 3893, "time_per_iteration": 2.541347026824951 }, { "auxiliary_loss_clip": 0.01180069, "auxiliary_loss_mlp": 0.01033018, "balance_loss_clip": 1.05271125, "balance_loss_mlp": 1.02514374, "epoch": 0.4682258161486202, "flos": 30117852984480.0, "grad_norm": 2.958906031799426, "language_loss": 0.75967705, "learning_rate": 2.3018619828709454e-06, "loss": 0.7818079, "num_input_tokens_seen": 83784220, "step": 3894, "time_per_iteration": 3.2704524993896484 }, { "auxiliary_loss_clip": 0.01167438, "auxiliary_loss_mlp": 0.0076263, "balance_loss_clip": 1.0547564, "balance_loss_mlp": 1.00030375, "epoch": 0.4683460590392593, "flos": 25293360417600.0, "grad_norm": 1.9672416958480128, "language_loss": 0.82239789, "learning_rate": 2.3010919120114084e-06, "loss": 0.84169859, "num_input_tokens_seen": 83800750, "step": 3895, "time_per_iteration": 2.509147882461548 }, { "auxiliary_loss_clip": 0.01165528, "auxiliary_loss_mlp": 0.01026496, "balance_loss_clip": 1.04814684, "balance_loss_mlp": 1.01807988, "epoch": 0.4684663019298984, "flos": 15368332718880.0, "grad_norm": 2.1857780278825465, "language_loss": 0.65969217, "learning_rate": 2.3003217954764672e-06, "loss": 0.68161237, "num_input_tokens_seen": 83815455, "step": 3896, "time_per_iteration": 2.4526915550231934 }, { "auxiliary_loss_clip": 0.01168528, "auxiliary_loss_mlp": 0.0103189, "balance_loss_clip": 1.04752898, "balance_loss_mlp": 1.02391458, "epoch": 0.46858654482053747, "flos": 27778846564320.0, "grad_norm": 1.6934892010063098, "language_loss": 0.79150057, "learning_rate": 2.299551633382949e-06, "loss": 0.8135047, "num_input_tokens_seen": 83835765, "step": 3897, "time_per_iteration": 2.550488233566284 }, { "auxiliary_loss_clip": 0.01148665, "auxiliary_loss_mlp": 0.0103291, "balance_loss_clip": 1.04862738, "balance_loss_mlp": 1.02467299, "epoch": 0.4687067877111766, "flos": 18040626949920.0, "grad_norm": 2.002213673395219, "language_loss": 0.85606432, "learning_rate": 2.2987814258476854e-06, "loss": 0.87788004, "num_input_tokens_seen": 83853565, "step": 3898, "time_per_iteration": 2.501339912414551 }, { "auxiliary_loss_clip": 0.0113088, "auxiliary_loss_mlp": 0.01031763, "balance_loss_clip": 1.04409683, "balance_loss_mlp": 1.02355552, "epoch": 0.4688270306018157, "flos": 16977382859520.0, "grad_norm": 2.415440358578403, "language_loss": 0.67548323, "learning_rate": 2.2980111729875177e-06, "loss": 0.69710958, "num_input_tokens_seen": 83869815, "step": 3899, "time_per_iteration": 2.5708131790161133 }, { "auxiliary_loss_clip": 0.01150517, "auxiliary_loss_mlp": 0.01028586, "balance_loss_clip": 1.05030084, "balance_loss_mlp": 1.02037263, "epoch": 0.46894727349245474, "flos": 17821643657760.0, "grad_norm": 1.7338924033433991, "language_loss": 0.82052946, "learning_rate": 2.2972408749192917e-06, "loss": 0.84232044, "num_input_tokens_seen": 83887545, "step": 3900, "time_per_iteration": 2.527144193649292 }, { "auxiliary_loss_clip": 0.01166963, "auxiliary_loss_mlp": 0.0076258, "balance_loss_clip": 1.0523113, "balance_loss_mlp": 1.00028062, "epoch": 0.46906751638309385, "flos": 21471353221440.0, "grad_norm": 2.4522634560597427, "language_loss": 0.6661045, "learning_rate": 2.296470531759861e-06, "loss": 0.68539989, "num_input_tokens_seen": 83905645, "step": 3901, "time_per_iteration": 2.5089166164398193 }, { "auxiliary_loss_clip": 0.01131508, "auxiliary_loss_mlp": 0.01026517, "balance_loss_clip": 1.04492152, "balance_loss_mlp": 1.01804137, "epoch": 0.46918775927373296, "flos": 20337832416960.0, "grad_norm": 1.8446052320469137, "language_loss": 0.78928131, "learning_rate": 2.2957001436260866e-06, "loss": 0.81086159, "num_input_tokens_seen": 83922705, "step": 3902, "time_per_iteration": 2.547924518585205 }, { "auxiliary_loss_clip": 0.01150635, "auxiliary_loss_mlp": 0.01028853, "balance_loss_clip": 1.04879344, "balance_loss_mlp": 1.02072239, "epoch": 0.469308002164372, "flos": 18403252528800.0, "grad_norm": 1.7374705039253837, "language_loss": 0.72978038, "learning_rate": 2.294929710634836e-06, "loss": 0.75157523, "num_input_tokens_seen": 83940795, "step": 3903, "time_per_iteration": 2.5400331020355225 }, { "auxiliary_loss_clip": 0.01166268, "auxiliary_loss_mlp": 0.01025588, "balance_loss_clip": 1.04873109, "balance_loss_mlp": 1.01752973, "epoch": 0.46942824505501113, "flos": 37962072487200.0, "grad_norm": 1.668004984009852, "language_loss": 0.61623955, "learning_rate": 2.2941592329029823e-06, "loss": 0.63815814, "num_input_tokens_seen": 83961900, "step": 3904, "time_per_iteration": 2.6487255096435547 }, { "auxiliary_loss_clip": 0.0116417, "auxiliary_loss_mlp": 0.01033145, "balance_loss_clip": 1.05082798, "balance_loss_mlp": 1.02524149, "epoch": 0.46954848794565024, "flos": 21872511310080.0, "grad_norm": 1.8012239536375676, "language_loss": 0.79293346, "learning_rate": 2.2933887105474067e-06, "loss": 0.81490672, "num_input_tokens_seen": 83980075, "step": 3905, "time_per_iteration": 2.5186848640441895 }, { "auxiliary_loss_clip": 0.01164422, "auxiliary_loss_mlp": 0.01026344, "balance_loss_clip": 1.05153012, "balance_loss_mlp": 1.01906037, "epoch": 0.4696687308362893, "flos": 22016548683360.0, "grad_norm": 1.6106568660269354, "language_loss": 0.81222653, "learning_rate": 2.2926181436849974e-06, "loss": 0.83413416, "num_input_tokens_seen": 83999430, "step": 3906, "time_per_iteration": 2.5024514198303223 }, { "auxiliary_loss_clip": 0.01167536, "auxiliary_loss_mlp": 0.01028757, "balance_loss_clip": 1.05159366, "balance_loss_mlp": 1.02106166, "epoch": 0.4697889737269284, "flos": 21613666580640.0, "grad_norm": 1.732516130026799, "language_loss": 0.7251687, "learning_rate": 2.2918475324326478e-06, "loss": 0.74713165, "num_input_tokens_seen": 84019150, "step": 3907, "time_per_iteration": 2.5241711139678955 }, { "auxiliary_loss_clip": 0.01173444, "auxiliary_loss_mlp": 0.00763013, "balance_loss_clip": 1.05235982, "balance_loss_mlp": 1.00039482, "epoch": 0.46990921661756746, "flos": 25228327579680.0, "grad_norm": 2.1852250489499476, "language_loss": 0.91121441, "learning_rate": 2.2910768769072603e-06, "loss": 0.93057901, "num_input_tokens_seen": 84037930, "step": 3908, "time_per_iteration": 2.5241293907165527 }, { "auxiliary_loss_clip": 0.01160625, "auxiliary_loss_mlp": 0.01030911, "balance_loss_clip": 1.0486238, "balance_loss_mlp": 1.02367163, "epoch": 0.47002945950820657, "flos": 13844033827200.0, "grad_norm": 1.9774774611380719, "language_loss": 0.76325142, "learning_rate": 2.2903061772257417e-06, "loss": 0.78516674, "num_input_tokens_seen": 84055915, "step": 3909, "time_per_iteration": 2.4805495738983154 }, { "auxiliary_loss_clip": 0.01167653, "auxiliary_loss_mlp": 0.01031308, "balance_loss_clip": 1.05213523, "balance_loss_mlp": 1.02350008, "epoch": 0.4701497023988457, "flos": 26247005111040.0, "grad_norm": 1.524240453578169, "language_loss": 0.78609788, "learning_rate": 2.289535433505007e-06, "loss": 0.80808747, "num_input_tokens_seen": 84077270, "step": 3910, "time_per_iteration": 2.532378673553467 }, { "auxiliary_loss_clip": 0.01153771, "auxiliary_loss_mlp": 0.01028853, "balance_loss_clip": 1.04907393, "balance_loss_mlp": 1.02098489, "epoch": 0.47026994528948474, "flos": 25629521585280.0, "grad_norm": 1.773467365226604, "language_loss": 0.63959169, "learning_rate": 2.2887646458619767e-06, "loss": 0.66141796, "num_input_tokens_seen": 84098635, "step": 3911, "time_per_iteration": 2.568596363067627 }, { "auxiliary_loss_clip": 0.01146952, "auxiliary_loss_mlp": 0.01030253, "balance_loss_clip": 1.04965973, "balance_loss_mlp": 1.02196193, "epoch": 0.47039018818012385, "flos": 20554409272800.0, "grad_norm": 2.049680897525602, "language_loss": 0.76487494, "learning_rate": 2.2879938144135797e-06, "loss": 0.78664696, "num_input_tokens_seen": 84114740, "step": 3912, "time_per_iteration": 3.347365379333496 }, { "auxiliary_loss_clip": 0.01137294, "auxiliary_loss_mlp": 0.00761914, "balance_loss_clip": 1.04679322, "balance_loss_mlp": 1.00034046, "epoch": 0.47051043107076296, "flos": 21577253171520.0, "grad_norm": 1.6334529435115108, "language_loss": 0.75456411, "learning_rate": 2.2872229392767496e-06, "loss": 0.77355617, "num_input_tokens_seen": 84134845, "step": 3913, "time_per_iteration": 3.3255321979522705 }, { "auxiliary_loss_clip": 0.01169701, "auxiliary_loss_mlp": 0.01027806, "balance_loss_clip": 1.04952621, "balance_loss_mlp": 1.02057588, "epoch": 0.470630673961402, "flos": 18953189029440.0, "grad_norm": 1.6548870962686206, "language_loss": 0.74486125, "learning_rate": 2.286452020568428e-06, "loss": 0.76683629, "num_input_tokens_seen": 84152920, "step": 3914, "time_per_iteration": 2.506464719772339 }, { "auxiliary_loss_clip": 0.01189705, "auxiliary_loss_mlp": 0.01031013, "balance_loss_clip": 1.05515599, "balance_loss_mlp": 1.02251935, "epoch": 0.4707509168520411, "flos": 19938973013760.0, "grad_norm": 1.8849910498421043, "language_loss": 0.7280916, "learning_rate": 2.2856810584055637e-06, "loss": 0.75029886, "num_input_tokens_seen": 84170455, "step": 3915, "time_per_iteration": 2.457611560821533 }, { "auxiliary_loss_clip": 0.01171312, "auxiliary_loss_mlp": 0.01032981, "balance_loss_clip": 1.05284953, "balance_loss_mlp": 1.02505064, "epoch": 0.47087115974268023, "flos": 40118760443040.0, "grad_norm": 1.4771547041600808, "language_loss": 0.67793715, "learning_rate": 2.2849100529051085e-06, "loss": 0.69998014, "num_input_tokens_seen": 84197390, "step": 3916, "time_per_iteration": 3.4725465774536133 }, { "auxiliary_loss_clip": 0.01177696, "auxiliary_loss_mlp": 0.01026941, "balance_loss_clip": 1.05040538, "balance_loss_mlp": 1.0197829, "epoch": 0.4709914026333193, "flos": 13552726554240.0, "grad_norm": 3.2995320547191267, "language_loss": 0.79769456, "learning_rate": 2.284139004184026e-06, "loss": 0.81974089, "num_input_tokens_seen": 84214620, "step": 3917, "time_per_iteration": 2.4635050296783447 }, { "auxiliary_loss_clip": 0.01183566, "auxiliary_loss_mlp": 0.01025625, "balance_loss_clip": 1.0541743, "balance_loss_mlp": 1.01772785, "epoch": 0.4711116455239584, "flos": 19974632166720.0, "grad_norm": 2.0699538549617422, "language_loss": 0.74607986, "learning_rate": 2.2833679123592814e-06, "loss": 0.76817179, "num_input_tokens_seen": 84231880, "step": 3918, "time_per_iteration": 2.486990451812744 }, { "auxiliary_loss_clip": 0.01153262, "auxiliary_loss_mlp": 0.01029273, "balance_loss_clip": 1.05059981, "balance_loss_mlp": 1.02123284, "epoch": 0.4712318884145975, "flos": 32124828687360.0, "grad_norm": 1.7362563398108815, "language_loss": 0.63503206, "learning_rate": 2.2825967775478508e-06, "loss": 0.65685737, "num_input_tokens_seen": 84252980, "step": 3919, "time_per_iteration": 2.6277947425842285 }, { "auxiliary_loss_clip": 0.01183756, "auxiliary_loss_mlp": 0.01034334, "balance_loss_clip": 1.05313563, "balance_loss_mlp": 1.02619183, "epoch": 0.47135213130523657, "flos": 20047854071520.0, "grad_norm": 2.139338956651933, "language_loss": 0.83520508, "learning_rate": 2.2818255998667135e-06, "loss": 0.85738593, "num_input_tokens_seen": 84271490, "step": 3920, "time_per_iteration": 3.227675199508667 }, { "auxiliary_loss_clip": 0.01170111, "auxiliary_loss_mlp": 0.01028706, "balance_loss_clip": 1.05456185, "balance_loss_mlp": 1.02137518, "epoch": 0.4714723741958757, "flos": 19426994351520.0, "grad_norm": 1.5236239469097606, "language_loss": 0.78824532, "learning_rate": 2.2810543794328566e-06, "loss": 0.81023347, "num_input_tokens_seen": 84290525, "step": 3921, "time_per_iteration": 2.531712770462036 }, { "auxiliary_loss_clip": 0.01174143, "auxiliary_loss_mlp": 0.01026847, "balance_loss_clip": 1.05370569, "balance_loss_mlp": 1.01881194, "epoch": 0.4715926170865148, "flos": 20373886656480.0, "grad_norm": 1.6534959837936545, "language_loss": 0.82432222, "learning_rate": 2.2802831163632735e-06, "loss": 0.84633213, "num_input_tokens_seen": 84309245, "step": 3922, "time_per_iteration": 2.5275635719299316 }, { "auxiliary_loss_clip": 0.01116436, "auxiliary_loss_mlp": 0.01029691, "balance_loss_clip": 1.04769754, "balance_loss_mlp": 1.02141774, "epoch": 0.47171285997715384, "flos": 22672887971520.0, "grad_norm": 1.6795906622331844, "language_loss": 0.74344629, "learning_rate": 2.279511810774965e-06, "loss": 0.7649076, "num_input_tokens_seen": 84330775, "step": 3923, "time_per_iteration": 2.6700899600982666 }, { "auxiliary_loss_clip": 0.01183614, "auxiliary_loss_mlp": 0.01022441, "balance_loss_clip": 1.05359793, "balance_loss_mlp": 1.01489496, "epoch": 0.47183310286779295, "flos": 21105423282240.0, "grad_norm": 2.1349905556887068, "language_loss": 0.71254635, "learning_rate": 2.2787404627849364e-06, "loss": 0.73460692, "num_input_tokens_seen": 84349985, "step": 3924, "time_per_iteration": 2.455476999282837 }, { "auxiliary_loss_clip": 0.011516, "auxiliary_loss_mlp": 0.01025817, "balance_loss_clip": 1.04938817, "balance_loss_mlp": 1.01856351, "epoch": 0.471953345758432, "flos": 21726570337920.0, "grad_norm": 1.715043891199663, "language_loss": 0.78942573, "learning_rate": 2.277969072510202e-06, "loss": 0.81119984, "num_input_tokens_seen": 84368965, "step": 3925, "time_per_iteration": 2.5437164306640625 }, { "auxiliary_loss_clip": 0.01154714, "auxiliary_loss_mlp": 0.0102811, "balance_loss_clip": 1.05005765, "balance_loss_mlp": 1.02106833, "epoch": 0.4720735886490711, "flos": 19861081904160.0, "grad_norm": 1.53845354617761, "language_loss": 0.80897218, "learning_rate": 2.2771976400677803e-06, "loss": 0.83080041, "num_input_tokens_seen": 84387795, "step": 3926, "time_per_iteration": 2.5151233673095703 }, { "auxiliary_loss_clip": 0.0111366, "auxiliary_loss_mlp": 0.01030088, "balance_loss_clip": 1.04275811, "balance_loss_mlp": 1.0228076, "epoch": 0.47219383153971023, "flos": 19171813152000.0, "grad_norm": 1.67887641127534, "language_loss": 0.78840828, "learning_rate": 2.2764261655746965e-06, "loss": 0.80984575, "num_input_tokens_seen": 84405290, "step": 3927, "time_per_iteration": 2.610752820968628 }, { "auxiliary_loss_clip": 0.01139158, "auxiliary_loss_mlp": 0.01034604, "balance_loss_clip": 1.04928398, "balance_loss_mlp": 1.0267005, "epoch": 0.4723140744303493, "flos": 23224009731840.0, "grad_norm": 1.552728667144456, "language_loss": 0.75777459, "learning_rate": 2.2756546491479832e-06, "loss": 0.77951223, "num_input_tokens_seen": 84426205, "step": 3928, "time_per_iteration": 2.663649320602417 }, { "auxiliary_loss_clip": 0.01183984, "auxiliary_loss_mlp": 0.00762736, "balance_loss_clip": 1.05216527, "balance_loss_mlp": 1.00048327, "epoch": 0.4724343173209884, "flos": 18223268666880.0, "grad_norm": 3.6387458209614847, "language_loss": 0.80437779, "learning_rate": 2.274883090904679e-06, "loss": 0.82384497, "num_input_tokens_seen": 84443970, "step": 3929, "time_per_iteration": 2.459042549133301 }, { "auxiliary_loss_clip": 0.01185792, "auxiliary_loss_mlp": 0.01029279, "balance_loss_clip": 1.05514872, "balance_loss_mlp": 1.02128589, "epoch": 0.4725545602116275, "flos": 21251041001760.0, "grad_norm": 2.0943985770616766, "language_loss": 0.67564797, "learning_rate": 2.2741114909618283e-06, "loss": 0.69779873, "num_input_tokens_seen": 84459865, "step": 3930, "time_per_iteration": 2.4532694816589355 }, { "auxiliary_loss_clip": 0.01141839, "auxiliary_loss_mlp": 0.01029876, "balance_loss_clip": 1.04983854, "balance_loss_mlp": 1.02197242, "epoch": 0.47267480310226656, "flos": 21434005971360.0, "grad_norm": 1.8092340730175083, "language_loss": 0.72050732, "learning_rate": 2.2733398494364828e-06, "loss": 0.74222445, "num_input_tokens_seen": 84479110, "step": 3931, "time_per_iteration": 2.5739574432373047 }, { "auxiliary_loss_clip": 0.01150759, "auxiliary_loss_mlp": 0.0102903, "balance_loss_clip": 1.05126977, "balance_loss_mlp": 1.0218358, "epoch": 0.47279504599290567, "flos": 18770511395520.0, "grad_norm": 2.8754368583077015, "language_loss": 0.84658301, "learning_rate": 2.272568166445699e-06, "loss": 0.8683809, "num_input_tokens_seen": 84497675, "step": 3932, "time_per_iteration": 2.495279312133789 }, { "auxiliary_loss_clip": 0.0116839, "auxiliary_loss_mlp": 0.01028269, "balance_loss_clip": 1.05134046, "balance_loss_mlp": 1.02041876, "epoch": 0.4729152888835448, "flos": 21105746534880.0, "grad_norm": 2.014665939351259, "language_loss": 0.64327979, "learning_rate": 2.271796442106541e-06, "loss": 0.66524637, "num_input_tokens_seen": 84517030, "step": 3933, "time_per_iteration": 2.5064477920532227 }, { "auxiliary_loss_clip": 0.01039575, "auxiliary_loss_mlp": 0.01005688, "balance_loss_clip": 1.01167917, "balance_loss_mlp": 1.00439477, "epoch": 0.47303553177418384, "flos": 70201888115520.0, "grad_norm": 0.8032908600302903, "language_loss": 0.56591427, "learning_rate": 2.271024676536079e-06, "loss": 0.58636695, "num_input_tokens_seen": 84577290, "step": 3934, "time_per_iteration": 3.1223533153533936 }, { "auxiliary_loss_clip": 0.01164443, "auxiliary_loss_mlp": 0.01030301, "balance_loss_clip": 1.056494, "balance_loss_mlp": 1.02152157, "epoch": 0.47315577466482295, "flos": 22455125856000.0, "grad_norm": 1.9097515350580194, "language_loss": 0.73050523, "learning_rate": 2.2702528698513894e-06, "loss": 0.75245261, "num_input_tokens_seen": 84598415, "step": 3935, "time_per_iteration": 2.571075201034546 }, { "auxiliary_loss_clip": 0.01153214, "auxiliary_loss_mlp": 0.01041885, "balance_loss_clip": 1.04617333, "balance_loss_mlp": 1.03357601, "epoch": 0.47327601755546206, "flos": 24352861331520.0, "grad_norm": 1.8015630078559544, "language_loss": 0.78972143, "learning_rate": 2.269481022169554e-06, "loss": 0.81167245, "num_input_tokens_seen": 84617010, "step": 3936, "time_per_iteration": 2.5721518993377686 }, { "auxiliary_loss_clip": 0.01164209, "auxiliary_loss_mlp": 0.01033498, "balance_loss_clip": 1.05168271, "balance_loss_mlp": 1.02474809, "epoch": 0.4733962604461011, "flos": 22926776160480.0, "grad_norm": 1.6860742601210952, "language_loss": 0.80465615, "learning_rate": 2.2687091336076614e-06, "loss": 0.82663321, "num_input_tokens_seen": 84636350, "step": 3937, "time_per_iteration": 2.5629184246063232 }, { "auxiliary_loss_clip": 0.01165552, "auxiliary_loss_mlp": 0.01031652, "balance_loss_clip": 1.05059099, "balance_loss_mlp": 1.02393937, "epoch": 0.4735165033367402, "flos": 18327372768960.0, "grad_norm": 1.8214266294360344, "language_loss": 0.79785168, "learning_rate": 2.267937204282807e-06, "loss": 0.8198238, "num_input_tokens_seen": 84653490, "step": 3938, "time_per_iteration": 3.274352788925171 }, { "auxiliary_loss_clip": 0.01175321, "auxiliary_loss_mlp": 0.01031724, "balance_loss_clip": 1.05335045, "balance_loss_mlp": 1.02346313, "epoch": 0.4736367462273793, "flos": 23037022062720.0, "grad_norm": 2.245823799975877, "language_loss": 0.78791165, "learning_rate": 2.2671652343120926e-06, "loss": 0.80998212, "num_input_tokens_seen": 84673965, "step": 3939, "time_per_iteration": 3.2745163440704346 }, { "auxiliary_loss_clip": 0.01182744, "auxiliary_loss_mlp": 0.01032054, "balance_loss_clip": 1.05415416, "balance_loss_mlp": 1.02392364, "epoch": 0.4737569891180184, "flos": 25374340385760.0, "grad_norm": 1.6124775985633397, "language_loss": 0.80788922, "learning_rate": 2.2663932238126236e-06, "loss": 0.83003724, "num_input_tokens_seen": 84692525, "step": 3940, "time_per_iteration": 2.511505603790283 }, { "auxiliary_loss_clip": 0.01166581, "auxiliary_loss_mlp": 0.0102967, "balance_loss_clip": 1.0492543, "balance_loss_mlp": 1.02184701, "epoch": 0.4738772320086575, "flos": 25849330967520.0, "grad_norm": 1.3885816164467246, "language_loss": 0.80187416, "learning_rate": 2.265621172901515e-06, "loss": 0.82383662, "num_input_tokens_seen": 84715640, "step": 3941, "time_per_iteration": 2.5416457653045654 }, { "auxiliary_loss_clip": 0.01191093, "auxiliary_loss_mlp": 0.01036658, "balance_loss_clip": 1.05924368, "balance_loss_mlp": 1.02852178, "epoch": 0.47399747489929656, "flos": 27564424726080.0, "grad_norm": 3.0271087401772565, "language_loss": 0.71331596, "learning_rate": 2.2648490816958854e-06, "loss": 0.73559344, "num_input_tokens_seen": 84736635, "step": 3942, "time_per_iteration": 3.2910635471343994 }, { "auxiliary_loss_clip": 0.01168178, "auxiliary_loss_mlp": 0.01031811, "balance_loss_clip": 1.04969168, "balance_loss_mlp": 1.02366924, "epoch": 0.47411771778993567, "flos": 24863654734080.0, "grad_norm": 2.4374871780873146, "language_loss": 0.73465955, "learning_rate": 2.264076950312861e-06, "loss": 0.75665951, "num_input_tokens_seen": 84755445, "step": 3943, "time_per_iteration": 2.52079176902771 }, { "auxiliary_loss_clip": 0.01162948, "auxiliary_loss_mlp": 0.01031474, "balance_loss_clip": 1.05243456, "balance_loss_mlp": 1.02281904, "epoch": 0.4742379606805748, "flos": 22748013475200.0, "grad_norm": 1.9888306553591324, "language_loss": 0.82518137, "learning_rate": 2.2633047788695727e-06, "loss": 0.84712559, "num_input_tokens_seen": 84775750, "step": 3944, "time_per_iteration": 2.531388759613037 }, { "auxiliary_loss_clip": 0.01156738, "auxiliary_loss_mlp": 0.01031669, "balance_loss_clip": 1.05345547, "balance_loss_mlp": 1.02430165, "epoch": 0.47435820357121383, "flos": 19681133959200.0, "grad_norm": 1.6733989863062657, "language_loss": 0.6407479, "learning_rate": 2.262532567483159e-06, "loss": 0.66263199, "num_input_tokens_seen": 84794310, "step": 3945, "time_per_iteration": 2.5083730220794678 }, { "auxiliary_loss_clip": 0.01186188, "auxiliary_loss_mlp": 0.0076359, "balance_loss_clip": 1.05563319, "balance_loss_mlp": 1.00054765, "epoch": 0.47447844646185294, "flos": 25228722666240.0, "grad_norm": 2.143820131005961, "language_loss": 0.80025667, "learning_rate": 2.2617603162707635e-06, "loss": 0.81975448, "num_input_tokens_seen": 84814720, "step": 3946, "time_per_iteration": 3.196302890777588 }, { "auxiliary_loss_clip": 0.01182282, "auxiliary_loss_mlp": 0.01028891, "balance_loss_clip": 1.05326104, "balance_loss_mlp": 1.02124357, "epoch": 0.47459868935249205, "flos": 24570623447040.0, "grad_norm": 1.6602715708711473, "language_loss": 0.82633346, "learning_rate": 2.2609880253495363e-06, "loss": 0.84844518, "num_input_tokens_seen": 84834355, "step": 3947, "time_per_iteration": 2.515347480773926 }, { "auxiliary_loss_clip": 0.01145787, "auxiliary_loss_mlp": 0.01028883, "balance_loss_clip": 1.04541707, "balance_loss_mlp": 1.02089024, "epoch": 0.4747189322431311, "flos": 20558503806240.0, "grad_norm": 2.46222281134205, "language_loss": 0.8645224, "learning_rate": 2.260215694836633e-06, "loss": 0.88626909, "num_input_tokens_seen": 84853530, "step": 3948, "time_per_iteration": 2.5494139194488525 }, { "auxiliary_loss_clip": 0.01128672, "auxiliary_loss_mlp": 0.00763514, "balance_loss_clip": 1.04488778, "balance_loss_mlp": 1.00041699, "epoch": 0.4748391751337702, "flos": 25995236022720.0, "grad_norm": 2.195514558578593, "language_loss": 0.6485604, "learning_rate": 2.2594433248492157e-06, "loss": 0.66748226, "num_input_tokens_seen": 84872505, "step": 3949, "time_per_iteration": 2.6540606021881104 }, { "auxiliary_loss_clip": 0.01175378, "auxiliary_loss_mlp": 0.01030369, "balance_loss_clip": 1.05201089, "balance_loss_mlp": 1.02172875, "epoch": 0.47495941802440933, "flos": 22821055795200.0, "grad_norm": 1.9005770223548626, "language_loss": 0.80335176, "learning_rate": 2.2586709155044527e-06, "loss": 0.82540923, "num_input_tokens_seen": 84893105, "step": 3950, "time_per_iteration": 2.517587423324585 }, { "auxiliary_loss_clip": 0.01184635, "auxiliary_loss_mlp": 0.01026132, "balance_loss_clip": 1.05462027, "balance_loss_mlp": 1.01849675, "epoch": 0.4750796609150484, "flos": 27891786238560.0, "grad_norm": 1.7236727903578994, "language_loss": 0.75980383, "learning_rate": 2.2578984669195167e-06, "loss": 0.78191155, "num_input_tokens_seen": 84914070, "step": 3951, "time_per_iteration": 2.5469565391540527 }, { "auxiliary_loss_clip": 0.01164877, "auxiliary_loss_mlp": 0.01027848, "balance_loss_clip": 1.04785478, "balance_loss_mlp": 1.02081442, "epoch": 0.4751999038056875, "flos": 35660090064480.0, "grad_norm": 1.6682044424986298, "language_loss": 0.67977995, "learning_rate": 2.2571259792115887e-06, "loss": 0.70170718, "num_input_tokens_seen": 84935290, "step": 3952, "time_per_iteration": 2.6143341064453125 }, { "auxiliary_loss_clip": 0.01164923, "auxiliary_loss_mlp": 0.01031606, "balance_loss_clip": 1.05067277, "balance_loss_mlp": 1.02441502, "epoch": 0.4753201466963266, "flos": 22090884013920.0, "grad_norm": 6.82110078806485, "language_loss": 0.78889555, "learning_rate": 2.2563534524978544e-06, "loss": 0.81086087, "num_input_tokens_seen": 84952760, "step": 3953, "time_per_iteration": 2.5286285877227783 }, { "auxiliary_loss_clip": 0.01137031, "auxiliary_loss_mlp": 0.01024315, "balance_loss_clip": 1.05086911, "balance_loss_mlp": 1.01634562, "epoch": 0.47544038958696566, "flos": 30190859387520.0, "grad_norm": 33.146433111570936, "language_loss": 0.70916098, "learning_rate": 2.2555808868955052e-06, "loss": 0.73077446, "num_input_tokens_seen": 84974890, "step": 3954, "time_per_iteration": 2.613459348678589 }, { "auxiliary_loss_clip": 0.01131101, "auxiliary_loss_mlp": 0.01032886, "balance_loss_clip": 1.05006135, "balance_loss_mlp": 1.02442205, "epoch": 0.47556063247760477, "flos": 23472223041120.0, "grad_norm": 2.4021272770759916, "language_loss": 0.73595178, "learning_rate": 2.254808282521738e-06, "loss": 0.75759172, "num_input_tokens_seen": 84993640, "step": 3955, "time_per_iteration": 2.6419057846069336 }, { "auxiliary_loss_clip": 0.01144726, "auxiliary_loss_mlp": 0.00762394, "balance_loss_clip": 1.04816389, "balance_loss_mlp": 1.00034964, "epoch": 0.4756808753682438, "flos": 25155213425760.0, "grad_norm": 1.683495276623269, "language_loss": 0.80905998, "learning_rate": 2.2540356394937573e-06, "loss": 0.82813114, "num_input_tokens_seen": 85012340, "step": 3956, "time_per_iteration": 2.600496530532837 }, { "auxiliary_loss_clip": 0.01149092, "auxiliary_loss_mlp": 0.01028492, "balance_loss_clip": 1.0501523, "balance_loss_mlp": 1.01964664, "epoch": 0.47580111825888294, "flos": 15669732657600.0, "grad_norm": 2.150297444104931, "language_loss": 0.84174478, "learning_rate": 2.253262957928772e-06, "loss": 0.86352056, "num_input_tokens_seen": 85029225, "step": 3957, "time_per_iteration": 2.5599679946899414 }, { "auxiliary_loss_clip": 0.01149564, "auxiliary_loss_mlp": 0.01026146, "balance_loss_clip": 1.04688132, "balance_loss_mlp": 1.01811147, "epoch": 0.47592136114952205, "flos": 17636559587520.0, "grad_norm": 1.813191045795013, "language_loss": 0.72274041, "learning_rate": 2.2524902379439976e-06, "loss": 0.74449754, "num_input_tokens_seen": 85047895, "step": 3958, "time_per_iteration": 2.524853229522705 }, { "auxiliary_loss_clip": 0.01025041, "auxiliary_loss_mlp": 0.01008081, "balance_loss_clip": 1.02237916, "balance_loss_mlp": 1.00673985, "epoch": 0.4760416040401611, "flos": 61417169526720.0, "grad_norm": 0.7414002593774108, "language_loss": 0.63734794, "learning_rate": 2.251717479656655e-06, "loss": 0.65767914, "num_input_tokens_seen": 85112690, "step": 3959, "time_per_iteration": 3.247274875640869 }, { "auxiliary_loss_clip": 0.0118694, "auxiliary_loss_mlp": 0.01026068, "balance_loss_clip": 1.05508077, "balance_loss_mlp": 1.01826024, "epoch": 0.4761618469308002, "flos": 18405874466880.0, "grad_norm": 1.9103512114768713, "language_loss": 0.7639963, "learning_rate": 2.2509446831839704e-06, "loss": 0.78612638, "num_input_tokens_seen": 85132130, "step": 3960, "time_per_iteration": 2.45940899848938 }, { "auxiliary_loss_clip": 0.01158292, "auxiliary_loss_mlp": 0.01030709, "balance_loss_clip": 1.04937434, "balance_loss_mlp": 1.02266514, "epoch": 0.4762820898214393, "flos": 18040914285600.0, "grad_norm": 5.14696483778665, "language_loss": 0.82772326, "learning_rate": 2.250171848643177e-06, "loss": 0.84961319, "num_input_tokens_seen": 85149420, "step": 3961, "time_per_iteration": 2.539044141769409 }, { "auxiliary_loss_clip": 0.01152084, "auxiliary_loss_mlp": 0.01030084, "balance_loss_clip": 1.05204844, "balance_loss_mlp": 1.02292228, "epoch": 0.4764023327120784, "flos": 19318256961600.0, "grad_norm": 3.061781000431281, "language_loss": 0.8593297, "learning_rate": 2.249398976151513e-06, "loss": 0.88115132, "num_input_tokens_seen": 85166970, "step": 3962, "time_per_iteration": 2.538337469100952 }, { "auxiliary_loss_clip": 0.01182389, "auxiliary_loss_mlp": 0.01033169, "balance_loss_clip": 1.05361247, "balance_loss_mlp": 1.02548027, "epoch": 0.4765225756027175, "flos": 22747259219040.0, "grad_norm": 2.0604009579347204, "language_loss": 0.78543496, "learning_rate": 2.248626065826223e-06, "loss": 0.8075906, "num_input_tokens_seen": 85185175, "step": 3963, "time_per_iteration": 2.479644298553467 }, { "auxiliary_loss_clip": 0.0107361, "auxiliary_loss_mlp": 0.01000996, "balance_loss_clip": 1.01195562, "balance_loss_mlp": 0.99965507, "epoch": 0.4766428184933566, "flos": 65933402015520.0, "grad_norm": 0.766192426207629, "language_loss": 0.62572867, "learning_rate": 2.2478531177845564e-06, "loss": 0.64647472, "num_input_tokens_seen": 85246170, "step": 3964, "time_per_iteration": 3.774199962615967 }, { "auxiliary_loss_clip": 0.01159393, "auxiliary_loss_mlp": 0.0103617, "balance_loss_clip": 1.05222762, "balance_loss_mlp": 1.02843904, "epoch": 0.47676306138399566, "flos": 24136499977440.0, "grad_norm": 2.081629133942867, "language_loss": 0.84917402, "learning_rate": 2.247080132143769e-06, "loss": 0.87112963, "num_input_tokens_seen": 85268525, "step": 3965, "time_per_iteration": 3.3363771438598633 }, { "auxiliary_loss_clip": 0.01142627, "auxiliary_loss_mlp": 0.01025657, "balance_loss_clip": 1.04640675, "balance_loss_mlp": 1.01723528, "epoch": 0.47688330427463477, "flos": 12604325736960.0, "grad_norm": 2.150748197105462, "language_loss": 0.69512761, "learning_rate": 2.246307109021121e-06, "loss": 0.71681046, "num_input_tokens_seen": 85285930, "step": 3966, "time_per_iteration": 2.547494649887085 }, { "auxiliary_loss_clip": 0.0115122, "auxiliary_loss_mlp": 0.01030574, "balance_loss_clip": 1.04874349, "balance_loss_mlp": 1.02311754, "epoch": 0.4770035471652739, "flos": 21390588755040.0, "grad_norm": 1.6286069975889028, "language_loss": 0.82295215, "learning_rate": 2.2455340485338817e-06, "loss": 0.84477007, "num_input_tokens_seen": 85303565, "step": 3967, "time_per_iteration": 2.5290274620056152 }, { "auxiliary_loss_clip": 0.01171012, "auxiliary_loss_mlp": 0.01024306, "balance_loss_clip": 1.05130506, "balance_loss_mlp": 1.01696324, "epoch": 0.47712379005591293, "flos": 25156255017600.0, "grad_norm": 1.9204140037310518, "language_loss": 0.68032742, "learning_rate": 2.244760950799322e-06, "loss": 0.70228064, "num_input_tokens_seen": 85321835, "step": 3968, "time_per_iteration": 3.277637243270874 }, { "auxiliary_loss_clip": 0.01128092, "auxiliary_loss_mlp": 0.01028516, "balance_loss_clip": 1.04773605, "balance_loss_mlp": 1.02037978, "epoch": 0.47724403294655204, "flos": 22054326936960.0, "grad_norm": 1.7365263074368928, "language_loss": 0.72718215, "learning_rate": 2.2439878159347203e-06, "loss": 0.74874824, "num_input_tokens_seen": 85341260, "step": 3969, "time_per_iteration": 2.5919840335845947 }, { "auxiliary_loss_clip": 0.01071983, "auxiliary_loss_mlp": 0.01000458, "balance_loss_clip": 1.01057029, "balance_loss_mlp": 0.99909335, "epoch": 0.4773642758371911, "flos": 70229400506880.0, "grad_norm": 0.7315992381905618, "language_loss": 0.55304837, "learning_rate": 2.2432146440573616e-06, "loss": 0.57377279, "num_input_tokens_seen": 85407220, "step": 3970, "time_per_iteration": 3.201645851135254 }, { "auxiliary_loss_clip": 0.0115535, "auxiliary_loss_mlp": 0.01030936, "balance_loss_clip": 1.05221665, "balance_loss_mlp": 1.02279425, "epoch": 0.4774845187278302, "flos": 23548605638400.0, "grad_norm": 1.7542759691207441, "language_loss": 0.66631192, "learning_rate": 2.242441435284534e-06, "loss": 0.68817472, "num_input_tokens_seen": 85426095, "step": 3971, "time_per_iteration": 2.5626115798950195 }, { "auxiliary_loss_clip": 0.01173258, "auxiliary_loss_mlp": 0.01034881, "balance_loss_clip": 1.05362689, "balance_loss_mlp": 1.02667308, "epoch": 0.4776047616184693, "flos": 23075374987680.0, "grad_norm": 2.1840296437950606, "language_loss": 0.85376114, "learning_rate": 2.2416681897335337e-06, "loss": 0.87584251, "num_input_tokens_seen": 85444245, "step": 3972, "time_per_iteration": 3.2138402462005615 }, { "auxiliary_loss_clip": 0.01130559, "auxiliary_loss_mlp": 0.01029015, "balance_loss_clip": 1.05220377, "balance_loss_mlp": 1.02096844, "epoch": 0.4777250045091084, "flos": 31898123248800.0, "grad_norm": 1.7949372628247948, "language_loss": 0.66776371, "learning_rate": 2.240894907521661e-06, "loss": 0.68935949, "num_input_tokens_seen": 85463325, "step": 3973, "time_per_iteration": 2.683502435684204 }, { "auxiliary_loss_clip": 0.0115526, "auxiliary_loss_mlp": 0.01026894, "balance_loss_clip": 1.05049324, "balance_loss_mlp": 1.01828122, "epoch": 0.4778452473997475, "flos": 24278166831360.0, "grad_norm": 2.275643047123937, "language_loss": 0.63775343, "learning_rate": 2.240121588766223e-06, "loss": 0.65957493, "num_input_tokens_seen": 85483375, "step": 3974, "time_per_iteration": 2.5618324279785156 }, { "auxiliary_loss_clip": 0.01149612, "auxiliary_loss_mlp": 0.01029141, "balance_loss_clip": 1.04857135, "balance_loss_mlp": 1.02173233, "epoch": 0.4779654902903866, "flos": 31575035854560.0, "grad_norm": 1.8906949423464736, "language_loss": 0.71400809, "learning_rate": 2.239348233584531e-06, "loss": 0.73579562, "num_input_tokens_seen": 85504230, "step": 3975, "time_per_iteration": 2.613981008529663 }, { "auxiliary_loss_clip": 0.01172398, "auxiliary_loss_mlp": 0.01030812, "balance_loss_clip": 1.05295527, "balance_loss_mlp": 1.02294397, "epoch": 0.47808573318102565, "flos": 19500431758080.0, "grad_norm": 1.8428699206563632, "language_loss": 0.80654389, "learning_rate": 2.2385748420939013e-06, "loss": 0.82857597, "num_input_tokens_seen": 85523425, "step": 3976, "time_per_iteration": 2.480757236480713 }, { "auxiliary_loss_clip": 0.01183921, "auxiliary_loss_mlp": 0.01031494, "balance_loss_clip": 1.0571686, "balance_loss_mlp": 1.02439499, "epoch": 0.47820597607166476, "flos": 22601138662080.0, "grad_norm": 1.6833731776794962, "language_loss": 0.71916091, "learning_rate": 2.2378014144116583e-06, "loss": 0.74131507, "num_input_tokens_seen": 85542235, "step": 3977, "time_per_iteration": 2.4810941219329834 }, { "auxiliary_loss_clip": 0.01190783, "auxiliary_loss_mlp": 0.01031419, "balance_loss_clip": 1.0583626, "balance_loss_mlp": 1.02405834, "epoch": 0.4783262189623039, "flos": 23003015089920.0, "grad_norm": 1.9122108017426405, "language_loss": 0.79359573, "learning_rate": 2.23702795065513e-06, "loss": 0.81581777, "num_input_tokens_seen": 85561815, "step": 3978, "time_per_iteration": 2.470142126083374 }, { "auxiliary_loss_clip": 0.01062379, "auxiliary_loss_mlp": 0.01005364, "balance_loss_clip": 1.00960994, "balance_loss_mlp": 1.00394523, "epoch": 0.47844646185294293, "flos": 49772809868160.0, "grad_norm": 0.993284520216907, "language_loss": 0.67477894, "learning_rate": 2.2362544509416493e-06, "loss": 0.69545639, "num_input_tokens_seen": 85613930, "step": 3979, "time_per_iteration": 2.9678423404693604 }, { "auxiliary_loss_clip": 0.01146828, "auxiliary_loss_mlp": 0.01028664, "balance_loss_clip": 1.04744196, "balance_loss_mlp": 1.02136803, "epoch": 0.47856670474358204, "flos": 20229561947520.0, "grad_norm": 2.221842327106322, "language_loss": 0.82891798, "learning_rate": 2.2354809153885572e-06, "loss": 0.85067296, "num_input_tokens_seen": 85631000, "step": 3980, "time_per_iteration": 2.5304598808288574 }, { "auxiliary_loss_clip": 0.01170363, "auxiliary_loss_mlp": 0.01029922, "balance_loss_clip": 1.05297029, "balance_loss_mlp": 1.02160692, "epoch": 0.47868694763422115, "flos": 20990939178720.0, "grad_norm": 1.7843070520637236, "language_loss": 0.82573187, "learning_rate": 2.234707344113197e-06, "loss": 0.84773469, "num_input_tokens_seen": 85649095, "step": 3981, "time_per_iteration": 2.487266778945923 }, { "auxiliary_loss_clip": 0.01179474, "auxiliary_loss_mlp": 0.01025944, "balance_loss_clip": 1.05218887, "balance_loss_mlp": 1.01894689, "epoch": 0.4788071905248602, "flos": 19026554602080.0, "grad_norm": 1.6309021954347074, "language_loss": 0.77509248, "learning_rate": 2.233933737232919e-06, "loss": 0.79714668, "num_input_tokens_seen": 85666875, "step": 3982, "time_per_iteration": 2.454329490661621 }, { "auxiliary_loss_clip": 0.01118478, "auxiliary_loss_mlp": 0.00762754, "balance_loss_clip": 1.04542863, "balance_loss_mlp": 1.00042689, "epoch": 0.4789274334154993, "flos": 23002224916800.0, "grad_norm": 1.7610372386388677, "language_loss": 0.78315961, "learning_rate": 2.2331600948650793e-06, "loss": 0.80197191, "num_input_tokens_seen": 85687020, "step": 3983, "time_per_iteration": 2.601560354232788 }, { "auxiliary_loss_clip": 0.01128872, "auxiliary_loss_mlp": 0.00763769, "balance_loss_clip": 1.04727328, "balance_loss_mlp": 1.00040579, "epoch": 0.4790476763061384, "flos": 23075590489440.0, "grad_norm": 1.4367593907719105, "language_loss": 0.80044347, "learning_rate": 2.2323864171270386e-06, "loss": 0.81936991, "num_input_tokens_seen": 85708290, "step": 3984, "time_per_iteration": 2.6611573696136475 }, { "auxiliary_loss_clip": 0.01144668, "auxiliary_loss_mlp": 0.01032659, "balance_loss_clip": 1.04808605, "balance_loss_mlp": 1.02464783, "epoch": 0.4791679191967775, "flos": 21179291692320.0, "grad_norm": 1.8141732448123125, "language_loss": 0.72522426, "learning_rate": 2.231612704136164e-06, "loss": 0.74699754, "num_input_tokens_seen": 85728660, "step": 3985, "time_per_iteration": 2.5996570587158203 }, { "auxiliary_loss_clip": 0.01164855, "auxiliary_loss_mlp": 0.01025236, "balance_loss_clip": 1.05044913, "balance_loss_mlp": 1.01736248, "epoch": 0.4792881620874166, "flos": 22301498654400.0, "grad_norm": 2.6568763833964937, "language_loss": 0.75594556, "learning_rate": 2.2308389560098253e-06, "loss": 0.77784646, "num_input_tokens_seen": 85745035, "step": 3986, "time_per_iteration": 2.4964137077331543 }, { "auxiliary_loss_clip": 0.01143545, "auxiliary_loss_mlp": 0.01028762, "balance_loss_clip": 1.05087519, "balance_loss_mlp": 1.0203402, "epoch": 0.47940840497805565, "flos": 17420880655680.0, "grad_norm": 1.9063715937320749, "language_loss": 0.76928031, "learning_rate": 2.2300651728654008e-06, "loss": 0.79100335, "num_input_tokens_seen": 85760295, "step": 3987, "time_per_iteration": 2.530477285385132 }, { "auxiliary_loss_clip": 0.01055786, "auxiliary_loss_mlp": 0.00752779, "balance_loss_clip": 1.0088973, "balance_loss_mlp": 0.99982566, "epoch": 0.47952864786869476, "flos": 65358186363360.0, "grad_norm": 0.7468897829871619, "language_loss": 0.60237336, "learning_rate": 2.229291354820272e-06, "loss": 0.62045902, "num_input_tokens_seen": 85821305, "step": 3988, "time_per_iteration": 3.1134891510009766 }, { "auxiliary_loss_clip": 0.01170025, "auxiliary_loss_mlp": 0.01029172, "balance_loss_clip": 1.05081344, "balance_loss_mlp": 1.02096438, "epoch": 0.47964889075933387, "flos": 16799805433920.0, "grad_norm": 2.046088854949112, "language_loss": 0.7633419, "learning_rate": 2.228517501991828e-06, "loss": 0.78533387, "num_input_tokens_seen": 85840105, "step": 3989, "time_per_iteration": 2.473224401473999 }, { "auxiliary_loss_clip": 0.01047105, "auxiliary_loss_mlp": 0.01003604, "balance_loss_clip": 1.00948572, "balance_loss_mlp": 1.00222731, "epoch": 0.4797691336499729, "flos": 70079257250400.0, "grad_norm": 0.8142666695275398, "language_loss": 0.61051852, "learning_rate": 2.22774361449746e-06, "loss": 0.63102561, "num_input_tokens_seen": 85896585, "step": 3990, "time_per_iteration": 3.9802653789520264 }, { "auxiliary_loss_clip": 0.0111697, "auxiliary_loss_mlp": 0.01026179, "balance_loss_clip": 1.04793239, "balance_loss_mlp": 1.01807892, "epoch": 0.47988937654061203, "flos": 18953332697280.0, "grad_norm": 2.580056594689465, "language_loss": 0.70357609, "learning_rate": 2.2269696924545668e-06, "loss": 0.72500759, "num_input_tokens_seen": 85914415, "step": 3991, "time_per_iteration": 3.4189059734344482 }, { "auxiliary_loss_clip": 0.01142699, "auxiliary_loss_mlp": 0.0103302, "balance_loss_clip": 1.05149877, "balance_loss_mlp": 1.02517033, "epoch": 0.48000961943125114, "flos": 14461984273440.0, "grad_norm": 2.5497188889398688, "language_loss": 0.78182375, "learning_rate": 2.2261957359805523e-06, "loss": 0.80358094, "num_input_tokens_seen": 85931650, "step": 3992, "time_per_iteration": 2.5299506187438965 }, { "auxiliary_loss_clip": 0.01185191, "auxiliary_loss_mlp": 0.01032594, "balance_loss_clip": 1.05422485, "balance_loss_mlp": 1.02452326, "epoch": 0.4801298623218902, "flos": 27051153053280.0, "grad_norm": 2.8776797826328915, "language_loss": 0.74322182, "learning_rate": 2.225421745192823e-06, "loss": 0.76539969, "num_input_tokens_seen": 85951805, "step": 3993, "time_per_iteration": 2.545273780822754 }, { "auxiliary_loss_clip": 0.01170119, "auxiliary_loss_mlp": 0.01033309, "balance_loss_clip": 1.05387616, "balance_loss_mlp": 1.02538705, "epoch": 0.4802501052125293, "flos": 26355239663520.0, "grad_norm": 1.7952599291614761, "language_loss": 0.78018486, "learning_rate": 2.2246477202087955e-06, "loss": 0.80221915, "num_input_tokens_seen": 85972485, "step": 3994, "time_per_iteration": 3.2922751903533936 }, { "auxiliary_loss_clip": 0.01154895, "auxiliary_loss_mlp": 0.01029133, "balance_loss_clip": 1.04835701, "balance_loss_mlp": 1.02116418, "epoch": 0.4803703481031684, "flos": 20993920286400.0, "grad_norm": 1.5270116710033648, "language_loss": 0.82698596, "learning_rate": 2.223873661145887e-06, "loss": 0.84882617, "num_input_tokens_seen": 85992540, "step": 3995, "time_per_iteration": 2.557039976119995 }, { "auxiliary_loss_clip": 0.0115561, "auxiliary_loss_mlp": 0.00762745, "balance_loss_clip": 1.05385935, "balance_loss_mlp": 1.00032771, "epoch": 0.4804905909938075, "flos": 20703726439200.0, "grad_norm": 1.8966314694361168, "language_loss": 0.71249628, "learning_rate": 2.2230995681215226e-06, "loss": 0.7316798, "num_input_tokens_seen": 86012065, "step": 3996, "time_per_iteration": 2.5378801822662354 }, { "auxiliary_loss_clip": 0.0113882, "auxiliary_loss_mlp": 0.01029746, "balance_loss_clip": 1.04920578, "balance_loss_mlp": 1.02158642, "epoch": 0.4806108338844466, "flos": 16654834219680.0, "grad_norm": 2.0273354822622065, "language_loss": 0.78105545, "learning_rate": 2.2223254412531305e-06, "loss": 0.80274105, "num_input_tokens_seen": 86029435, "step": 3997, "time_per_iteration": 3.344120740890503 }, { "auxiliary_loss_clip": 0.01139553, "auxiliary_loss_mlp": 0.0103428, "balance_loss_clip": 1.04466057, "balance_loss_mlp": 1.0265615, "epoch": 0.4807310767750857, "flos": 20011332911520.0, "grad_norm": 1.7202245318002838, "language_loss": 0.82232165, "learning_rate": 2.221551280658146e-06, "loss": 0.84405994, "num_input_tokens_seen": 86048495, "step": 3998, "time_per_iteration": 2.5135958194732666 }, { "auxiliary_loss_clip": 0.01123581, "auxiliary_loss_mlp": 0.01034604, "balance_loss_clip": 1.0474515, "balance_loss_mlp": 1.02662849, "epoch": 0.48085131966572475, "flos": 23185261720320.0, "grad_norm": 1.6338021878996016, "language_loss": 0.74090171, "learning_rate": 2.2207770864540085e-06, "loss": 0.7624836, "num_input_tokens_seen": 86067470, "step": 3999, "time_per_iteration": 2.614104747772217 }, { "auxiliary_loss_clip": 0.01147193, "auxiliary_loss_mlp": 0.01026832, "balance_loss_clip": 1.0492382, "balance_loss_mlp": 1.01882684, "epoch": 0.48097156255636386, "flos": 20558647474080.0, "grad_norm": 1.922174858258385, "language_loss": 0.7273342, "learning_rate": 2.220002858758162e-06, "loss": 0.74907446, "num_input_tokens_seen": 86085460, "step": 4000, "time_per_iteration": 2.531238079071045 }, { "auxiliary_loss_clip": 0.01064824, "auxiliary_loss_mlp": 0.01001186, "balance_loss_clip": 1.01124465, "balance_loss_mlp": 0.99976104, "epoch": 0.481091805447003, "flos": 70511620788960.0, "grad_norm": 1.2450105996943166, "language_loss": 0.60826075, "learning_rate": 2.2192285976880573e-06, "loss": 0.62892085, "num_input_tokens_seen": 86149715, "step": 4001, "time_per_iteration": 3.1046674251556396 }, { "auxiliary_loss_clip": 0.01144558, "auxiliary_loss_mlp": 0.00762308, "balance_loss_clip": 1.0487566, "balance_loss_mlp": 1.00035596, "epoch": 0.48121204833764203, "flos": 36428219684160.0, "grad_norm": 1.643078117958673, "language_loss": 0.8058638, "learning_rate": 2.2184543033611485e-06, "loss": 0.82493246, "num_input_tokens_seen": 86170795, "step": 4002, "time_per_iteration": 2.7389655113220215 }, { "auxiliary_loss_clip": 0.01168906, "auxiliary_loss_mlp": 0.01026954, "balance_loss_clip": 1.04918647, "balance_loss_mlp": 1.01854944, "epoch": 0.48133229122828114, "flos": 27490268980320.0, "grad_norm": 2.4117616312364247, "language_loss": 0.82040167, "learning_rate": 2.2176799758948957e-06, "loss": 0.84236026, "num_input_tokens_seen": 86190955, "step": 4003, "time_per_iteration": 2.535400390625 }, { "auxiliary_loss_clip": 0.01148367, "auxiliary_loss_mlp": 0.01032161, "balance_loss_clip": 1.04890609, "balance_loss_mlp": 1.02437675, "epoch": 0.4814525341189202, "flos": 43072808035680.0, "grad_norm": 1.660687311242782, "language_loss": 0.73580682, "learning_rate": 2.2169056154067635e-06, "loss": 0.75761211, "num_input_tokens_seen": 86214875, "step": 4004, "time_per_iteration": 2.746112823486328 }, { "auxiliary_loss_clip": 0.01169699, "auxiliary_loss_mlp": 0.00762912, "balance_loss_clip": 1.05361009, "balance_loss_mlp": 1.00030935, "epoch": 0.4815727770095593, "flos": 24236904632640.0, "grad_norm": 1.8525246466138428, "language_loss": 0.82605243, "learning_rate": 2.216131222014222e-06, "loss": 0.84537852, "num_input_tokens_seen": 86232950, "step": 4005, "time_per_iteration": 2.5238146781921387 }, { "auxiliary_loss_clip": 0.01132181, "auxiliary_loss_mlp": 0.01030216, "balance_loss_clip": 1.04594636, "balance_loss_mlp": 1.02204394, "epoch": 0.4816930199001984, "flos": 18113633352960.0, "grad_norm": 2.358057716247965, "language_loss": 0.80475521, "learning_rate": 2.2153567958347455e-06, "loss": 0.82637918, "num_input_tokens_seen": 86249160, "step": 4006, "time_per_iteration": 2.538282871246338 }, { "auxiliary_loss_clip": 0.01157244, "auxiliary_loss_mlp": 0.01033662, "balance_loss_clip": 1.05340266, "balance_loss_mlp": 1.02563643, "epoch": 0.48181326279083747, "flos": 17274724181760.0, "grad_norm": 1.803712129153883, "language_loss": 0.79611909, "learning_rate": 2.214582336985815e-06, "loss": 0.81802821, "num_input_tokens_seen": 86267060, "step": 4007, "time_per_iteration": 2.514528512954712 }, { "auxiliary_loss_clip": 0.01149437, "auxiliary_loss_mlp": 0.01027867, "balance_loss_clip": 1.05143583, "balance_loss_mlp": 1.01983809, "epoch": 0.4819335056814766, "flos": 14903255218080.0, "grad_norm": 2.3077771276271832, "language_loss": 0.66281271, "learning_rate": 2.2138078455849142e-06, "loss": 0.68458569, "num_input_tokens_seen": 86285055, "step": 4008, "time_per_iteration": 2.507171869277954 }, { "auxiliary_loss_clip": 0.01172206, "auxiliary_loss_mlp": 0.01030584, "balance_loss_clip": 1.0526638, "balance_loss_mlp": 1.02311563, "epoch": 0.4820537485721157, "flos": 19244891388960.0, "grad_norm": 2.192835654037777, "language_loss": 0.7855612, "learning_rate": 2.2130333217495334e-06, "loss": 0.80758905, "num_input_tokens_seen": 86304225, "step": 4009, "time_per_iteration": 2.5010287761688232 }, { "auxiliary_loss_clip": 0.01152427, "auxiliary_loss_mlp": 0.01028442, "balance_loss_clip": 1.05084896, "balance_loss_mlp": 1.01954341, "epoch": 0.48217399146275475, "flos": 16033794914880.0, "grad_norm": 3.061610124724525, "language_loss": 0.67962492, "learning_rate": 2.2122587655971665e-06, "loss": 0.70143354, "num_input_tokens_seen": 86319170, "step": 4010, "time_per_iteration": 2.500555992126465 }, { "auxiliary_loss_clip": 0.01154249, "auxiliary_loss_mlp": 0.01029583, "balance_loss_clip": 1.04856181, "balance_loss_mlp": 1.02219474, "epoch": 0.48229423435339386, "flos": 24134201292000.0, "grad_norm": 1.7947042733265686, "language_loss": 0.63992608, "learning_rate": 2.211484177245314e-06, "loss": 0.66176438, "num_input_tokens_seen": 86338760, "step": 4011, "time_per_iteration": 2.5788962841033936 }, { "auxiliary_loss_clip": 0.01186027, "auxiliary_loss_mlp": 0.01026806, "balance_loss_clip": 1.05448365, "balance_loss_mlp": 1.01821089, "epoch": 0.48241447724403297, "flos": 23805439018080.0, "grad_norm": 1.999149731562044, "language_loss": 0.72490907, "learning_rate": 2.21070955681148e-06, "loss": 0.74703741, "num_input_tokens_seen": 86357865, "step": 4012, "time_per_iteration": 2.5094151496887207 }, { "auxiliary_loss_clip": 0.01132538, "auxiliary_loss_mlp": 0.01034634, "balance_loss_clip": 1.04766226, "balance_loss_mlp": 1.02677536, "epoch": 0.482534720134672, "flos": 23110315801440.0, "grad_norm": 1.4946653551137086, "language_loss": 0.78094697, "learning_rate": 2.209934904413174e-06, "loss": 0.80261874, "num_input_tokens_seen": 86379470, "step": 4013, "time_per_iteration": 2.5914807319641113 }, { "auxiliary_loss_clip": 0.01108528, "auxiliary_loss_mlp": 0.01034509, "balance_loss_clip": 1.03924727, "balance_loss_mlp": 1.02639055, "epoch": 0.48265496302531113, "flos": 20923823157120.0, "grad_norm": 1.811002525532359, "language_loss": 0.71612525, "learning_rate": 2.2091602201679095e-06, "loss": 0.73755562, "num_input_tokens_seen": 86399080, "step": 4014, "time_per_iteration": 2.6815762519836426 }, { "auxiliary_loss_clip": 0.01145981, "auxiliary_loss_mlp": 0.01031367, "balance_loss_clip": 1.04818666, "balance_loss_mlp": 1.02433074, "epoch": 0.48277520591595025, "flos": 15231155484960.0, "grad_norm": 2.759022546826354, "language_loss": 0.83375883, "learning_rate": 2.208385504193206e-06, "loss": 0.85553235, "num_input_tokens_seen": 86416580, "step": 4015, "time_per_iteration": 2.552795171737671 }, { "auxiliary_loss_clip": 0.0118136, "auxiliary_loss_mlp": 0.01027914, "balance_loss_clip": 1.05079639, "balance_loss_mlp": 1.01976585, "epoch": 0.4828954488065893, "flos": 17858667655200.0, "grad_norm": 1.8854538116420685, "language_loss": 0.81214386, "learning_rate": 2.2076107566065873e-06, "loss": 0.83423662, "num_input_tokens_seen": 86434365, "step": 4016, "time_per_iteration": 3.2437613010406494 }, { "auxiliary_loss_clip": 0.0117208, "auxiliary_loss_mlp": 0.01024738, "balance_loss_clip": 1.05328906, "balance_loss_mlp": 1.01697135, "epoch": 0.4830156916972284, "flos": 32087409603360.0, "grad_norm": 2.359616749191738, "language_loss": 0.7561512, "learning_rate": 2.2068359775255816e-06, "loss": 0.77811944, "num_input_tokens_seen": 86452675, "step": 4017, "time_per_iteration": 3.3734400272369385 }, { "auxiliary_loss_clip": 0.0111891, "auxiliary_loss_mlp": 0.0102945, "balance_loss_clip": 1.044981, "balance_loss_mlp": 1.02171898, "epoch": 0.48313593458786747, "flos": 21871721136960.0, "grad_norm": 2.4104499063462046, "language_loss": 0.78913212, "learning_rate": 2.206061167067723e-06, "loss": 0.81061566, "num_input_tokens_seen": 86470785, "step": 4018, "time_per_iteration": 2.594900131225586 }, { "auxiliary_loss_clip": 0.01139622, "auxiliary_loss_mlp": 0.01027439, "balance_loss_clip": 1.04663551, "balance_loss_mlp": 1.01932704, "epoch": 0.4832561774785066, "flos": 22601210496000.0, "grad_norm": 2.001435054677529, "language_loss": 0.79515946, "learning_rate": 2.205286325350549e-06, "loss": 0.8168301, "num_input_tokens_seen": 86489850, "step": 4019, "time_per_iteration": 2.601135730743408 }, { "auxiliary_loss_clip": 0.01125402, "auxiliary_loss_mlp": 0.01034747, "balance_loss_clip": 1.04580021, "balance_loss_mlp": 1.02701318, "epoch": 0.4833764203691457, "flos": 13437344526720.0, "grad_norm": 2.2536540970138885, "language_loss": 0.72494441, "learning_rate": 2.204511452491603e-06, "loss": 0.74654591, "num_input_tokens_seen": 86506475, "step": 4020, "time_per_iteration": 3.2756805419921875 }, { "auxiliary_loss_clip": 0.01178855, "auxiliary_loss_mlp": 0.01028117, "balance_loss_clip": 1.05299187, "balance_loss_mlp": 1.02053547, "epoch": 0.48349666325978474, "flos": 44128042644000.0, "grad_norm": 3.269128116081908, "language_loss": 0.74723214, "learning_rate": 2.2037365486084316e-06, "loss": 0.76930189, "num_input_tokens_seen": 86529715, "step": 4021, "time_per_iteration": 2.663517475128174 }, { "auxiliary_loss_clip": 0.01147308, "auxiliary_loss_mlp": 0.01026755, "balance_loss_clip": 1.04643285, "balance_loss_mlp": 1.01863086, "epoch": 0.48361690615042385, "flos": 26028380988480.0, "grad_norm": 1.7751339816710718, "language_loss": 0.7800445, "learning_rate": 2.2029616138185886e-06, "loss": 0.80178511, "num_input_tokens_seen": 86548715, "step": 4022, "time_per_iteration": 2.60111141204834 }, { "auxiliary_loss_clip": 0.011372, "auxiliary_loss_mlp": 0.01030361, "balance_loss_clip": 1.05027151, "balance_loss_mlp": 1.02270472, "epoch": 0.48373714904106296, "flos": 22273310229120.0, "grad_norm": 1.6615390859076293, "language_loss": 0.82588947, "learning_rate": 2.202186648239629e-06, "loss": 0.84756511, "num_input_tokens_seen": 86568650, "step": 4023, "time_per_iteration": 3.3179240226745605 }, { "auxiliary_loss_clip": 0.01166809, "auxiliary_loss_mlp": 0.01025498, "balance_loss_clip": 1.05090523, "balance_loss_mlp": 1.01756477, "epoch": 0.483857391931702, "flos": 28292297821920.0, "grad_norm": 1.6284525089568045, "language_loss": 0.71803904, "learning_rate": 2.201411651989117e-06, "loss": 0.7399621, "num_input_tokens_seen": 86590630, "step": 4024, "time_per_iteration": 2.5545506477355957 }, { "auxiliary_loss_clip": 0.01154522, "auxiliary_loss_mlp": 0.00763013, "balance_loss_clip": 1.05212164, "balance_loss_mlp": 1.00023997, "epoch": 0.48397763482234113, "flos": 27418052750400.0, "grad_norm": 1.9389593917962245, "language_loss": 0.78678417, "learning_rate": 2.2006366251846167e-06, "loss": 0.80595946, "num_input_tokens_seen": 86611270, "step": 4025, "time_per_iteration": 2.5697245597839355 }, { "auxiliary_loss_clip": 0.01154837, "auxiliary_loss_mlp": 0.01028057, "balance_loss_clip": 1.0517838, "balance_loss_mlp": 1.02082086, "epoch": 0.48409787771298024, "flos": 16797255329760.0, "grad_norm": 2.1691153225880373, "language_loss": 0.75379163, "learning_rate": 2.1998615679436997e-06, "loss": 0.77562058, "num_input_tokens_seen": 86628810, "step": 4026, "time_per_iteration": 2.5111141204833984 }, { "auxiliary_loss_clip": 0.01160769, "auxiliary_loss_mlp": 0.01028232, "balance_loss_clip": 1.04870629, "balance_loss_mlp": 1.02043569, "epoch": 0.4842181206036193, "flos": 25083500033280.0, "grad_norm": 2.1362902292749273, "language_loss": 0.76872003, "learning_rate": 2.199086480383942e-06, "loss": 0.79061007, "num_input_tokens_seen": 86648185, "step": 4027, "time_per_iteration": 2.5471670627593994 }, { "auxiliary_loss_clip": 0.01166464, "auxiliary_loss_mlp": 0.0103017, "balance_loss_clip": 1.05059183, "balance_loss_mlp": 1.02124739, "epoch": 0.4843383634942584, "flos": 30372315844800.0, "grad_norm": 2.7673022708727206, "language_loss": 0.67705882, "learning_rate": 2.1983113626229234e-06, "loss": 0.69902509, "num_input_tokens_seen": 86667435, "step": 4028, "time_per_iteration": 2.6212661266326904 }, { "auxiliary_loss_clip": 0.01132135, "auxiliary_loss_mlp": 0.00762696, "balance_loss_clip": 1.04340959, "balance_loss_mlp": 1.00032377, "epoch": 0.4844586063848975, "flos": 20413568508960.0, "grad_norm": 1.6338292309171, "language_loss": 0.78764665, "learning_rate": 2.1975362147782293e-06, "loss": 0.80659491, "num_input_tokens_seen": 86686630, "step": 4029, "time_per_iteration": 2.5725903511047363 }, { "auxiliary_loss_clip": 0.01053477, "auxiliary_loss_mlp": 0.01001493, "balance_loss_clip": 1.0175643, "balance_loss_mlp": 1.00012231, "epoch": 0.48457884927553657, "flos": 70303735837440.0, "grad_norm": 0.6972109499362049, "language_loss": 0.54188502, "learning_rate": 2.196761036967448e-06, "loss": 0.56243473, "num_input_tokens_seen": 86754595, "step": 4030, "time_per_iteration": 3.275477409362793 }, { "auxiliary_loss_clip": 0.01161191, "auxiliary_loss_mlp": 0.01031015, "balance_loss_clip": 1.04884243, "balance_loss_mlp": 1.02386558, "epoch": 0.4846990921661757, "flos": 19934519310720.0, "grad_norm": 1.596139339489997, "language_loss": 0.77463233, "learning_rate": 2.1959858293081743e-06, "loss": 0.79655439, "num_input_tokens_seen": 86773730, "step": 4031, "time_per_iteration": 2.4968342781066895 }, { "auxiliary_loss_clip": 0.01139729, "auxiliary_loss_mlp": 0.01023782, "balance_loss_clip": 1.04978514, "balance_loss_mlp": 1.01581287, "epoch": 0.4848193350568148, "flos": 23075949659040.0, "grad_norm": 1.6663334828293446, "language_loss": 0.75897014, "learning_rate": 2.1952105919180056e-06, "loss": 0.7806052, "num_input_tokens_seen": 86792985, "step": 4032, "time_per_iteration": 2.599210500717163 }, { "auxiliary_loss_clip": 0.01152616, "auxiliary_loss_mlp": 0.0102499, "balance_loss_clip": 1.05082393, "balance_loss_mlp": 1.01743841, "epoch": 0.48493957794745385, "flos": 22455485025600.0, "grad_norm": 2.3626867996707306, "language_loss": 0.67745745, "learning_rate": 2.1944353249145456e-06, "loss": 0.69923353, "num_input_tokens_seen": 86812095, "step": 4033, "time_per_iteration": 2.5231547355651855 }, { "auxiliary_loss_clip": 0.01184015, "auxiliary_loss_mlp": 0.01026914, "balance_loss_clip": 1.05534887, "balance_loss_mlp": 1.01893282, "epoch": 0.48505982083809296, "flos": 25046116866240.0, "grad_norm": 1.5129864561927904, "language_loss": 0.74142218, "learning_rate": 2.193660028415401e-06, "loss": 0.76353145, "num_input_tokens_seen": 86832875, "step": 4034, "time_per_iteration": 2.520139217376709 }, { "auxiliary_loss_clip": 0.0114691, "auxiliary_loss_mlp": 0.01026243, "balance_loss_clip": 1.04961431, "balance_loss_mlp": 1.01827931, "epoch": 0.485180063728732, "flos": 26761390209600.0, "grad_norm": 1.7638105553833487, "language_loss": 0.82019621, "learning_rate": 2.1928847025381852e-06, "loss": 0.84192777, "num_input_tokens_seen": 86853480, "step": 4035, "time_per_iteration": 2.581977605819702 }, { "auxiliary_loss_clip": 0.01162962, "auxiliary_loss_mlp": 0.01027418, "balance_loss_clip": 1.0459497, "balance_loss_mlp": 1.01963997, "epoch": 0.4853003066193711, "flos": 24059147622240.0, "grad_norm": 1.6738333441837436, "language_loss": 0.83797091, "learning_rate": 2.192109347400512e-06, "loss": 0.85987478, "num_input_tokens_seen": 86873695, "step": 4036, "time_per_iteration": 2.5349905490875244 }, { "auxiliary_loss_clip": 0.01155585, "auxiliary_loss_mlp": 0.01026706, "balance_loss_clip": 1.04976797, "balance_loss_mlp": 1.01885617, "epoch": 0.48542054951001024, "flos": 23076380662560.0, "grad_norm": 2.0755424493187453, "language_loss": 0.79147315, "learning_rate": 2.191333963120004e-06, "loss": 0.81329602, "num_input_tokens_seen": 86892675, "step": 4037, "time_per_iteration": 2.5662460327148438 }, { "auxiliary_loss_clip": 0.01153576, "auxiliary_loss_mlp": 0.01030027, "balance_loss_clip": 1.04979324, "balance_loss_mlp": 1.02206993, "epoch": 0.4855407924006493, "flos": 25664893402560.0, "grad_norm": 2.7818056600399412, "language_loss": 0.70061731, "learning_rate": 2.190558549814286e-06, "loss": 0.72245342, "num_input_tokens_seen": 86912835, "step": 4038, "time_per_iteration": 2.565373420715332 }, { "auxiliary_loss_clip": 0.01151402, "auxiliary_loss_mlp": 0.01033757, "balance_loss_clip": 1.04781508, "balance_loss_mlp": 1.02661014, "epoch": 0.4856610352912884, "flos": 23987937067200.0, "grad_norm": 1.669828689554297, "language_loss": 0.79674244, "learning_rate": 2.1897831076009872e-06, "loss": 0.81859404, "num_input_tokens_seen": 86932475, "step": 4039, "time_per_iteration": 2.5508687496185303 }, { "auxiliary_loss_clip": 0.01166204, "auxiliary_loss_mlp": 0.01028399, "balance_loss_clip": 1.04951119, "balance_loss_mlp": 1.02079368, "epoch": 0.4857812781819275, "flos": 24096818124960.0, "grad_norm": 1.6913957001498652, "language_loss": 0.79638577, "learning_rate": 2.1890076365977426e-06, "loss": 0.81833184, "num_input_tokens_seen": 86952300, "step": 4040, "time_per_iteration": 2.5306570529937744 }, { "auxiliary_loss_clip": 0.01049437, "auxiliary_loss_mlp": 0.0100157, "balance_loss_clip": 1.01087856, "balance_loss_mlp": 1.00027108, "epoch": 0.48590152107256657, "flos": 56266644374880.0, "grad_norm": 1.418840216954893, "language_loss": 0.52817631, "learning_rate": 2.188232136922189e-06, "loss": 0.54868639, "num_input_tokens_seen": 87010420, "step": 4041, "time_per_iteration": 3.0511622428894043 }, { "auxiliary_loss_clip": 0.01103119, "auxiliary_loss_mlp": 0.01029669, "balance_loss_clip": 1.043661, "balance_loss_mlp": 1.02231073, "epoch": 0.4860217639632057, "flos": 20046992064480.0, "grad_norm": 1.9186856470699838, "language_loss": 0.7612381, "learning_rate": 2.187456608691971e-06, "loss": 0.78256595, "num_input_tokens_seen": 87029295, "step": 4042, "time_per_iteration": 3.3853471279144287 }, { "auxiliary_loss_clip": 0.01146349, "auxiliary_loss_mlp": 0.01022421, "balance_loss_clip": 1.05220485, "balance_loss_mlp": 1.01474428, "epoch": 0.4861420068538448, "flos": 17822146495200.0, "grad_norm": 1.9661596351786124, "language_loss": 0.87582791, "learning_rate": 2.1866810520247334e-06, "loss": 0.89751565, "num_input_tokens_seen": 87048165, "step": 4043, "time_per_iteration": 3.3308095932006836 }, { "auxiliary_loss_clip": 0.01170529, "auxiliary_loss_mlp": 0.01027024, "balance_loss_clip": 1.04868662, "balance_loss_mlp": 1.01913881, "epoch": 0.48626224974448384, "flos": 26250129886560.0, "grad_norm": 1.9957045899261117, "language_loss": 0.65133679, "learning_rate": 2.185905467038129e-06, "loss": 0.67331237, "num_input_tokens_seen": 87067070, "step": 4044, "time_per_iteration": 2.5555038452148438 }, { "auxiliary_loss_clip": 0.01182289, "auxiliary_loss_mlp": 0.01028686, "balance_loss_clip": 1.05467737, "balance_loss_mlp": 1.02097869, "epoch": 0.48638249263512295, "flos": 22054506521760.0, "grad_norm": 1.7032606153179595, "language_loss": 0.777107, "learning_rate": 2.1851298538498127e-06, "loss": 0.79921675, "num_input_tokens_seen": 87086785, "step": 4045, "time_per_iteration": 2.51045298576355 }, { "auxiliary_loss_clip": 0.01177064, "auxiliary_loss_mlp": 0.00763389, "balance_loss_clip": 1.05458522, "balance_loss_mlp": 1.00047922, "epoch": 0.48650273552576206, "flos": 25119949359360.0, "grad_norm": 1.9456105625291318, "language_loss": 0.79861426, "learning_rate": 2.184354212577446e-06, "loss": 0.81801879, "num_input_tokens_seen": 87107090, "step": 4046, "time_per_iteration": 3.237257242202759 }, { "auxiliary_loss_clip": 0.01186905, "auxiliary_loss_mlp": 0.0103109, "balance_loss_clip": 1.05360031, "balance_loss_mlp": 1.02301335, "epoch": 0.4866229784164011, "flos": 17456934895200.0, "grad_norm": 3.109898640354858, "language_loss": 0.6281113, "learning_rate": 2.1835785433386907e-06, "loss": 0.65029126, "num_input_tokens_seen": 87125905, "step": 4047, "time_per_iteration": 2.443817138671875 }, { "auxiliary_loss_clip": 0.01131619, "auxiliary_loss_mlp": 0.01031734, "balance_loss_clip": 1.04875875, "balance_loss_mlp": 1.02392554, "epoch": 0.48674322130704023, "flos": 23331130858560.0, "grad_norm": 1.831952475247587, "language_loss": 0.65344715, "learning_rate": 2.182802846251216e-06, "loss": 0.67508066, "num_input_tokens_seen": 87146175, "step": 4048, "time_per_iteration": 2.6026225090026855 }, { "auxiliary_loss_clip": 0.01142028, "auxiliary_loss_mlp": 0.0102377, "balance_loss_clip": 1.04510403, "balance_loss_mlp": 1.01628399, "epoch": 0.4868634641976793, "flos": 28804348318080.0, "grad_norm": 1.8029736321154628, "language_loss": 0.72234946, "learning_rate": 2.182027121432696e-06, "loss": 0.74400747, "num_input_tokens_seen": 87166800, "step": 4049, "time_per_iteration": 3.365053415298462 }, { "auxiliary_loss_clip": 0.01183316, "auxiliary_loss_mlp": 0.01027661, "balance_loss_clip": 1.05178952, "balance_loss_mlp": 1.01907218, "epoch": 0.4869837070883184, "flos": 19025979930720.0, "grad_norm": 2.1094227782688257, "language_loss": 0.81976461, "learning_rate": 2.1812513690008054e-06, "loss": 0.84187436, "num_input_tokens_seen": 87185920, "step": 4050, "time_per_iteration": 2.483416795730591 }, { "auxiliary_loss_clip": 0.01175115, "auxiliary_loss_mlp": 0.01033207, "balance_loss_clip": 1.05108726, "balance_loss_mlp": 1.02528572, "epoch": 0.4871039499789575, "flos": 15121412420160.0, "grad_norm": 2.0665144982851937, "language_loss": 0.7956239, "learning_rate": 2.180475589073227e-06, "loss": 0.81770712, "num_input_tokens_seen": 87203620, "step": 4051, "time_per_iteration": 2.480064630508423 }, { "auxiliary_loss_clip": 0.01156913, "auxiliary_loss_mlp": 0.01025964, "balance_loss_clip": 1.04745793, "balance_loss_mlp": 1.01868069, "epoch": 0.48722419286959656, "flos": 26174070541920.0, "grad_norm": 1.6269666395023596, "language_loss": 0.73596209, "learning_rate": 2.1796997817676456e-06, "loss": 0.75779086, "num_input_tokens_seen": 87224630, "step": 4052, "time_per_iteration": 2.5426878929138184 }, { "auxiliary_loss_clip": 0.01170893, "auxiliary_loss_mlp": 0.0076216, "balance_loss_clip": 1.05226779, "balance_loss_mlp": 1.00032973, "epoch": 0.4873444357602357, "flos": 24026146324320.0, "grad_norm": 1.6044887403551225, "language_loss": 0.67582077, "learning_rate": 2.1789239472017494e-06, "loss": 0.69515127, "num_input_tokens_seen": 87246280, "step": 4053, "time_per_iteration": 2.542314052581787 }, { "auxiliary_loss_clip": 0.01137799, "auxiliary_loss_mlp": 0.01034345, "balance_loss_clip": 1.04582691, "balance_loss_mlp": 1.02607763, "epoch": 0.4874646786508748, "flos": 22820445206880.0, "grad_norm": 2.2169342187445267, "language_loss": 0.73020971, "learning_rate": 2.1781480854932326e-06, "loss": 0.75193107, "num_input_tokens_seen": 87266045, "step": 4054, "time_per_iteration": 2.582216739654541 }, { "auxiliary_loss_clip": 0.01123008, "auxiliary_loss_mlp": 0.01028625, "balance_loss_clip": 1.0478512, "balance_loss_mlp": 1.02104354, "epoch": 0.48758492154151384, "flos": 21287598078720.0, "grad_norm": 5.160387939338385, "language_loss": 0.79066932, "learning_rate": 2.1773721967597933e-06, "loss": 0.8121857, "num_input_tokens_seen": 87284495, "step": 4055, "time_per_iteration": 2.5968892574310303 }, { "auxiliary_loss_clip": 0.01045964, "auxiliary_loss_mlp": 0.01001046, "balance_loss_clip": 1.0114553, "balance_loss_mlp": 0.99980658, "epoch": 0.48770516443215295, "flos": 62244118350240.0, "grad_norm": 0.8418703401613191, "language_loss": 0.57419842, "learning_rate": 2.1765962811191322e-06, "loss": 0.59466857, "num_input_tokens_seen": 87338960, "step": 4056, "time_per_iteration": 3.097783327102661 }, { "auxiliary_loss_clip": 0.01025536, "auxiliary_loss_mlp": 0.01003453, "balance_loss_clip": 1.01061058, "balance_loss_mlp": 1.00220716, "epoch": 0.48782540732279206, "flos": 66133463458080.0, "grad_norm": 0.8321433779936758, "language_loss": 0.62103987, "learning_rate": 2.1758203386889566e-06, "loss": 0.64132977, "num_input_tokens_seen": 87401730, "step": 4057, "time_per_iteration": 3.2168476581573486 }, { "auxiliary_loss_clip": 0.0113891, "auxiliary_loss_mlp": 0.0076329, "balance_loss_clip": 1.04748595, "balance_loss_mlp": 1.00036657, "epoch": 0.4879456502134311, "flos": 14607925245600.0, "grad_norm": 1.829988610535916, "language_loss": 0.839562, "learning_rate": 2.1750443695869746e-06, "loss": 0.85858405, "num_input_tokens_seen": 87417300, "step": 4058, "time_per_iteration": 2.538055896759033 }, { "auxiliary_loss_clip": 0.01171384, "auxiliary_loss_mlp": 0.0102473, "balance_loss_clip": 1.0511837, "balance_loss_mlp": 1.01725531, "epoch": 0.4880658931040702, "flos": 19500467675040.0, "grad_norm": 1.8065712776786784, "language_loss": 0.85676408, "learning_rate": 2.174268373930901e-06, "loss": 0.87872523, "num_input_tokens_seen": 87434815, "step": 4059, "time_per_iteration": 2.509207010269165 }, { "auxiliary_loss_clip": 0.01132994, "auxiliary_loss_mlp": 0.00762444, "balance_loss_clip": 1.04838228, "balance_loss_mlp": 1.00041151, "epoch": 0.48818613599470934, "flos": 16723063667040.0, "grad_norm": 1.982404819974681, "language_loss": 0.79956162, "learning_rate": 2.1734923518384537e-06, "loss": 0.81851602, "num_input_tokens_seen": 87451420, "step": 4060, "time_per_iteration": 2.5430855751037598 }, { "auxiliary_loss_clip": 0.0112483, "auxiliary_loss_mlp": 0.01029201, "balance_loss_clip": 1.04643369, "balance_loss_mlp": 1.02188134, "epoch": 0.4883063788853484, "flos": 26756936506560.0, "grad_norm": 1.8184496701062343, "language_loss": 0.82200134, "learning_rate": 2.1727163034273547e-06, "loss": 0.84354162, "num_input_tokens_seen": 87469585, "step": 4061, "time_per_iteration": 2.624725580215454 }, { "auxiliary_loss_clip": 0.0116882, "auxiliary_loss_mlp": 0.01026525, "balance_loss_clip": 1.04969788, "balance_loss_mlp": 1.01889825, "epoch": 0.4884266217759875, "flos": 16763392024800.0, "grad_norm": 2.4766510061628044, "language_loss": 0.78949618, "learning_rate": 2.17194022881533e-06, "loss": 0.81144965, "num_input_tokens_seen": 87485675, "step": 4062, "time_per_iteration": 2.4789772033691406 }, { "auxiliary_loss_clip": 0.01158051, "auxiliary_loss_mlp": 0.01029152, "balance_loss_clip": 1.04923892, "balance_loss_mlp": 1.02064013, "epoch": 0.4885468646666266, "flos": 24207135861120.0, "grad_norm": 1.6429348755366397, "language_loss": 0.6779632, "learning_rate": 2.1711641281201092e-06, "loss": 0.69983518, "num_input_tokens_seen": 87505605, "step": 4063, "time_per_iteration": 2.5798404216766357 }, { "auxiliary_loss_clip": 0.01167923, "auxiliary_loss_mlp": 0.0102522, "balance_loss_clip": 1.05248725, "balance_loss_mlp": 1.0181179, "epoch": 0.48866710755726567, "flos": 14610798602400.0, "grad_norm": 2.4491180672450064, "language_loss": 0.79298067, "learning_rate": 2.1703880014594264e-06, "loss": 0.81491202, "num_input_tokens_seen": 87523195, "step": 4064, "time_per_iteration": 2.486459970474243 }, { "auxiliary_loss_clip": 0.01118458, "auxiliary_loss_mlp": 0.01031364, "balance_loss_clip": 1.04697895, "balance_loss_mlp": 1.02379107, "epoch": 0.4887873504479048, "flos": 28804456068960.0, "grad_norm": 1.958098164360088, "language_loss": 0.73341048, "learning_rate": 2.1696118489510182e-06, "loss": 0.75490868, "num_input_tokens_seen": 87544125, "step": 4065, "time_per_iteration": 2.6454732418060303 }, { "auxiliary_loss_clip": 0.0114782, "auxiliary_loss_mlp": 0.00762419, "balance_loss_clip": 1.05062318, "balance_loss_mlp": 1.00031114, "epoch": 0.48890759333854383, "flos": 22784390967360.0, "grad_norm": 1.7601057124651436, "language_loss": 0.72625363, "learning_rate": 2.1688356707126286e-06, "loss": 0.74535602, "num_input_tokens_seen": 87563745, "step": 4066, "time_per_iteration": 2.5789740085601807 }, { "auxiliary_loss_clip": 0.01135876, "auxiliary_loss_mlp": 0.01029101, "balance_loss_clip": 1.04887724, "balance_loss_mlp": 1.0210191, "epoch": 0.48902783622918294, "flos": 17786092255680.0, "grad_norm": 1.8003184832558599, "language_loss": 0.69971538, "learning_rate": 2.168059466862001e-06, "loss": 0.72136515, "num_input_tokens_seen": 87581895, "step": 4067, "time_per_iteration": 2.5490429401397705 }, { "auxiliary_loss_clip": 0.01153461, "auxiliary_loss_mlp": 0.01028205, "balance_loss_clip": 1.04621601, "balance_loss_mlp": 1.02042627, "epoch": 0.48914807911982205, "flos": 22310298309600.0, "grad_norm": 1.911533166027528, "language_loss": 0.81429195, "learning_rate": 2.167283237516887e-06, "loss": 0.83610857, "num_input_tokens_seen": 87600170, "step": 4068, "time_per_iteration": 3.327479839324951 }, { "auxiliary_loss_clip": 0.0115731, "auxiliary_loss_mlp": 0.01023509, "balance_loss_clip": 1.04961634, "balance_loss_mlp": 1.0155282, "epoch": 0.4892683220104611, "flos": 16363024109280.0, "grad_norm": 1.7938742710683386, "language_loss": 0.74765694, "learning_rate": 2.1665069827950383e-06, "loss": 0.76946521, "num_input_tokens_seen": 87617455, "step": 4069, "time_per_iteration": 3.23183012008667 }, { "auxiliary_loss_clip": 0.01151232, "auxiliary_loss_mlp": 0.01027437, "balance_loss_clip": 1.04874265, "balance_loss_mlp": 1.02007914, "epoch": 0.4893885649011002, "flos": 15739147364640.0, "grad_norm": 1.7788582265581572, "language_loss": 0.8650924, "learning_rate": 2.1657307028142126e-06, "loss": 0.88687909, "num_input_tokens_seen": 87634995, "step": 4070, "time_per_iteration": 2.511720657348633 }, { "auxiliary_loss_clip": 0.01155969, "auxiliary_loss_mlp": 0.01027554, "balance_loss_clip": 1.05145788, "balance_loss_mlp": 1.01923943, "epoch": 0.48950880779173933, "flos": 28581988831680.0, "grad_norm": 1.9453501391356156, "language_loss": 0.67357194, "learning_rate": 2.164954397692171e-06, "loss": 0.69540727, "num_input_tokens_seen": 87654420, "step": 4071, "time_per_iteration": 2.604642391204834 }, { "auxiliary_loss_clip": 0.01058968, "auxiliary_loss_mlp": 0.01003435, "balance_loss_clip": 1.01478386, "balance_loss_mlp": 1.00223148, "epoch": 0.4896290506823784, "flos": 66186320446560.0, "grad_norm": 1.0867915048663765, "language_loss": 0.77342784, "learning_rate": 2.164178067546678e-06, "loss": 0.79405189, "num_input_tokens_seen": 87713585, "step": 4072, "time_per_iteration": 3.923360586166382 }, { "auxiliary_loss_clip": 0.01157086, "auxiliary_loss_mlp": 0.01032897, "balance_loss_clip": 1.04605901, "balance_loss_mlp": 1.02566385, "epoch": 0.4897492935730175, "flos": 12531067915200.0, "grad_norm": 1.8491229126865965, "language_loss": 0.90676618, "learning_rate": 2.163401712495504e-06, "loss": 0.92866606, "num_input_tokens_seen": 87731280, "step": 4073, "time_per_iteration": 2.5898241996765137 }, { "auxiliary_loss_clip": 0.01127581, "auxiliary_loss_mlp": 0.0102752, "balance_loss_clip": 1.046422, "balance_loss_mlp": 1.01949763, "epoch": 0.4898695364636566, "flos": 23476820412000.0, "grad_norm": 1.5278761419482159, "language_loss": 0.79105675, "learning_rate": 2.1626253326564194e-06, "loss": 0.81260777, "num_input_tokens_seen": 87750230, "step": 4074, "time_per_iteration": 2.683238983154297 }, { "auxiliary_loss_clip": 0.0115346, "auxiliary_loss_mlp": 0.01030078, "balance_loss_clip": 1.04806721, "balance_loss_mlp": 1.02193594, "epoch": 0.48998977935429566, "flos": 27160213695840.0, "grad_norm": 2.0104066865304304, "language_loss": 0.76990312, "learning_rate": 2.161848928147201e-06, "loss": 0.79173851, "num_input_tokens_seen": 87770500, "step": 4075, "time_per_iteration": 3.28778076171875 }, { "auxiliary_loss_clip": 0.01170633, "auxiliary_loss_mlp": 0.01028783, "balance_loss_clip": 1.05290627, "balance_loss_mlp": 1.02114201, "epoch": 0.4901100222449348, "flos": 20339592348000.0, "grad_norm": 1.9870254039373014, "language_loss": 0.80963528, "learning_rate": 2.161072499085629e-06, "loss": 0.8316294, "num_input_tokens_seen": 87789495, "step": 4076, "time_per_iteration": 2.498622179031372 }, { "auxiliary_loss_clip": 0.01140732, "auxiliary_loss_mlp": 0.01034871, "balance_loss_clip": 1.04715335, "balance_loss_mlp": 1.02758729, "epoch": 0.4902302651355739, "flos": 30446363839680.0, "grad_norm": 1.7820222435168724, "language_loss": 0.83191442, "learning_rate": 2.160296045589487e-06, "loss": 0.85367048, "num_input_tokens_seen": 87812955, "step": 4077, "time_per_iteration": 2.653218984603882 }, { "auxiliary_loss_clip": 0.01167039, "auxiliary_loss_mlp": 0.01032645, "balance_loss_clip": 1.05192578, "balance_loss_mlp": 1.02484918, "epoch": 0.49035050802621294, "flos": 19174183671360.0, "grad_norm": 1.8832850189517978, "language_loss": 0.69953799, "learning_rate": 2.159519567776562e-06, "loss": 0.72153485, "num_input_tokens_seen": 87832605, "step": 4078, "time_per_iteration": 2.512356758117676 }, { "auxiliary_loss_clip": 0.01127603, "auxiliary_loss_mlp": 0.01026302, "balance_loss_clip": 1.04404783, "balance_loss_mlp": 1.01855302, "epoch": 0.49047075091685205, "flos": 22228492251360.0, "grad_norm": 2.7958018768804496, "language_loss": 0.70564032, "learning_rate": 2.1587430657646463e-06, "loss": 0.72717941, "num_input_tokens_seen": 87846040, "step": 4079, "time_per_iteration": 2.5713553428649902 }, { "auxiliary_loss_clip": 0.01154022, "auxiliary_loss_mlp": 0.01029553, "balance_loss_clip": 1.05112743, "balance_loss_mlp": 1.02216756, "epoch": 0.4905909938074911, "flos": 20156519627520.0, "grad_norm": 1.8628066069411002, "language_loss": 0.77894425, "learning_rate": 2.157966539671533e-06, "loss": 0.80077994, "num_input_tokens_seen": 87865680, "step": 4080, "time_per_iteration": 2.5318288803100586 }, { "auxiliary_loss_clip": 0.01144261, "auxiliary_loss_mlp": 0.01025696, "balance_loss_clip": 1.04835653, "balance_loss_mlp": 1.01852608, "epoch": 0.4907112366981302, "flos": 17202220616160.0, "grad_norm": 13.868131110637385, "language_loss": 0.67629313, "learning_rate": 2.157189989615021e-06, "loss": 0.69799268, "num_input_tokens_seen": 87884270, "step": 4081, "time_per_iteration": 2.571187734603882 }, { "auxiliary_loss_clip": 0.01170524, "auxiliary_loss_mlp": 0.00763638, "balance_loss_clip": 1.05039334, "balance_loss_mlp": 1.00049734, "epoch": 0.4908314795887693, "flos": 21688971669120.0, "grad_norm": 2.07262666377095, "language_loss": 0.75170565, "learning_rate": 2.156413415712913e-06, "loss": 0.77104723, "num_input_tokens_seen": 87906320, "step": 4082, "time_per_iteration": 2.552931070327759 }, { "auxiliary_loss_clip": 0.01157354, "auxiliary_loss_mlp": 0.00763131, "balance_loss_clip": 1.04829037, "balance_loss_mlp": 1.000453, "epoch": 0.4909517224794084, "flos": 26213680560480.0, "grad_norm": 1.638972190313164, "language_loss": 0.78292739, "learning_rate": 2.155636818083014e-06, "loss": 0.80213225, "num_input_tokens_seen": 87927690, "step": 4083, "time_per_iteration": 2.587545871734619 }, { "auxiliary_loss_clip": 0.01149986, "auxiliary_loss_mlp": 0.01028556, "balance_loss_clip": 1.04969192, "balance_loss_mlp": 1.02106416, "epoch": 0.4910719653700475, "flos": 23148381390720.0, "grad_norm": 1.9368991328995953, "language_loss": 0.84278417, "learning_rate": 2.154860196843134e-06, "loss": 0.8645696, "num_input_tokens_seen": 87946885, "step": 4084, "time_per_iteration": 2.5438051223754883 }, { "auxiliary_loss_clip": 0.01182581, "auxiliary_loss_mlp": 0.01029707, "balance_loss_clip": 1.05166268, "balance_loss_mlp": 1.02191055, "epoch": 0.4911922082606866, "flos": 23331849197760.0, "grad_norm": 1.628417914080019, "language_loss": 0.76803219, "learning_rate": 2.154083552111085e-06, "loss": 0.79015505, "num_input_tokens_seen": 87966055, "step": 4085, "time_per_iteration": 2.5283236503601074 }, { "auxiliary_loss_clip": 0.01182529, "auxiliary_loss_mlp": 0.01031393, "balance_loss_clip": 1.0503757, "balance_loss_mlp": 1.0235486, "epoch": 0.49131245115132566, "flos": 29203243638240.0, "grad_norm": 2.0855726825397274, "language_loss": 0.81994402, "learning_rate": 2.1533068840046834e-06, "loss": 0.84208322, "num_input_tokens_seen": 87986320, "step": 4086, "time_per_iteration": 2.523937702178955 }, { "auxiliary_loss_clip": 0.01148451, "auxiliary_loss_mlp": 0.00763171, "balance_loss_clip": 1.04873013, "balance_loss_mlp": 1.00038302, "epoch": 0.49143269404196477, "flos": 20147468553600.0, "grad_norm": 2.4379275856742337, "language_loss": 0.61599672, "learning_rate": 2.152530192641749e-06, "loss": 0.63511288, "num_input_tokens_seen": 88001230, "step": 4087, "time_per_iteration": 2.530839204788208 }, { "auxiliary_loss_clip": 0.01170883, "auxiliary_loss_mlp": 0.01028443, "balance_loss_clip": 1.05233204, "balance_loss_mlp": 1.02116489, "epoch": 0.4915529369326039, "flos": 24389813495040.0, "grad_norm": 1.8613653265455574, "language_loss": 0.72505683, "learning_rate": 2.1517534781401068e-06, "loss": 0.74705011, "num_input_tokens_seen": 88019110, "step": 4088, "time_per_iteration": 2.532093048095703 }, { "auxiliary_loss_clip": 0.01166985, "auxiliary_loss_mlp": 0.01028486, "balance_loss_clip": 1.05025482, "balance_loss_mlp": 1.02052903, "epoch": 0.49167317982324293, "flos": 10524307714080.0, "grad_norm": 2.2264680670944386, "language_loss": 0.6975978, "learning_rate": 2.150976740617581e-06, "loss": 0.71955246, "num_input_tokens_seen": 88035670, "step": 4089, "time_per_iteration": 2.4816243648529053 }, { "auxiliary_loss_clip": 0.01157533, "auxiliary_loss_mlp": 0.0102723, "balance_loss_clip": 1.05053449, "balance_loss_mlp": 1.01962113, "epoch": 0.49179342271388204, "flos": 25593431428800.0, "grad_norm": 1.8375963619383147, "language_loss": 0.71615565, "learning_rate": 2.150199980192006e-06, "loss": 0.73800325, "num_input_tokens_seen": 88054790, "step": 4090, "time_per_iteration": 2.5625691413879395 }, { "auxiliary_loss_clip": 0.01148949, "auxiliary_loss_mlp": 0.01022416, "balance_loss_clip": 1.04935026, "balance_loss_mlp": 1.01540625, "epoch": 0.49191366560452116, "flos": 21102047088000.0, "grad_norm": 1.5976330467530417, "language_loss": 0.80838126, "learning_rate": 2.1494231969812114e-06, "loss": 0.83009487, "num_input_tokens_seen": 88073780, "step": 4091, "time_per_iteration": 2.5294651985168457 }, { "auxiliary_loss_clip": 0.01146272, "auxiliary_loss_mlp": 0.01034361, "balance_loss_clip": 1.04985976, "balance_loss_mlp": 1.02719378, "epoch": 0.4920339084951602, "flos": 26067524086560.0, "grad_norm": 2.543475040092491, "language_loss": 0.81099755, "learning_rate": 2.1486463911030372e-06, "loss": 0.83280385, "num_input_tokens_seen": 88094430, "step": 4092, "time_per_iteration": 2.6041460037231445 }, { "auxiliary_loss_clip": 0.01152775, "auxiliary_loss_mlp": 0.01031546, "balance_loss_clip": 1.04812884, "balance_loss_mlp": 1.02353525, "epoch": 0.4921541513857993, "flos": 25081273181760.0, "grad_norm": 1.6611522743090883, "language_loss": 0.74437034, "learning_rate": 2.147869562675324e-06, "loss": 0.76621354, "num_input_tokens_seen": 88113400, "step": 4093, "time_per_iteration": 2.5888993740081787 }, { "auxiliary_loss_clip": 0.0117, "auxiliary_loss_mlp": 0.01027383, "balance_loss_clip": 1.05345786, "balance_loss_mlp": 1.01981342, "epoch": 0.49227439427643843, "flos": 24389813495040.0, "grad_norm": 1.7564698631388485, "language_loss": 0.72519171, "learning_rate": 2.147092711815915e-06, "loss": 0.74716556, "num_input_tokens_seen": 88132750, "step": 4094, "time_per_iteration": 2.5859196186065674 }, { "auxiliary_loss_clip": 0.01138261, "auxiliary_loss_mlp": 0.010283, "balance_loss_clip": 1.04982364, "balance_loss_mlp": 1.02058995, "epoch": 0.4923946371670775, "flos": 11363755639680.0, "grad_norm": 2.467314663038252, "language_loss": 0.86561185, "learning_rate": 2.1463158386426593e-06, "loss": 0.88727748, "num_input_tokens_seen": 88150560, "step": 4095, "time_per_iteration": 3.9827609062194824 }, { "auxiliary_loss_clip": 0.01158978, "auxiliary_loss_mlp": 0.01030344, "balance_loss_clip": 1.051229, "balance_loss_mlp": 1.02145076, "epoch": 0.4925148800577166, "flos": 30445968753120.0, "grad_norm": 3.334355104390387, "language_loss": 0.76941562, "learning_rate": 2.145538943273407e-06, "loss": 0.79130888, "num_input_tokens_seen": 88170835, "step": 4096, "time_per_iteration": 2.603121042251587 }, { "auxiliary_loss_clip": 0.01184309, "auxiliary_loss_mlp": 0.01030921, "balance_loss_clip": 1.0535562, "balance_loss_mlp": 1.02330923, "epoch": 0.49263512294835565, "flos": 20850457584480.0, "grad_norm": 1.728353253839469, "language_loss": 0.71793258, "learning_rate": 2.144762025826013e-06, "loss": 0.74008489, "num_input_tokens_seen": 88189925, "step": 4097, "time_per_iteration": 2.514362096786499 }, { "auxiliary_loss_clip": 0.01172261, "auxiliary_loss_mlp": 0.01029541, "balance_loss_clip": 1.05052257, "balance_loss_mlp": 1.021667, "epoch": 0.49275536583899476, "flos": 23767481179680.0, "grad_norm": 2.265454115126669, "language_loss": 0.86894894, "learning_rate": 2.143985086418334e-06, "loss": 0.89096695, "num_input_tokens_seen": 88205105, "step": 4098, "time_per_iteration": 3.2738871574401855 }, { "auxiliary_loss_clip": 0.01155919, "auxiliary_loss_mlp": 0.01024351, "balance_loss_clip": 1.05010247, "balance_loss_mlp": 1.01704621, "epoch": 0.4928756087296339, "flos": 22273525730880.0, "grad_norm": 1.3247930993844221, "language_loss": 0.76472163, "learning_rate": 2.1432081251682324e-06, "loss": 0.7865243, "num_input_tokens_seen": 88225475, "step": 4099, "time_per_iteration": 2.5376405715942383 }, { "auxiliary_loss_clip": 0.01169594, "auxiliary_loss_mlp": 0.01028939, "balance_loss_clip": 1.05511892, "balance_loss_mlp": 1.02126765, "epoch": 0.49299585162027293, "flos": 19645474806240.0, "grad_norm": 1.6537840945373592, "language_loss": 0.8695333, "learning_rate": 2.142431142193572e-06, "loss": 0.89151859, "num_input_tokens_seen": 88243255, "step": 4100, "time_per_iteration": 3.268444776535034 }, { "auxiliary_loss_clip": 0.01180872, "auxiliary_loss_mlp": 0.01028098, "balance_loss_clip": 1.05311596, "balance_loss_mlp": 1.02070069, "epoch": 0.49311609451091204, "flos": 38837143648800.0, "grad_norm": 1.984322534600522, "language_loss": 0.7210356, "learning_rate": 2.1416541376122207e-06, "loss": 0.74312532, "num_input_tokens_seen": 88263435, "step": 4101, "time_per_iteration": 2.6103734970092773 }, { "auxiliary_loss_clip": 0.0118223, "auxiliary_loss_mlp": 0.01031664, "balance_loss_clip": 1.05199313, "balance_loss_mlp": 1.02395713, "epoch": 0.49323633740155115, "flos": 28329106317600.0, "grad_norm": 2.4352958104567537, "language_loss": 0.72998273, "learning_rate": 2.1408771115420496e-06, "loss": 0.75212169, "num_input_tokens_seen": 88283295, "step": 4102, "time_per_iteration": 2.539581060409546 }, { "auxiliary_loss_clip": 0.01131277, "auxiliary_loss_mlp": 0.01031858, "balance_loss_clip": 1.05181479, "balance_loss_mlp": 1.02486002, "epoch": 0.4933565802921902, "flos": 21135587140320.0, "grad_norm": 1.7056352144491147, "language_loss": 0.64763582, "learning_rate": 2.140100064100932e-06, "loss": 0.66926718, "num_input_tokens_seen": 88299270, "step": 4103, "time_per_iteration": 2.5699424743652344 }, { "auxiliary_loss_clip": 0.01163513, "auxiliary_loss_mlp": 0.01026088, "balance_loss_clip": 1.04998469, "balance_loss_mlp": 1.01905775, "epoch": 0.4934768231828293, "flos": 18039010686720.0, "grad_norm": 4.405105526897506, "language_loss": 0.75833666, "learning_rate": 2.139322995406746e-06, "loss": 0.78023267, "num_input_tokens_seen": 88316905, "step": 4104, "time_per_iteration": 2.476728916168213 }, { "auxiliary_loss_clip": 0.01184856, "auxiliary_loss_mlp": 0.01035041, "balance_loss_clip": 1.05492711, "balance_loss_mlp": 1.02761412, "epoch": 0.4935970660734684, "flos": 23469960272640.0, "grad_norm": 2.089737225011473, "language_loss": 0.79690462, "learning_rate": 2.1385459055773727e-06, "loss": 0.8191036, "num_input_tokens_seen": 88335095, "step": 4105, "time_per_iteration": 2.4878954887390137 }, { "auxiliary_loss_clip": 0.01110738, "auxiliary_loss_mlp": 0.00762016, "balance_loss_clip": 1.04326665, "balance_loss_mlp": 1.00035334, "epoch": 0.4937173089641075, "flos": 64479265698720.0, "grad_norm": 2.071978154783207, "language_loss": 0.73940253, "learning_rate": 2.137768794730696e-06, "loss": 0.75813007, "num_input_tokens_seen": 88358545, "step": 4106, "time_per_iteration": 2.9964516162872314 }, { "auxiliary_loss_clip": 0.01161465, "auxiliary_loss_mlp": 0.01032454, "balance_loss_clip": 1.05372977, "balance_loss_mlp": 1.02384102, "epoch": 0.4938375518547466, "flos": 22346029296480.0, "grad_norm": 1.9370274072364626, "language_loss": 0.80510467, "learning_rate": 2.1369916629846026e-06, "loss": 0.82704383, "num_input_tokens_seen": 88378295, "step": 4107, "time_per_iteration": 2.5500009059906006 }, { "auxiliary_loss_clip": 0.01153506, "auxiliary_loss_mlp": 0.01026423, "balance_loss_clip": 1.04784536, "balance_loss_mlp": 1.01820302, "epoch": 0.4939577947453857, "flos": 17858703572160.0, "grad_norm": 3.5841462502535086, "language_loss": 0.7517522, "learning_rate": 2.136214510456983e-06, "loss": 0.77355146, "num_input_tokens_seen": 88396750, "step": 4108, "time_per_iteration": 2.5413548946380615 }, { "auxiliary_loss_clip": 0.01045333, "auxiliary_loss_mlp": 0.00752374, "balance_loss_clip": 1.01869571, "balance_loss_mlp": 0.99999142, "epoch": 0.49407803763602476, "flos": 70066758148320.0, "grad_norm": 0.8853972606127529, "language_loss": 0.63243103, "learning_rate": 2.1354373372657296e-06, "loss": 0.65040803, "num_input_tokens_seen": 88455190, "step": 4109, "time_per_iteration": 3.2041215896606445 }, { "auxiliary_loss_clip": 0.01185668, "auxiliary_loss_mlp": 0.01036159, "balance_loss_clip": 1.05731678, "balance_loss_mlp": 1.02887487, "epoch": 0.49419828052666387, "flos": 24317489514240.0, "grad_norm": 1.449800119885922, "language_loss": 0.70921463, "learning_rate": 2.1346601435287404e-06, "loss": 0.73143286, "num_input_tokens_seen": 88477460, "step": 4110, "time_per_iteration": 2.498927116394043 }, { "auxiliary_loss_clip": 0.0115371, "auxiliary_loss_mlp": 0.01028033, "balance_loss_clip": 1.05032909, "balance_loss_mlp": 1.0208807, "epoch": 0.494318523417303, "flos": 29386064940000.0, "grad_norm": 2.0572387572597677, "language_loss": 0.80449998, "learning_rate": 2.1338829293639144e-06, "loss": 0.82631737, "num_input_tokens_seen": 88497820, "step": 4111, "time_per_iteration": 2.60697340965271 }, { "auxiliary_loss_clip": 0.01127068, "auxiliary_loss_mlp": 0.0102864, "balance_loss_clip": 1.04742718, "balance_loss_mlp": 1.02089715, "epoch": 0.49443876630794203, "flos": 15268287233280.0, "grad_norm": 1.9733824251219332, "language_loss": 0.831617, "learning_rate": 2.1331056948891547e-06, "loss": 0.85317415, "num_input_tokens_seen": 88514920, "step": 4112, "time_per_iteration": 2.5868546962738037 }, { "auxiliary_loss_clip": 0.01150085, "auxiliary_loss_mlp": 0.01029816, "balance_loss_clip": 1.04943013, "balance_loss_mlp": 1.02229369, "epoch": 0.49455900919858115, "flos": 12347456440320.0, "grad_norm": 2.098400194491306, "language_loss": 0.76084048, "learning_rate": 2.1323284402223666e-06, "loss": 0.7826395, "num_input_tokens_seen": 88530910, "step": 4113, "time_per_iteration": 2.4966461658477783 }, { "auxiliary_loss_clip": 0.01182508, "auxiliary_loss_mlp": 0.00761453, "balance_loss_clip": 1.05682182, "balance_loss_mlp": 1.00035381, "epoch": 0.4946792520892202, "flos": 22779614011680.0, "grad_norm": 1.952119666234262, "language_loss": 0.88049185, "learning_rate": 2.1315511654814597e-06, "loss": 0.89993143, "num_input_tokens_seen": 88549320, "step": 4114, "time_per_iteration": 2.491408586502075 }, { "auxiliary_loss_clip": 0.01146724, "auxiliary_loss_mlp": 0.0102495, "balance_loss_clip": 1.05086744, "balance_loss_mlp": 1.01811385, "epoch": 0.4947994949798593, "flos": 23148129972000.0, "grad_norm": 1.844857090968137, "language_loss": 0.78355634, "learning_rate": 2.1307738707843456e-06, "loss": 0.80527312, "num_input_tokens_seen": 88568985, "step": 4115, "time_per_iteration": 2.558570146560669 }, { "auxiliary_loss_clip": 0.01175833, "auxiliary_loss_mlp": 0.01024988, "balance_loss_clip": 1.05479932, "balance_loss_mlp": 1.01745117, "epoch": 0.4949197378704984, "flos": 23660000883360.0, "grad_norm": 2.097778391107894, "language_loss": 0.68785024, "learning_rate": 2.1299965562489385e-06, "loss": 0.70985842, "num_input_tokens_seen": 88588790, "step": 4116, "time_per_iteration": 2.532836437225342 }, { "auxiliary_loss_clip": 0.01166719, "auxiliary_loss_mlp": 0.01029656, "balance_loss_clip": 1.05030823, "balance_loss_mlp": 1.02231908, "epoch": 0.4950399807611375, "flos": 26911497549120.0, "grad_norm": 1.308295874127114, "language_loss": 0.78839934, "learning_rate": 2.129219221993158e-06, "loss": 0.81036305, "num_input_tokens_seen": 88613575, "step": 4117, "time_per_iteration": 2.638080358505249 }, { "auxiliary_loss_clip": 0.01044136, "auxiliary_loss_mlp": 0.01013374, "balance_loss_clip": 1.01383448, "balance_loss_mlp": 1.0119555, "epoch": 0.4951602236517766, "flos": 67315279797120.0, "grad_norm": 0.7904050005461004, "language_loss": 0.59976709, "learning_rate": 2.128441868134924e-06, "loss": 0.6203422, "num_input_tokens_seen": 88675510, "step": 4118, "time_per_iteration": 3.186309814453125 }, { "auxiliary_loss_clip": 0.01141135, "auxiliary_loss_mlp": 0.01026387, "balance_loss_clip": 1.04696965, "balance_loss_mlp": 1.01931465, "epoch": 0.4952804665424157, "flos": 19901446178880.0, "grad_norm": 2.092427682678262, "language_loss": 0.82684469, "learning_rate": 2.1276644947921606e-06, "loss": 0.84851992, "num_input_tokens_seen": 88694425, "step": 4119, "time_per_iteration": 2.5837790966033936 }, { "auxiliary_loss_clip": 0.01168512, "auxiliary_loss_mlp": 0.01028883, "balance_loss_clip": 1.05133843, "balance_loss_mlp": 1.02079439, "epoch": 0.49540070943305475, "flos": 18806817053760.0, "grad_norm": 1.9006828595213392, "language_loss": 0.82612604, "learning_rate": 2.126887102082795e-06, "loss": 0.84810001, "num_input_tokens_seen": 88714450, "step": 4120, "time_per_iteration": 3.2514305114746094 }, { "auxiliary_loss_clip": 0.01137266, "auxiliary_loss_mlp": 0.01024757, "balance_loss_clip": 1.04514515, "balance_loss_mlp": 1.01713324, "epoch": 0.49552095232369386, "flos": 24934182866880.0, "grad_norm": 1.7125897339271656, "language_loss": 0.70258772, "learning_rate": 2.126109690124757e-06, "loss": 0.724208, "num_input_tokens_seen": 88735265, "step": 4121, "time_per_iteration": 3.38631010055542 }, { "auxiliary_loss_clip": 0.01128764, "auxiliary_loss_mlp": 0.01027953, "balance_loss_clip": 1.04794979, "balance_loss_mlp": 1.02070546, "epoch": 0.495641195214333, "flos": 22857253702560.0, "grad_norm": 2.3176482629477015, "language_loss": 0.71149373, "learning_rate": 2.1253322590359786e-06, "loss": 0.7330609, "num_input_tokens_seen": 88754600, "step": 4122, "time_per_iteration": 2.6120007038116455 }, { "auxiliary_loss_clip": 0.01166002, "auxiliary_loss_mlp": 0.0102573, "balance_loss_clip": 1.050174, "balance_loss_mlp": 1.0184046, "epoch": 0.49576143810497203, "flos": 25769751760800.0, "grad_norm": 2.021073585011941, "language_loss": 0.7392872, "learning_rate": 2.124554808934397e-06, "loss": 0.76120454, "num_input_tokens_seen": 88775180, "step": 4123, "time_per_iteration": 2.5496442317962646 }, { "auxiliary_loss_clip": 0.01118284, "auxiliary_loss_mlp": 0.010225, "balance_loss_clip": 1.04266644, "balance_loss_mlp": 1.01473367, "epoch": 0.49588168099561114, "flos": 22128841852320.0, "grad_norm": 2.0996731239123676, "language_loss": 0.7335, "learning_rate": 2.1237773399379496e-06, "loss": 0.75490785, "num_input_tokens_seen": 88796145, "step": 4124, "time_per_iteration": 3.373563289642334 }, { "auxiliary_loss_clip": 0.01156656, "auxiliary_loss_mlp": 0.01026412, "balance_loss_clip": 1.04656756, "balance_loss_mlp": 1.01869059, "epoch": 0.49600192388625025, "flos": 24387335224800.0, "grad_norm": 1.8127362474632407, "language_loss": 0.86825776, "learning_rate": 2.122999852164578e-06, "loss": 0.89008844, "num_input_tokens_seen": 88816765, "step": 4125, "time_per_iteration": 2.560992479324341 }, { "auxiliary_loss_clip": 0.01118978, "auxiliary_loss_mlp": 0.01028692, "balance_loss_clip": 1.04676747, "balance_loss_mlp": 1.02087748, "epoch": 0.4961221667768893, "flos": 22857433287360.0, "grad_norm": 2.639076021970221, "language_loss": 0.58229387, "learning_rate": 2.122222345732227e-06, "loss": 0.60377061, "num_input_tokens_seen": 88836680, "step": 4126, "time_per_iteration": 3.370112895965576 }, { "auxiliary_loss_clip": 0.01141989, "auxiliary_loss_mlp": 0.01023072, "balance_loss_clip": 1.04884326, "balance_loss_mlp": 1.01528728, "epoch": 0.4962424096675284, "flos": 17858092983840.0, "grad_norm": 1.6724661255755353, "language_loss": 0.82829767, "learning_rate": 2.121444820758843e-06, "loss": 0.84994829, "num_input_tokens_seen": 88855320, "step": 4127, "time_per_iteration": 2.562345504760742 }, { "auxiliary_loss_clip": 0.0112633, "auxiliary_loss_mlp": 0.01038542, "balance_loss_clip": 1.04877651, "balance_loss_mlp": 1.02955985, "epoch": 0.49636265255816747, "flos": 21793614525600.0, "grad_norm": 2.145508372546882, "language_loss": 0.78565288, "learning_rate": 2.120667277362376e-06, "loss": 0.80730164, "num_input_tokens_seen": 88874035, "step": 4128, "time_per_iteration": 2.5763392448425293 }, { "auxiliary_loss_clip": 0.01186252, "auxiliary_loss_mlp": 0.01028121, "balance_loss_clip": 1.05568612, "balance_loss_mlp": 1.02043176, "epoch": 0.4964828954488066, "flos": 16358606323200.0, "grad_norm": 1.9408086581618673, "language_loss": 0.84869409, "learning_rate": 2.1198897156607796e-06, "loss": 0.87083781, "num_input_tokens_seen": 88891390, "step": 4129, "time_per_iteration": 2.538864850997925 }, { "auxiliary_loss_clip": 0.01173073, "auxiliary_loss_mlp": 0.01024224, "balance_loss_clip": 1.05230796, "balance_loss_mlp": 1.01688695, "epoch": 0.4966031383394457, "flos": 24711105041280.0, "grad_norm": 2.1740796004106566, "language_loss": 0.73486078, "learning_rate": 2.1191121357720085e-06, "loss": 0.75683367, "num_input_tokens_seen": 88909450, "step": 4130, "time_per_iteration": 2.52626633644104 }, { "auxiliary_loss_clip": 0.01117305, "auxiliary_loss_mlp": 0.0103744, "balance_loss_clip": 1.04534841, "balance_loss_mlp": 1.0292089, "epoch": 0.49672338123008475, "flos": 22930619275200.0, "grad_norm": 1.7788644174897523, "language_loss": 0.74707353, "learning_rate": 2.1183345378140206e-06, "loss": 0.76862097, "num_input_tokens_seen": 88929195, "step": 4131, "time_per_iteration": 2.630999803543091 }, { "auxiliary_loss_clip": 0.01067335, "auxiliary_loss_mlp": 0.01001432, "balance_loss_clip": 1.01453853, "balance_loss_mlp": 1.00012672, "epoch": 0.49684362412072386, "flos": 65976747397920.0, "grad_norm": 0.8532080200553789, "language_loss": 0.62003297, "learning_rate": 2.1175569219047783e-06, "loss": 0.64072061, "num_input_tokens_seen": 88990635, "step": 4132, "time_per_iteration": 3.209033489227295 }, { "auxiliary_loss_clip": 0.01182841, "auxiliary_loss_mlp": 0.01028933, "balance_loss_clip": 1.05332351, "balance_loss_mlp": 1.02111268, "epoch": 0.49696386701136297, "flos": 19971291889440.0, "grad_norm": 1.6964347978563525, "language_loss": 0.73151922, "learning_rate": 2.1167792881622437e-06, "loss": 0.75363696, "num_input_tokens_seen": 89009655, "step": 4133, "time_per_iteration": 2.5142290592193604 }, { "auxiliary_loss_clip": 0.01151636, "auxiliary_loss_mlp": 0.01027467, "balance_loss_clip": 1.05052114, "balance_loss_mlp": 1.02020729, "epoch": 0.497084109902002, "flos": 24750822810720.0, "grad_norm": 1.5362724998225097, "language_loss": 0.80957723, "learning_rate": 2.116001636704384e-06, "loss": 0.83136827, "num_input_tokens_seen": 89030040, "step": 4134, "time_per_iteration": 2.557209014892578 }, { "auxiliary_loss_clip": 0.011335, "auxiliary_loss_mlp": 0.01025516, "balance_loss_clip": 1.04804897, "balance_loss_mlp": 1.01747525, "epoch": 0.49720435279264114, "flos": 21871828887840.0, "grad_norm": 1.8089221415546404, "language_loss": 0.79976559, "learning_rate": 2.1152239676491685e-06, "loss": 0.82135576, "num_input_tokens_seen": 89048145, "step": 4135, "time_per_iteration": 2.613236665725708 }, { "auxiliary_loss_clip": 0.01158426, "auxiliary_loss_mlp": 0.01027017, "balance_loss_clip": 1.04819834, "balance_loss_mlp": 1.0198015, "epoch": 0.49732459568328025, "flos": 23805798187680.0, "grad_norm": 1.709442866804807, "language_loss": 0.73316753, "learning_rate": 2.114446281114569e-06, "loss": 0.75502193, "num_input_tokens_seen": 89067165, "step": 4136, "time_per_iteration": 2.5575599670410156 }, { "auxiliary_loss_clip": 0.01142964, "auxiliary_loss_mlp": 0.0102776, "balance_loss_clip": 1.04671311, "balance_loss_mlp": 1.02033353, "epoch": 0.4974448385739193, "flos": 20047746320640.0, "grad_norm": 1.7906438225834462, "language_loss": 0.76175952, "learning_rate": 2.1136685772185587e-06, "loss": 0.78346676, "num_input_tokens_seen": 89086190, "step": 4137, "time_per_iteration": 2.5463926792144775 }, { "auxiliary_loss_clip": 0.01155491, "auxiliary_loss_mlp": 0.0076281, "balance_loss_clip": 1.04637611, "balance_loss_mlp": 1.00039268, "epoch": 0.4975650814645584, "flos": 24821350943520.0, "grad_norm": 1.8224930161933564, "language_loss": 0.77623391, "learning_rate": 2.1128908560791163e-06, "loss": 0.79541695, "num_input_tokens_seen": 89106020, "step": 4138, "time_per_iteration": 2.57181978225708 }, { "auxiliary_loss_clip": 0.01182863, "auxiliary_loss_mlp": 0.01029003, "balance_loss_clip": 1.05412233, "balance_loss_mlp": 1.0213083, "epoch": 0.4976853243551975, "flos": 19829481367680.0, "grad_norm": 1.9653611330205178, "language_loss": 0.7808727, "learning_rate": 2.1121131178142203e-06, "loss": 0.80299133, "num_input_tokens_seen": 89125385, "step": 4139, "time_per_iteration": 2.4841277599334717 }, { "auxiliary_loss_clip": 0.01153217, "auxiliary_loss_mlp": 0.01028126, "balance_loss_clip": 1.04897738, "balance_loss_mlp": 1.02028823, "epoch": 0.4978055672458366, "flos": 23142993846720.0, "grad_norm": 1.4731728279660181, "language_loss": 0.82416803, "learning_rate": 2.1113353625418544e-06, "loss": 0.84598148, "num_input_tokens_seen": 89143935, "step": 4140, "time_per_iteration": 2.5360443592071533 }, { "auxiliary_loss_clip": 0.01161385, "auxiliary_loss_mlp": 0.01026274, "balance_loss_clip": 1.05264521, "balance_loss_mlp": 1.01940513, "epoch": 0.4979258101364757, "flos": 15559917758880.0, "grad_norm": 1.597001003100953, "language_loss": 0.78745151, "learning_rate": 2.1105575903800017e-06, "loss": 0.80932808, "num_input_tokens_seen": 89162655, "step": 4141, "time_per_iteration": 2.4980225563049316 }, { "auxiliary_loss_clip": 0.0116855, "auxiliary_loss_mlp": 0.01029451, "balance_loss_clip": 1.05027032, "balance_loss_mlp": 1.02197051, "epoch": 0.4980460530271148, "flos": 26356173504480.0, "grad_norm": 2.0778290297923547, "language_loss": 0.85410053, "learning_rate": 2.1097798014466502e-06, "loss": 0.87608057, "num_input_tokens_seen": 89182255, "step": 4142, "time_per_iteration": 2.5415496826171875 }, { "auxiliary_loss_clip": 0.01172835, "auxiliary_loss_mlp": 0.01025854, "balance_loss_clip": 1.05213761, "balance_loss_mlp": 1.01755428, "epoch": 0.49816629591775385, "flos": 17274544596960.0, "grad_norm": 2.7240611383492728, "language_loss": 0.58933675, "learning_rate": 2.109001995859791e-06, "loss": 0.61132365, "num_input_tokens_seen": 89201155, "step": 4143, "time_per_iteration": 2.4882030487060547 }, { "auxiliary_loss_clip": 0.01052229, "auxiliary_loss_mlp": 0.01001382, "balance_loss_clip": 1.01349938, "balance_loss_mlp": 1.00004661, "epoch": 0.49828653880839296, "flos": 64930958950080.0, "grad_norm": 0.7917160350089935, "language_loss": 0.60080779, "learning_rate": 2.108224173737415e-06, "loss": 0.62134397, "num_input_tokens_seen": 89264455, "step": 4144, "time_per_iteration": 3.118910312652588 }, { "auxiliary_loss_clip": 0.01151435, "auxiliary_loss_mlp": 0.01029366, "balance_loss_clip": 1.04759407, "balance_loss_mlp": 1.02083647, "epoch": 0.498406781699032, "flos": 27484809602400.0, "grad_norm": 1.9983888838326038, "language_loss": 0.76143283, "learning_rate": 2.1074463351975183e-06, "loss": 0.7832408, "num_input_tokens_seen": 89283340, "step": 4145, "time_per_iteration": 2.6076841354370117 }, { "auxiliary_loss_clip": 0.01141931, "auxiliary_loss_mlp": 0.0102942, "balance_loss_clip": 1.04841256, "balance_loss_mlp": 1.02219009, "epoch": 0.49852702458967113, "flos": 31499874433920.0, "grad_norm": 1.723286635309191, "language_loss": 0.71516883, "learning_rate": 2.106668480358098e-06, "loss": 0.73688233, "num_input_tokens_seen": 89303565, "step": 4146, "time_per_iteration": 3.4132585525512695 }, { "auxiliary_loss_clip": 0.01148305, "auxiliary_loss_mlp": 0.01031328, "balance_loss_clip": 1.04673767, "balance_loss_mlp": 1.02283406, "epoch": 0.49864726748031024, "flos": 22852871833440.0, "grad_norm": 1.6288376514108662, "language_loss": 0.7102381, "learning_rate": 2.105890609337154e-06, "loss": 0.73203444, "num_input_tokens_seen": 89322080, "step": 4147, "time_per_iteration": 3.3527936935424805 }, { "auxiliary_loss_clip": 0.01074815, "auxiliary_loss_mlp": 0.0100258, "balance_loss_clip": 1.01291656, "balance_loss_mlp": 1.00129306, "epoch": 0.4987675103709493, "flos": 70405720838880.0, "grad_norm": 0.6936705807028521, "language_loss": 0.63824052, "learning_rate": 2.1051127222526883e-06, "loss": 0.65901446, "num_input_tokens_seen": 89394195, "step": 4148, "time_per_iteration": 3.1819605827331543 }, { "auxiliary_loss_clip": 0.01166359, "auxiliary_loss_mlp": 0.01028746, "balance_loss_clip": 1.05314302, "balance_loss_mlp": 1.02144444, "epoch": 0.4988877532615884, "flos": 28767575739360.0, "grad_norm": 1.5420890969629788, "language_loss": 0.81083781, "learning_rate": 2.1043348192227067e-06, "loss": 0.83278883, "num_input_tokens_seen": 89414565, "step": 4149, "time_per_iteration": 2.564061164855957 }, { "auxiliary_loss_clip": 0.01129292, "auxiliary_loss_mlp": 0.01028835, "balance_loss_clip": 1.04815316, "balance_loss_mlp": 1.02153397, "epoch": 0.4990079961522275, "flos": 16872704086080.0, "grad_norm": 1.7782294971652672, "language_loss": 0.61599171, "learning_rate": 2.1035569003652156e-06, "loss": 0.637573, "num_input_tokens_seen": 89433195, "step": 4150, "time_per_iteration": 3.340402364730835 }, { "auxiliary_loss_clip": 0.01123932, "auxiliary_loss_mlp": 0.01028569, "balance_loss_clip": 1.0468148, "balance_loss_mlp": 1.01918173, "epoch": 0.4991282390428666, "flos": 13291044384960.0, "grad_norm": 1.944803214274958, "language_loss": 0.8180356, "learning_rate": 2.1027789657982255e-06, "loss": 0.83956063, "num_input_tokens_seen": 89447410, "step": 4151, "time_per_iteration": 2.557072162628174 }, { "auxiliary_loss_clip": 0.01123784, "auxiliary_loss_mlp": 0.01027397, "balance_loss_clip": 1.04849708, "balance_loss_mlp": 1.01975513, "epoch": 0.4992484819335057, "flos": 21537499485120.0, "grad_norm": 1.901129582072146, "language_loss": 0.77097511, "learning_rate": 2.1020010156397482e-06, "loss": 0.79248691, "num_input_tokens_seen": 89464630, "step": 4152, "time_per_iteration": 2.5716359615325928 }, { "auxiliary_loss_clip": 0.01166165, "auxiliary_loss_mlp": 0.0103314, "balance_loss_clip": 1.04825759, "balance_loss_mlp": 1.02587986, "epoch": 0.4993687248241448, "flos": 24860098955040.0, "grad_norm": 1.4966247146878453, "language_loss": 0.77370691, "learning_rate": 2.101223050007797e-06, "loss": 0.79569989, "num_input_tokens_seen": 89483180, "step": 4153, "time_per_iteration": 3.270751476287842 }, { "auxiliary_loss_clip": 0.0107323, "auxiliary_loss_mlp": 0.01000508, "balance_loss_clip": 1.01153326, "balance_loss_mlp": 0.99923211, "epoch": 0.49948896771478385, "flos": 62941612086240.0, "grad_norm": 0.816220966190894, "language_loss": 0.53821927, "learning_rate": 2.1004450690203904e-06, "loss": 0.55895662, "num_input_tokens_seen": 89539260, "step": 4154, "time_per_iteration": 3.11563777923584 }, { "auxiliary_loss_clip": 0.01073024, "auxiliary_loss_mlp": 0.01001765, "balance_loss_clip": 1.01142979, "balance_loss_mlp": 1.00048316, "epoch": 0.49960921060542296, "flos": 68284225115520.0, "grad_norm": 0.8598035763804911, "language_loss": 0.63331211, "learning_rate": 2.099667072795546e-06, "loss": 0.65406001, "num_input_tokens_seen": 89601380, "step": 4155, "time_per_iteration": 3.1432831287384033 }, { "auxiliary_loss_clip": 0.01165974, "auxiliary_loss_mlp": 0.01024811, "balance_loss_clip": 1.04982412, "balance_loss_mlp": 1.01717007, "epoch": 0.49972945349606207, "flos": 23659354378080.0, "grad_norm": 1.658157231329812, "language_loss": 0.79699063, "learning_rate": 2.0988890614512864e-06, "loss": 0.81889844, "num_input_tokens_seen": 89621270, "step": 4156, "time_per_iteration": 2.5388028621673584 }, { "auxiliary_loss_clip": 0.0115784, "auxiliary_loss_mlp": 0.01025202, "balance_loss_clip": 1.051844, "balance_loss_mlp": 1.01741767, "epoch": 0.4998496963867011, "flos": 19755828459360.0, "grad_norm": 1.6490596505751625, "language_loss": 0.84237862, "learning_rate": 2.098111035105635e-06, "loss": 0.86420912, "num_input_tokens_seen": 89639695, "step": 4157, "time_per_iteration": 2.5259010791778564 }, { "auxiliary_loss_clip": 0.01122375, "auxiliary_loss_mlp": 0.01029752, "balance_loss_clip": 1.04841268, "balance_loss_mlp": 1.02190852, "epoch": 0.49996993927734024, "flos": 22265731750560.0, "grad_norm": 1.9703616725935913, "language_loss": 0.73207647, "learning_rate": 2.0973329938766176e-06, "loss": 0.75359774, "num_input_tokens_seen": 89657125, "step": 4158, "time_per_iteration": 2.583038806915283 }, { "auxiliary_loss_clip": 0.01174515, "auxiliary_loss_mlp": 0.01033026, "balance_loss_clip": 1.05133581, "balance_loss_mlp": 1.0248127, "epoch": 0.5000901821679793, "flos": 23327215909920.0, "grad_norm": 1.8465318752964375, "language_loss": 0.78656876, "learning_rate": 2.0965549378822618e-06, "loss": 0.80864418, "num_input_tokens_seen": 89678415, "step": 4159, "time_per_iteration": 2.5253520011901855 }, { "auxiliary_loss_clip": 0.01079729, "auxiliary_loss_mlp": 0.01030435, "balance_loss_clip": 1.04148853, "balance_loss_mlp": 1.02255511, "epoch": 0.5002104250586185, "flos": 20339017676640.0, "grad_norm": 1.844671056114194, "language_loss": 0.8433733, "learning_rate": 2.095776867240599e-06, "loss": 0.86447495, "num_input_tokens_seen": 89695405, "step": 4160, "time_per_iteration": 2.728943347930908 }, { "auxiliary_loss_clip": 0.01134854, "auxiliary_loss_mlp": 0.01028504, "balance_loss_clip": 1.0480845, "balance_loss_mlp": 1.02087498, "epoch": 0.5003306679492575, "flos": 13991375560800.0, "grad_norm": 1.9515617301797732, "language_loss": 0.82557648, "learning_rate": 2.094998782069661e-06, "loss": 0.84721005, "num_input_tokens_seen": 89713110, "step": 4161, "time_per_iteration": 2.674123764038086 }, { "auxiliary_loss_clip": 0.01181827, "auxiliary_loss_mlp": 0.01026659, "balance_loss_clip": 1.05324531, "balance_loss_mlp": 1.01927698, "epoch": 0.5004509108398966, "flos": 27672767029440.0, "grad_norm": 1.6269006801714592, "language_loss": 0.75959325, "learning_rate": 2.0942206824874845e-06, "loss": 0.78167814, "num_input_tokens_seen": 89735885, "step": 4162, "time_per_iteration": 2.533989667892456 }, { "auxiliary_loss_clip": 0.01165095, "auxiliary_loss_mlp": 0.01026261, "balance_loss_clip": 1.05090857, "balance_loss_mlp": 1.01770222, "epoch": 0.5005711537305357, "flos": 14976189787200.0, "grad_norm": 2.5258238162687188, "language_loss": 0.78911519, "learning_rate": 2.093442568612105e-06, "loss": 0.81102878, "num_input_tokens_seen": 89753690, "step": 4163, "time_per_iteration": 2.459322690963745 }, { "auxiliary_loss_clip": 0.01182371, "auxiliary_loss_mlp": 0.01025324, "balance_loss_clip": 1.05159688, "balance_loss_mlp": 1.0169909, "epoch": 0.5006913966211748, "flos": 26503263819360.0, "grad_norm": 1.5490899760052335, "language_loss": 0.85619855, "learning_rate": 2.0926644405615613e-06, "loss": 0.87827557, "num_input_tokens_seen": 89774590, "step": 4164, "time_per_iteration": 2.522548198699951 }, { "auxiliary_loss_clip": 0.01135963, "auxiliary_loss_mlp": 0.01025405, "balance_loss_clip": 1.0490644, "balance_loss_mlp": 1.01793027, "epoch": 0.5008116395118138, "flos": 20449299495840.0, "grad_norm": 1.8551807576985524, "language_loss": 0.81562281, "learning_rate": 2.091886298453897e-06, "loss": 0.83723652, "num_input_tokens_seen": 89792775, "step": 4165, "time_per_iteration": 2.544255256652832 }, { "auxiliary_loss_clip": 0.01166513, "auxiliary_loss_mlp": 0.01025569, "balance_loss_clip": 1.05123639, "balance_loss_mlp": 1.01760328, "epoch": 0.500931882402453, "flos": 21579875109600.0, "grad_norm": 2.0172982213154813, "language_loss": 0.72795647, "learning_rate": 2.091108142407153e-06, "loss": 0.74987727, "num_input_tokens_seen": 89811515, "step": 4166, "time_per_iteration": 2.5221707820892334 }, { "auxiliary_loss_clip": 0.01054204, "auxiliary_loss_mlp": 0.01000505, "balance_loss_clip": 1.01895189, "balance_loss_mlp": 0.99919397, "epoch": 0.5010521252930921, "flos": 57785023356480.0, "grad_norm": 0.8445553520350932, "language_loss": 0.62429261, "learning_rate": 2.090329972539377e-06, "loss": 0.64483964, "num_input_tokens_seen": 89870080, "step": 4167, "time_per_iteration": 3.160468578338623 }, { "auxiliary_loss_clip": 0.01075021, "auxiliary_loss_mlp": 0.01023932, "balance_loss_clip": 1.04181647, "balance_loss_mlp": 1.01619565, "epoch": 0.5011723681837311, "flos": 18625504264320.0, "grad_norm": 1.8885174046982451, "language_loss": 0.68639058, "learning_rate": 2.089551788968616e-06, "loss": 0.70738006, "num_input_tokens_seen": 89888045, "step": 4168, "time_per_iteration": 2.753002643585205 }, { "auxiliary_loss_clip": 0.01073743, "auxiliary_loss_mlp": 0.01002452, "balance_loss_clip": 1.0121274, "balance_loss_mlp": 1.00115287, "epoch": 0.5012926110743702, "flos": 55883157430560.0, "grad_norm": 0.8318173559699739, "language_loss": 0.60826051, "learning_rate": 2.08877359181292e-06, "loss": 0.62902242, "num_input_tokens_seen": 89944610, "step": 4169, "time_per_iteration": 3.436154842376709 }, { "auxiliary_loss_clip": 0.0114317, "auxiliary_loss_mlp": 0.01032196, "balance_loss_clip": 1.04545796, "balance_loss_mlp": 1.02449512, "epoch": 0.5014128539650093, "flos": 24238269477120.0, "grad_norm": 2.2335325112464965, "language_loss": 0.85644817, "learning_rate": 2.0879953811903396e-06, "loss": 0.87820184, "num_input_tokens_seen": 89959495, "step": 4170, "time_per_iteration": 2.5774950981140137 }, { "auxiliary_loss_clip": 0.01168482, "auxiliary_loss_mlp": 0.01030441, "balance_loss_clip": 1.05219531, "balance_loss_mlp": 1.02253711, "epoch": 0.5015330968556484, "flos": 27527472562560.0, "grad_norm": 1.8762468450902914, "language_loss": 0.78606141, "learning_rate": 2.08721715721893e-06, "loss": 0.80805063, "num_input_tokens_seen": 89978820, "step": 4171, "time_per_iteration": 2.572605848312378 }, { "auxiliary_loss_clip": 0.01166969, "auxiliary_loss_mlp": 0.01024495, "balance_loss_clip": 1.05197334, "balance_loss_mlp": 1.01693404, "epoch": 0.5016533397462875, "flos": 23800805730240.0, "grad_norm": 2.039607382892118, "language_loss": 0.77197039, "learning_rate": 2.0864389200167477e-06, "loss": 0.79388499, "num_input_tokens_seen": 89997075, "step": 4172, "time_per_iteration": 2.523106813430786 }, { "auxiliary_loss_clip": 0.01167681, "auxiliary_loss_mlp": 0.00763048, "balance_loss_clip": 1.05036974, "balance_loss_mlp": 1.00050783, "epoch": 0.5017735826369266, "flos": 25295012597760.0, "grad_norm": 1.887664820351186, "language_loss": 0.78831792, "learning_rate": 2.0856606697018504e-06, "loss": 0.80762517, "num_input_tokens_seen": 90015085, "step": 4173, "time_per_iteration": 4.098636865615845 }, { "auxiliary_loss_clip": 0.0115077, "auxiliary_loss_mlp": 0.01030493, "balance_loss_clip": 1.04732251, "balance_loss_mlp": 1.02246463, "epoch": 0.5018938255275657, "flos": 16873206923520.0, "grad_norm": 1.9950241759007497, "language_loss": 0.73283947, "learning_rate": 2.084882406392297e-06, "loss": 0.75465208, "num_input_tokens_seen": 90033045, "step": 4174, "time_per_iteration": 2.500641345977783 }, { "auxiliary_loss_clip": 0.01172571, "auxiliary_loss_mlp": 0.010314, "balance_loss_clip": 1.05402684, "balance_loss_mlp": 1.02345431, "epoch": 0.5020140684182047, "flos": 25515432568320.0, "grad_norm": 9.047408221286123, "language_loss": 0.71343738, "learning_rate": 2.0841041302061496e-06, "loss": 0.73547709, "num_input_tokens_seen": 90052505, "step": 4175, "time_per_iteration": 2.5403614044189453 }, { "auxiliary_loss_clip": 0.01142921, "auxiliary_loss_mlp": 0.01029897, "balance_loss_clip": 1.04554987, "balance_loss_mlp": 1.02223158, "epoch": 0.5021343113088439, "flos": 23659282544160.0, "grad_norm": 2.525641277727805, "language_loss": 0.75465536, "learning_rate": 2.083325841261473e-06, "loss": 0.77638352, "num_input_tokens_seen": 90071565, "step": 4176, "time_per_iteration": 3.2709083557128906 }, { "auxiliary_loss_clip": 0.01146337, "auxiliary_loss_mlp": 0.01028891, "balance_loss_clip": 1.04557228, "balance_loss_mlp": 1.02122021, "epoch": 0.502254554199483, "flos": 24534676958400.0, "grad_norm": 2.3020969546561294, "language_loss": 0.65999299, "learning_rate": 2.0825475396763322e-06, "loss": 0.68174529, "num_input_tokens_seen": 90092215, "step": 4177, "time_per_iteration": 2.569423198699951 }, { "auxiliary_loss_clip": 0.01075233, "auxiliary_loss_mlp": 0.0102974, "balance_loss_clip": 1.04038906, "balance_loss_mlp": 1.02183032, "epoch": 0.502374797090122, "flos": 34240290361440.0, "grad_norm": 1.3854553663217573, "language_loss": 0.65657347, "learning_rate": 2.081769225568796e-06, "loss": 0.67762321, "num_input_tokens_seen": 90114665, "step": 4178, "time_per_iteration": 2.893343925476074 }, { "auxiliary_loss_clip": 0.01168643, "auxiliary_loss_mlp": 0.0103015, "balance_loss_clip": 1.04873204, "balance_loss_mlp": 1.02269971, "epoch": 0.5024950399807612, "flos": 26031110677440.0, "grad_norm": 1.4222123937356907, "language_loss": 0.75995421, "learning_rate": 2.0809908990569327e-06, "loss": 0.78194213, "num_input_tokens_seen": 90136445, "step": 4179, "time_per_iteration": 3.5655932426452637 }, { "auxiliary_loss_clip": 0.0115168, "auxiliary_loss_mlp": 0.01031168, "balance_loss_clip": 1.04920173, "balance_loss_mlp": 1.02352667, "epoch": 0.5026152828714002, "flos": 21252441763200.0, "grad_norm": 1.674020259404014, "language_loss": 0.78865135, "learning_rate": 2.0802125602588146e-06, "loss": 0.81047976, "num_input_tokens_seen": 90155710, "step": 4180, "time_per_iteration": 2.5991642475128174 }, { "auxiliary_loss_clip": 0.01183288, "auxiliary_loss_mlp": 0.01026429, "balance_loss_clip": 1.05378044, "balance_loss_mlp": 1.01842415, "epoch": 0.5027355257620393, "flos": 30956115650400.0, "grad_norm": 3.0235521983130687, "language_loss": 0.66579121, "learning_rate": 2.0794342092925146e-06, "loss": 0.68788838, "num_input_tokens_seen": 90176845, "step": 4181, "time_per_iteration": 2.5531365871429443 }, { "auxiliary_loss_clip": 0.01170164, "auxiliary_loss_mlp": 0.01035207, "balance_loss_clip": 1.0524528, "balance_loss_mlp": 1.02788115, "epoch": 0.5028557686526784, "flos": 24791151168480.0, "grad_norm": 1.9274008325517207, "language_loss": 0.67556053, "learning_rate": 2.078655846276108e-06, "loss": 0.69761419, "num_input_tokens_seen": 90197175, "step": 4182, "time_per_iteration": 2.5533578395843506 }, { "auxiliary_loss_clip": 0.01146692, "auxiliary_loss_mlp": 0.01032786, "balance_loss_clip": 1.04772282, "balance_loss_mlp": 1.02482283, "epoch": 0.5029760115433175, "flos": 22966996767360.0, "grad_norm": 2.0585234748180548, "language_loss": 0.6860016, "learning_rate": 2.0778774713276727e-06, "loss": 0.70779634, "num_input_tokens_seen": 90216650, "step": 4183, "time_per_iteration": 2.5578765869140625 }, { "auxiliary_loss_clip": 0.011661, "auxiliary_loss_mlp": 0.01030922, "balance_loss_clip": 1.04920983, "balance_loss_mlp": 1.02314389, "epoch": 0.5030962544339566, "flos": 15305167562880.0, "grad_norm": 1.9007643229019358, "language_loss": 0.67588055, "learning_rate": 2.077099084565287e-06, "loss": 0.69785076, "num_input_tokens_seen": 90234055, "step": 4184, "time_per_iteration": 2.4928526878356934 }, { "auxiliary_loss_clip": 0.01146327, "auxiliary_loss_mlp": 0.01028905, "balance_loss_clip": 1.04489672, "balance_loss_mlp": 1.02026224, "epoch": 0.5032164973245957, "flos": 24494851438080.0, "grad_norm": 2.1441067212415885, "language_loss": 0.65457952, "learning_rate": 2.0763206861070313e-06, "loss": 0.67633188, "num_input_tokens_seen": 90253115, "step": 4185, "time_per_iteration": 2.5515592098236084 }, { "auxiliary_loss_clip": 0.01183668, "auxiliary_loss_mlp": 0.01031037, "balance_loss_clip": 1.05221856, "balance_loss_mlp": 1.02330625, "epoch": 0.5033367402152348, "flos": 16213455524160.0, "grad_norm": 1.9399644065840813, "language_loss": 0.7527777, "learning_rate": 2.0755422760709876e-06, "loss": 0.77492476, "num_input_tokens_seen": 90270515, "step": 4186, "time_per_iteration": 2.4496569633483887 }, { "auxiliary_loss_clip": 0.01116125, "auxiliary_loss_mlp": 0.0102814, "balance_loss_clip": 1.04363894, "balance_loss_mlp": 1.02017033, "epoch": 0.5034569831058738, "flos": 21391378928160.0, "grad_norm": 1.8134620603514107, "language_loss": 0.76745582, "learning_rate": 2.0747638545752417e-06, "loss": 0.78889841, "num_input_tokens_seen": 90289075, "step": 4187, "time_per_iteration": 2.6014273166656494 }, { "auxiliary_loss_clip": 0.01153289, "auxiliary_loss_mlp": 0.01028484, "balance_loss_clip": 1.05212522, "balance_loss_mlp": 1.02141452, "epoch": 0.503577225996513, "flos": 20558755224960.0, "grad_norm": 2.109197700630704, "language_loss": 0.82970667, "learning_rate": 2.073985421737878e-06, "loss": 0.85152441, "num_input_tokens_seen": 90306385, "step": 4188, "time_per_iteration": 2.521315574645996 }, { "auxiliary_loss_clip": 0.01171579, "auxiliary_loss_mlp": 0.01028689, "balance_loss_clip": 1.05266094, "balance_loss_mlp": 1.01999235, "epoch": 0.5036974688871521, "flos": 27229161482400.0, "grad_norm": 2.3326786115549494, "language_loss": 0.74109721, "learning_rate": 2.0732069776769844e-06, "loss": 0.76309991, "num_input_tokens_seen": 90323795, "step": 4189, "time_per_iteration": 2.570091724395752 }, { "auxiliary_loss_clip": 0.01184532, "auxiliary_loss_mlp": 0.01034387, "balance_loss_clip": 1.05430818, "balance_loss_mlp": 1.02579224, "epoch": 0.5038177117777911, "flos": 20412167747520.0, "grad_norm": 2.190688658474577, "language_loss": 0.73233402, "learning_rate": 2.072428522510651e-06, "loss": 0.75452322, "num_input_tokens_seen": 90340360, "step": 4190, "time_per_iteration": 2.452099084854126 }, { "auxiliary_loss_clip": 0.01135135, "auxiliary_loss_mlp": 0.01034695, "balance_loss_clip": 1.04835987, "balance_loss_mlp": 1.02662444, "epoch": 0.5039379546684303, "flos": 21907990878240.0, "grad_norm": 1.986662676946674, "language_loss": 0.76267087, "learning_rate": 2.071650056356968e-06, "loss": 0.78436923, "num_input_tokens_seen": 90357900, "step": 4191, "time_per_iteration": 2.568856716156006 }, { "auxiliary_loss_clip": 0.0118451, "auxiliary_loss_mlp": 0.01029426, "balance_loss_clip": 1.05407071, "balance_loss_mlp": 1.02164793, "epoch": 0.5040581975590693, "flos": 20010722323200.0, "grad_norm": 2.7973849135220106, "language_loss": 0.80201042, "learning_rate": 2.070871579334028e-06, "loss": 0.82414979, "num_input_tokens_seen": 90377010, "step": 4192, "time_per_iteration": 2.466614246368408 }, { "auxiliary_loss_clip": 0.01180494, "auxiliary_loss_mlp": 0.01027384, "balance_loss_clip": 1.04988241, "balance_loss_mlp": 1.01934958, "epoch": 0.5041784404497084, "flos": 20959841479680.0, "grad_norm": 1.5364108500542804, "language_loss": 0.71986705, "learning_rate": 2.0700930915599264e-06, "loss": 0.74194586, "num_input_tokens_seen": 90396740, "step": 4193, "time_per_iteration": 2.483855962753296 }, { "auxiliary_loss_clip": 0.01179317, "auxiliary_loss_mlp": 0.01030989, "balance_loss_clip": 1.05230522, "balance_loss_mlp": 1.02360415, "epoch": 0.5042986833403476, "flos": 12495085509600.0, "grad_norm": 1.98120079585688, "language_loss": 0.7867828, "learning_rate": 2.0693145931527583e-06, "loss": 0.80888587, "num_input_tokens_seen": 90413220, "step": 4194, "time_per_iteration": 2.435701847076416 }, { "auxiliary_loss_clip": 0.01148162, "auxiliary_loss_mlp": 0.01027943, "balance_loss_clip": 1.0491035, "balance_loss_mlp": 1.02046835, "epoch": 0.5044189262309866, "flos": 29202309797280.0, "grad_norm": 1.6284686320491755, "language_loss": 0.77910924, "learning_rate": 2.068536084230622e-06, "loss": 0.80087024, "num_input_tokens_seen": 90435085, "step": 4195, "time_per_iteration": 2.5958006381988525 }, { "auxiliary_loss_clip": 0.01167188, "auxiliary_loss_mlp": 0.01030501, "balance_loss_clip": 1.05136204, "balance_loss_mlp": 1.02274632, "epoch": 0.5045391691216257, "flos": 23873201544960.0, "grad_norm": 2.0000946072294545, "language_loss": 0.88879555, "learning_rate": 2.067757564911616e-06, "loss": 0.91077244, "num_input_tokens_seen": 90453660, "step": 4196, "time_per_iteration": 2.515608787536621 }, { "auxiliary_loss_clip": 0.01161695, "auxiliary_loss_mlp": 0.00762345, "balance_loss_clip": 1.05000365, "balance_loss_mlp": 1.00038803, "epoch": 0.5046594120122648, "flos": 24644994694560.0, "grad_norm": 2.2538244844787836, "language_loss": 0.92512238, "learning_rate": 2.0669790353138407e-06, "loss": 0.94436282, "num_input_tokens_seen": 90472625, "step": 4197, "time_per_iteration": 2.5764658451080322 }, { "auxiliary_loss_clip": 0.01134852, "auxiliary_loss_mlp": 0.00762557, "balance_loss_clip": 1.04842901, "balance_loss_mlp": 1.00042415, "epoch": 0.5047796549029039, "flos": 23362839145920.0, "grad_norm": 2.28702965796275, "language_loss": 0.72973013, "learning_rate": 2.0662004955553995e-06, "loss": 0.74870414, "num_input_tokens_seen": 90492325, "step": 4198, "time_per_iteration": 2.568605899810791 }, { "auxiliary_loss_clip": 0.01148424, "auxiliary_loss_mlp": 0.01027611, "balance_loss_clip": 1.0467416, "balance_loss_mlp": 1.02051783, "epoch": 0.5048998977935429, "flos": 17304097866720.0, "grad_norm": 2.23834466549969, "language_loss": 0.76814061, "learning_rate": 2.065421945754395e-06, "loss": 0.78990096, "num_input_tokens_seen": 90510055, "step": 4199, "time_per_iteration": 3.4811651706695557 }, { "auxiliary_loss_clip": 0.01131012, "auxiliary_loss_mlp": 0.01028935, "balance_loss_clip": 1.05113328, "balance_loss_mlp": 1.0218482, "epoch": 0.505020140684182, "flos": 34856983714080.0, "grad_norm": 1.774687883213498, "language_loss": 0.78034234, "learning_rate": 2.0646433860289344e-06, "loss": 0.80194181, "num_input_tokens_seen": 90528980, "step": 4200, "time_per_iteration": 2.7081902027130127 }, { "auxiliary_loss_clip": 0.01172408, "auxiliary_loss_mlp": 0.00763169, "balance_loss_clip": 1.05042708, "balance_loss_mlp": 1.00044179, "epoch": 0.5051403835748212, "flos": 24863977986720.0, "grad_norm": 1.8343355120386722, "language_loss": 0.82384837, "learning_rate": 2.0638648164971233e-06, "loss": 0.84320414, "num_input_tokens_seen": 90547445, "step": 4201, "time_per_iteration": 2.5430009365081787 }, { "auxiliary_loss_clip": 0.01154202, "auxiliary_loss_mlp": 0.01025981, "balance_loss_clip": 1.05243254, "balance_loss_mlp": 1.01902497, "epoch": 0.5052606264654602, "flos": 20959697811840.0, "grad_norm": 1.8288582885540805, "language_loss": 0.88398558, "learning_rate": 2.06308623727707e-06, "loss": 0.90578735, "num_input_tokens_seen": 90567545, "step": 4202, "time_per_iteration": 3.2359907627105713 }, { "auxiliary_loss_clip": 0.01164112, "auxiliary_loss_mlp": 0.01028497, "balance_loss_clip": 1.050035, "balance_loss_mlp": 1.0204922, "epoch": 0.5053808693560993, "flos": 19642385947680.0, "grad_norm": 2.183989420546069, "language_loss": 0.76368386, "learning_rate": 2.0623076484868846e-06, "loss": 0.78560996, "num_input_tokens_seen": 90585000, "step": 4203, "time_per_iteration": 2.4753880500793457 }, { "auxiliary_loss_clip": 0.01052673, "auxiliary_loss_mlp": 0.01003097, "balance_loss_clip": 1.01513028, "balance_loss_mlp": 1.00178552, "epoch": 0.5055011122467384, "flos": 67504925321280.0, "grad_norm": 0.8263074530508284, "language_loss": 0.60609066, "learning_rate": 2.061529050244679e-06, "loss": 0.62664837, "num_input_tokens_seen": 90644745, "step": 4204, "time_per_iteration": 3.0973265171051025 }, { "auxiliary_loss_clip": 0.01143239, "auxiliary_loss_mlp": 0.01023021, "balance_loss_clip": 1.04748058, "balance_loss_mlp": 1.0149951, "epoch": 0.5056213551373775, "flos": 16872955504800.0, "grad_norm": 2.00154743072006, "language_loss": 0.74437279, "learning_rate": 2.060750442668565e-06, "loss": 0.76603538, "num_input_tokens_seen": 90662500, "step": 4205, "time_per_iteration": 3.257826805114746 }, { "auxiliary_loss_clip": 0.01167315, "auxiliary_loss_mlp": 0.01025741, "balance_loss_clip": 1.05190921, "balance_loss_mlp": 1.01749194, "epoch": 0.5057415980280165, "flos": 15334182078240.0, "grad_norm": 2.4518910397723435, "language_loss": 0.6385442, "learning_rate": 2.059971825876657e-06, "loss": 0.66047478, "num_input_tokens_seen": 90677010, "step": 4206, "time_per_iteration": 2.474360942840576 }, { "auxiliary_loss_clip": 0.01170119, "auxiliary_loss_mlp": 0.01030052, "balance_loss_clip": 1.0512166, "balance_loss_mlp": 1.02244651, "epoch": 0.5058618409186557, "flos": 19025979930720.0, "grad_norm": 1.8968107965618624, "language_loss": 0.76533622, "learning_rate": 2.0591931999870713e-06, "loss": 0.7873379, "num_input_tokens_seen": 90695935, "step": 4207, "time_per_iteration": 2.489136219024658 }, { "auxiliary_loss_clip": 0.01060838, "auxiliary_loss_mlp": 0.01001895, "balance_loss_clip": 1.01391566, "balance_loss_mlp": 1.00058961, "epoch": 0.5059820838092948, "flos": 63453123828000.0, "grad_norm": 0.8162192334829667, "language_loss": 0.57587796, "learning_rate": 2.0584145651179234e-06, "loss": 0.59650534, "num_input_tokens_seen": 90751645, "step": 4208, "time_per_iteration": 3.124490261077881 }, { "auxiliary_loss_clip": 0.01157171, "auxiliary_loss_mlp": 0.0076251, "balance_loss_clip": 1.05370116, "balance_loss_mlp": 1.0003736, "epoch": 0.5061023266999338, "flos": 15441806042400.0, "grad_norm": 2.1475376127772097, "language_loss": 0.79894114, "learning_rate": 2.0576359213873327e-06, "loss": 0.81813794, "num_input_tokens_seen": 90766795, "step": 4209, "time_per_iteration": 2.524533271789551 }, { "auxiliary_loss_clip": 0.01162576, "auxiliary_loss_mlp": 0.01030351, "balance_loss_clip": 1.04910481, "balance_loss_mlp": 1.02202404, "epoch": 0.506222569590573, "flos": 22451067239520.0, "grad_norm": 6.969195629859045, "language_loss": 0.70236129, "learning_rate": 2.056857268913419e-06, "loss": 0.72429049, "num_input_tokens_seen": 90786845, "step": 4210, "time_per_iteration": 2.55413818359375 }, { "auxiliary_loss_clip": 0.0116472, "auxiliary_loss_mlp": 0.01024059, "balance_loss_clip": 1.04966533, "balance_loss_mlp": 1.01662004, "epoch": 0.506342812481212, "flos": 17558668477920.0, "grad_norm": 3.35647800911695, "language_loss": 0.83987749, "learning_rate": 2.056078607814303e-06, "loss": 0.86176527, "num_input_tokens_seen": 90802630, "step": 4211, "time_per_iteration": 2.463521957397461 }, { "auxiliary_loss_clip": 0.01167365, "auxiliary_loss_mlp": 0.0102586, "balance_loss_clip": 1.05172205, "balance_loss_mlp": 1.01859999, "epoch": 0.5064630553718511, "flos": 23402054077920.0, "grad_norm": 2.2418528005237324, "language_loss": 0.78836596, "learning_rate": 2.055299938208106e-06, "loss": 0.8102982, "num_input_tokens_seen": 90823620, "step": 4212, "time_per_iteration": 2.541578769683838 }, { "auxiliary_loss_clip": 0.01172528, "auxiliary_loss_mlp": 0.01025969, "balance_loss_clip": 1.05289805, "balance_loss_mlp": 1.0184381, "epoch": 0.5065832982624903, "flos": 23987039143200.0, "grad_norm": 1.6685391283035944, "language_loss": 0.86111617, "learning_rate": 2.0545212602129526e-06, "loss": 0.88310122, "num_input_tokens_seen": 90843475, "step": 4213, "time_per_iteration": 2.503352642059326 }, { "auxiliary_loss_clip": 0.01146366, "auxiliary_loss_mlp": 0.01030824, "balance_loss_clip": 1.04682434, "balance_loss_mlp": 1.02285457, "epoch": 0.5067035411531293, "flos": 21503061508800.0, "grad_norm": 1.9665111491102507, "language_loss": 0.658535, "learning_rate": 2.0537425739469673e-06, "loss": 0.68030685, "num_input_tokens_seen": 90862410, "step": 4214, "time_per_iteration": 2.5272891521453857 }, { "auxiliary_loss_clip": 0.01068258, "auxiliary_loss_mlp": 0.01001299, "balance_loss_clip": 1.01576996, "balance_loss_mlp": 0.99997586, "epoch": 0.5068237840437684, "flos": 65934407690400.0, "grad_norm": 0.8437777861061613, "language_loss": 0.59496152, "learning_rate": 2.052963879528276e-06, "loss": 0.61565709, "num_input_tokens_seen": 90922280, "step": 4215, "time_per_iteration": 3.0766563415527344 }, { "auxiliary_loss_clip": 0.01170355, "auxiliary_loss_mlp": 0.01033229, "balance_loss_clip": 1.05307472, "balance_loss_mlp": 1.02460456, "epoch": 0.5069440269344075, "flos": 27264210047040.0, "grad_norm": 2.073128309186873, "language_loss": 0.7683835, "learning_rate": 2.052185177075007e-06, "loss": 0.79041934, "num_input_tokens_seen": 90941850, "step": 4216, "time_per_iteration": 2.5510354042053223 }, { "auxiliary_loss_clip": 0.01169424, "auxiliary_loss_mlp": 0.01031381, "balance_loss_clip": 1.05050659, "balance_loss_mlp": 1.02320313, "epoch": 0.5070642698250466, "flos": 23366323091040.0, "grad_norm": 2.8146460835722946, "language_loss": 0.82837605, "learning_rate": 2.051406466705288e-06, "loss": 0.85038412, "num_input_tokens_seen": 90961390, "step": 4217, "time_per_iteration": 2.521271228790283 }, { "auxiliary_loss_clip": 0.01180321, "auxiliary_loss_mlp": 0.01020769, "balance_loss_clip": 1.05007851, "balance_loss_mlp": 1.013363, "epoch": 0.5071845127156857, "flos": 20340131102400.0, "grad_norm": 1.8056871784938062, "language_loss": 0.80943465, "learning_rate": 2.0506277485372486e-06, "loss": 0.83144546, "num_input_tokens_seen": 90980215, "step": 4218, "time_per_iteration": 2.4807863235473633 }, { "auxiliary_loss_clip": 0.01160133, "auxiliary_loss_mlp": 0.01026433, "balance_loss_clip": 1.04836369, "balance_loss_mlp": 1.0186727, "epoch": 0.5073047556063248, "flos": 12092957663040.0, "grad_norm": 1.9643887458094476, "language_loss": 0.66982585, "learning_rate": 2.04984902268902e-06, "loss": 0.69169152, "num_input_tokens_seen": 90997415, "step": 4219, "time_per_iteration": 2.490832567214966 }, { "auxiliary_loss_clip": 0.01176377, "auxiliary_loss_mlp": 0.01032706, "balance_loss_clip": 1.05176353, "balance_loss_mlp": 1.0244627, "epoch": 0.5074249984969639, "flos": 19682857973280.0, "grad_norm": 2.1468523875243455, "language_loss": 0.75591272, "learning_rate": 2.0490702892787345e-06, "loss": 0.77800357, "num_input_tokens_seen": 91016475, "step": 4220, "time_per_iteration": 2.5193331241607666 }, { "auxiliary_loss_clip": 0.01155395, "auxiliary_loss_mlp": 0.01024381, "balance_loss_clip": 1.04558063, "balance_loss_mlp": 1.01721966, "epoch": 0.5075452413876029, "flos": 28765708057440.0, "grad_norm": 2.1226803164454364, "language_loss": 0.61926138, "learning_rate": 2.0482915484245246e-06, "loss": 0.64105922, "num_input_tokens_seen": 91038095, "step": 4221, "time_per_iteration": 2.5611867904663086 }, { "auxiliary_loss_clip": 0.01119663, "auxiliary_loss_mlp": 0.01029516, "balance_loss_clip": 1.04670393, "balance_loss_mlp": 1.02154708, "epoch": 0.5076654842782421, "flos": 20339448680160.0, "grad_norm": 2.4452649863523632, "language_loss": 0.84074008, "learning_rate": 2.047512800244526e-06, "loss": 0.86223191, "num_input_tokens_seen": 91053360, "step": 4222, "time_per_iteration": 2.5786633491516113 }, { "auxiliary_loss_clip": 0.01168052, "auxiliary_loss_mlp": 0.01027581, "balance_loss_clip": 1.05148864, "balance_loss_mlp": 1.02008224, "epoch": 0.5077857271688812, "flos": 26359657449600.0, "grad_norm": 2.3934822641381444, "language_loss": 0.78643262, "learning_rate": 2.046734044856873e-06, "loss": 0.80838895, "num_input_tokens_seen": 91072770, "step": 4223, "time_per_iteration": 2.5644335746765137 }, { "auxiliary_loss_clip": 0.01168104, "auxiliary_loss_mlp": 0.01032588, "balance_loss_clip": 1.05171251, "balance_loss_mlp": 1.02519751, "epoch": 0.5079059700595202, "flos": 21798966152640.0, "grad_norm": 1.8496443287124595, "language_loss": 0.813667, "learning_rate": 2.045955282379702e-06, "loss": 0.83567393, "num_input_tokens_seen": 91091430, "step": 4224, "time_per_iteration": 2.4995036125183105 }, { "auxiliary_loss_clip": 0.01164439, "auxiliary_loss_mlp": 0.01025157, "balance_loss_clip": 1.04761827, "balance_loss_mlp": 1.01660371, "epoch": 0.5080262129501594, "flos": 13187945957760.0, "grad_norm": 3.6886369944347077, "language_loss": 0.76128399, "learning_rate": 2.045176512931152e-06, "loss": 0.78317994, "num_input_tokens_seen": 91106060, "step": 4225, "time_per_iteration": 3.9812190532684326 }, { "auxiliary_loss_clip": 0.01140242, "auxiliary_loss_mlp": 0.01027997, "balance_loss_clip": 1.04923213, "balance_loss_mlp": 1.02110708, "epoch": 0.5081464558407984, "flos": 25301477650560.0, "grad_norm": 1.9009024768019787, "language_loss": 0.75747752, "learning_rate": 2.0443977366293604e-06, "loss": 0.77916002, "num_input_tokens_seen": 91124100, "step": 4226, "time_per_iteration": 2.587684154510498 }, { "auxiliary_loss_clip": 0.0111184, "auxiliary_loss_mlp": 0.0102824, "balance_loss_clip": 1.04314768, "balance_loss_mlp": 1.01944804, "epoch": 0.5082666987314375, "flos": 30951230943840.0, "grad_norm": 1.5881226447832557, "language_loss": 0.76698208, "learning_rate": 2.043618953592468e-06, "loss": 0.78838289, "num_input_tokens_seen": 91146555, "step": 4227, "time_per_iteration": 2.7034695148468018 }, { "auxiliary_loss_clip": 0.01154744, "auxiliary_loss_mlp": 0.01034239, "balance_loss_clip": 1.0516783, "balance_loss_mlp": 1.02649617, "epoch": 0.5083869416220766, "flos": 19682498803680.0, "grad_norm": 1.6770652154333907, "language_loss": 0.81109381, "learning_rate": 2.0428401639386144e-06, "loss": 0.83298361, "num_input_tokens_seen": 91167120, "step": 4228, "time_per_iteration": 3.285646915435791 }, { "auxiliary_loss_clip": 0.01048646, "auxiliary_loss_mlp": 0.01001402, "balance_loss_clip": 1.01322007, "balance_loss_mlp": 0.99998897, "epoch": 0.5085071845127157, "flos": 71817546976800.0, "grad_norm": 0.8172038078616869, "language_loss": 0.58105874, "learning_rate": 2.042061367785943e-06, "loss": 0.60155916, "num_input_tokens_seen": 91220260, "step": 4229, "time_per_iteration": 3.0973715782165527 }, { "auxiliary_loss_clip": 0.01139588, "auxiliary_loss_mlp": 0.01024507, "balance_loss_clip": 1.04442644, "balance_loss_mlp": 1.01644802, "epoch": 0.5086274274033548, "flos": 35951612839200.0, "grad_norm": 2.03558728205703, "language_loss": 0.75293142, "learning_rate": 2.041282565252594e-06, "loss": 0.77457237, "num_input_tokens_seen": 91240425, "step": 4230, "time_per_iteration": 2.7002201080322266 }, { "auxiliary_loss_clip": 0.01138945, "auxiliary_loss_mlp": 0.01027294, "balance_loss_clip": 1.04734075, "balance_loss_mlp": 1.01948583, "epoch": 0.5087476702939938, "flos": 23513736658560.0, "grad_norm": 1.767577555074192, "language_loss": 0.77078891, "learning_rate": 2.040503756456714e-06, "loss": 0.79245138, "num_input_tokens_seen": 91259635, "step": 4231, "time_per_iteration": 3.2900426387786865 }, { "auxiliary_loss_clip": 0.01161856, "auxiliary_loss_mlp": 0.01024237, "balance_loss_clip": 1.04835916, "balance_loss_mlp": 1.01685786, "epoch": 0.508867913184633, "flos": 15122094842400.0, "grad_norm": 2.0613990265212996, "language_loss": 0.78714573, "learning_rate": 2.0397249415164456e-06, "loss": 0.80900669, "num_input_tokens_seen": 91276990, "step": 4232, "time_per_iteration": 2.4958488941192627 }, { "auxiliary_loss_clip": 0.01145393, "auxiliary_loss_mlp": 0.01030955, "balance_loss_clip": 1.04562521, "balance_loss_mlp": 1.02317691, "epoch": 0.508988156075272, "flos": 25885313373120.0, "grad_norm": 1.5262956480100731, "language_loss": 0.80010438, "learning_rate": 2.0389461205499354e-06, "loss": 0.82186782, "num_input_tokens_seen": 91296125, "step": 4233, "time_per_iteration": 2.5793702602386475 }, { "auxiliary_loss_clip": 0.01142439, "auxiliary_loss_mlp": 0.01027356, "balance_loss_clip": 1.04747605, "balance_loss_mlp": 1.01955926, "epoch": 0.5091083989659111, "flos": 13844859917280.0, "grad_norm": 1.7665639105945663, "language_loss": 0.73512238, "learning_rate": 2.03816729367533e-06, "loss": 0.75682032, "num_input_tokens_seen": 91314280, "step": 4234, "time_per_iteration": 2.5734801292419434 }, { "auxiliary_loss_clip": 0.01156164, "auxiliary_loss_mlp": 0.01030579, "balance_loss_clip": 1.05067968, "balance_loss_mlp": 1.02165055, "epoch": 0.5092286418565503, "flos": 21104884527840.0, "grad_norm": 2.2210799299690436, "language_loss": 0.71757853, "learning_rate": 2.0373884610107765e-06, "loss": 0.73944598, "num_input_tokens_seen": 91334595, "step": 4235, "time_per_iteration": 2.5446579456329346 }, { "auxiliary_loss_clip": 0.01170732, "auxiliary_loss_mlp": 0.01025607, "balance_loss_clip": 1.04914939, "balance_loss_mlp": 1.01781046, "epoch": 0.5093488847471893, "flos": 18621301980000.0, "grad_norm": 2.2308569634100928, "language_loss": 0.69295037, "learning_rate": 2.0366096226744225e-06, "loss": 0.71491373, "num_input_tokens_seen": 91349790, "step": 4236, "time_per_iteration": 2.49855637550354 }, { "auxiliary_loss_clip": 0.01157294, "auxiliary_loss_mlp": 0.01028438, "balance_loss_clip": 1.04800725, "balance_loss_mlp": 1.02080226, "epoch": 0.5094691276378284, "flos": 23803786837920.0, "grad_norm": 1.6705733521023216, "language_loss": 0.76867759, "learning_rate": 2.035830778784418e-06, "loss": 0.79053485, "num_input_tokens_seen": 91370465, "step": 4237, "time_per_iteration": 2.5406110286712646 }, { "auxiliary_loss_clip": 0.01156336, "auxiliary_loss_mlp": 0.01030359, "balance_loss_clip": 1.05339265, "balance_loss_mlp": 1.02335286, "epoch": 0.5095893705284675, "flos": 17420413735200.0, "grad_norm": 2.010847929591217, "language_loss": 0.79795879, "learning_rate": 2.0350519294589134e-06, "loss": 0.81982577, "num_input_tokens_seen": 91388505, "step": 4238, "time_per_iteration": 2.5579686164855957 }, { "auxiliary_loss_clip": 0.01124116, "auxiliary_loss_mlp": 0.01025284, "balance_loss_clip": 1.04653621, "balance_loss_mlp": 1.01719558, "epoch": 0.5097096134191066, "flos": 25849366884480.0, "grad_norm": 1.7552974354389352, "language_loss": 0.82683015, "learning_rate": 2.0342730748160588e-06, "loss": 0.84832418, "num_input_tokens_seen": 91408970, "step": 4239, "time_per_iteration": 2.651578903198242 }, { "auxiliary_loss_clip": 0.01152914, "auxiliary_loss_mlp": 0.01027985, "balance_loss_clip": 1.04730582, "balance_loss_mlp": 1.01961017, "epoch": 0.5098298563097456, "flos": 27745126927200.0, "grad_norm": 2.1774739414396476, "language_loss": 0.70475185, "learning_rate": 2.033494214974006e-06, "loss": 0.72656089, "num_input_tokens_seen": 91430115, "step": 4240, "time_per_iteration": 2.6048583984375 }, { "auxiliary_loss_clip": 0.01141741, "auxiliary_loss_mlp": 0.01028436, "balance_loss_clip": 1.04716277, "balance_loss_mlp": 1.02067208, "epoch": 0.5099500992003848, "flos": 21358916384640.0, "grad_norm": 1.8086823201336792, "language_loss": 0.84042883, "learning_rate": 2.0327153500509067e-06, "loss": 0.86213064, "num_input_tokens_seen": 91449140, "step": 4241, "time_per_iteration": 2.600757360458374 }, { "auxiliary_loss_clip": 0.01156837, "auxiliary_loss_mlp": 0.01029898, "balance_loss_clip": 1.05210423, "balance_loss_mlp": 1.02199435, "epoch": 0.5100703420910239, "flos": 19866002527680.0, "grad_norm": 1.911386686079991, "language_loss": 0.8452307, "learning_rate": 2.031936480164916e-06, "loss": 0.86709797, "num_input_tokens_seen": 91466880, "step": 4242, "time_per_iteration": 2.548578977584839 }, { "auxiliary_loss_clip": 0.01151478, "auxiliary_loss_mlp": 0.01029766, "balance_loss_clip": 1.05199146, "balance_loss_mlp": 1.02203488, "epoch": 0.5101905849816629, "flos": 24648801892320.0, "grad_norm": 2.650230992824248, "language_loss": 0.80092752, "learning_rate": 2.0311576054341857e-06, "loss": 0.8227399, "num_input_tokens_seen": 91487495, "step": 4243, "time_per_iteration": 2.6466379165649414 }, { "auxiliary_loss_clip": 0.01184499, "auxiliary_loss_mlp": 0.01024598, "balance_loss_clip": 1.05472994, "balance_loss_mlp": 1.01700115, "epoch": 0.5103108278723021, "flos": 22930080520800.0, "grad_norm": 1.5550206415444747, "language_loss": 0.62486553, "learning_rate": 2.0303787259768715e-06, "loss": 0.64695644, "num_input_tokens_seen": 91508395, "step": 4244, "time_per_iteration": 2.4935033321380615 }, { "auxiliary_loss_clip": 0.01156534, "auxiliary_loss_mlp": 0.01029115, "balance_loss_clip": 1.05273736, "balance_loss_mlp": 1.02098489, "epoch": 0.5104310707629411, "flos": 21506617287840.0, "grad_norm": 2.3997677810025455, "language_loss": 0.69082499, "learning_rate": 2.0295998419111294e-06, "loss": 0.71268153, "num_input_tokens_seen": 91525685, "step": 4245, "time_per_iteration": 2.5648152828216553 }, { "auxiliary_loss_clip": 0.01111456, "auxiliary_loss_mlp": 0.01030268, "balance_loss_clip": 1.0430119, "balance_loss_mlp": 1.02299619, "epoch": 0.5105513136535802, "flos": 14903183384160.0, "grad_norm": 3.151851445638551, "language_loss": 0.74346739, "learning_rate": 2.028820953355115e-06, "loss": 0.76488459, "num_input_tokens_seen": 91543785, "step": 4246, "time_per_iteration": 2.608222723007202 }, { "auxiliary_loss_clip": 0.01160972, "auxiliary_loss_mlp": 0.01033284, "balance_loss_clip": 1.04984117, "balance_loss_mlp": 1.0253979, "epoch": 0.5106715565442194, "flos": 22602216170880.0, "grad_norm": 1.713861786180026, "language_loss": 0.78236616, "learning_rate": 2.0280420604269834e-06, "loss": 0.80430877, "num_input_tokens_seen": 91563325, "step": 4247, "time_per_iteration": 2.582367181777954 }, { "auxiliary_loss_clip": 0.01062371, "auxiliary_loss_mlp": 0.01004179, "balance_loss_clip": 1.01363635, "balance_loss_mlp": 1.00289714, "epoch": 0.5107917994348584, "flos": 71027658067680.0, "grad_norm": 0.7065420874302002, "language_loss": 0.58929133, "learning_rate": 2.027263163244895e-06, "loss": 0.6099568, "num_input_tokens_seen": 91632450, "step": 4248, "time_per_iteration": 3.2534377574920654 }, { "auxiliary_loss_clip": 0.01166016, "auxiliary_loss_mlp": 0.01033089, "balance_loss_clip": 1.05211055, "balance_loss_mlp": 1.02519107, "epoch": 0.5109120423254975, "flos": 24827420909760.0, "grad_norm": 1.5875221630957883, "language_loss": 0.74415123, "learning_rate": 2.026484261927005e-06, "loss": 0.76614225, "num_input_tokens_seen": 91651945, "step": 4249, "time_per_iteration": 2.559859275817871 }, { "auxiliary_loss_clip": 0.01177564, "auxiliary_loss_mlp": 0.01028852, "balance_loss_clip": 1.05517817, "balance_loss_mlp": 1.02092767, "epoch": 0.5110322852161366, "flos": 21247664807520.0, "grad_norm": 2.1272864842452424, "language_loss": 0.74519795, "learning_rate": 2.025705356591475e-06, "loss": 0.7672621, "num_input_tokens_seen": 91669635, "step": 4250, "time_per_iteration": 2.4961700439453125 }, { "auxiliary_loss_clip": 0.01037486, "auxiliary_loss_mlp": 0.00752546, "balance_loss_clip": 1.01086974, "balance_loss_mlp": 0.99986506, "epoch": 0.5111525281067757, "flos": 66457125523680.0, "grad_norm": 0.7580002862426913, "language_loss": 0.57934892, "learning_rate": 2.024926447356462e-06, "loss": 0.59724915, "num_input_tokens_seen": 91731920, "step": 4251, "time_per_iteration": 4.58832859992981 }, { "auxiliary_loss_clip": 0.01168124, "auxiliary_loss_mlp": 0.01030854, "balance_loss_clip": 1.05153346, "balance_loss_mlp": 1.02264047, "epoch": 0.5112727709974147, "flos": 14866734058080.0, "grad_norm": 1.8652991157480223, "language_loss": 0.78196275, "learning_rate": 2.024147534340127e-06, "loss": 0.80395257, "num_input_tokens_seen": 91749780, "step": 4252, "time_per_iteration": 2.5037577152252197 }, { "auxiliary_loss_clip": 0.01145968, "auxiliary_loss_mlp": 0.01026329, "balance_loss_clip": 1.04576385, "balance_loss_mlp": 1.0188787, "epoch": 0.5113930138880539, "flos": 21177603595200.0, "grad_norm": 1.7277907222892455, "language_loss": 0.79716539, "learning_rate": 2.02336861766063e-06, "loss": 0.81888837, "num_input_tokens_seen": 91768840, "step": 4253, "time_per_iteration": 2.5542895793914795 }, { "auxiliary_loss_clip": 0.01174672, "auxiliary_loss_mlp": 0.0102953, "balance_loss_clip": 1.05186462, "balance_loss_mlp": 1.02198982, "epoch": 0.511513256778693, "flos": 20409114805920.0, "grad_norm": 1.7879272265068533, "language_loss": 0.78975677, "learning_rate": 2.0225896974361327e-06, "loss": 0.81179881, "num_input_tokens_seen": 91788945, "step": 4254, "time_per_iteration": 3.2857868671417236 }, { "auxiliary_loss_clip": 0.01041221, "auxiliary_loss_mlp": 0.01003595, "balance_loss_clip": 1.0127275, "balance_loss_mlp": 1.00233161, "epoch": 0.511633499669332, "flos": 69879985980960.0, "grad_norm": 1.095871033793626, "language_loss": 0.60014379, "learning_rate": 2.0218107737847962e-06, "loss": 0.62059188, "num_input_tokens_seen": 91850990, "step": 4255, "time_per_iteration": 3.193185567855835 }, { "auxiliary_loss_clip": 0.01180057, "auxiliary_loss_mlp": 0.01025542, "balance_loss_clip": 1.05224824, "balance_loss_mlp": 1.01795483, "epoch": 0.5117537425599712, "flos": 24097859716800.0, "grad_norm": 1.8587497403515167, "language_loss": 0.74934101, "learning_rate": 2.0210318468247826e-06, "loss": 0.77139699, "num_input_tokens_seen": 91869960, "step": 4256, "time_per_iteration": 2.521022081375122 }, { "auxiliary_loss_clip": 0.01151621, "auxiliary_loss_mlp": 0.01030033, "balance_loss_clip": 1.04729199, "balance_loss_mlp": 1.0229578, "epoch": 0.5118739854506102, "flos": 20959554144000.0, "grad_norm": 1.9526488123150907, "language_loss": 0.81817931, "learning_rate": 2.020252916674255e-06, "loss": 0.83999586, "num_input_tokens_seen": 91889075, "step": 4257, "time_per_iteration": 3.3430843353271484 }, { "auxiliary_loss_clip": 0.01169326, "auxiliary_loss_mlp": 0.01031381, "balance_loss_clip": 1.05028892, "balance_loss_mlp": 1.02335238, "epoch": 0.5119942283412493, "flos": 17457329981760.0, "grad_norm": 2.189643567303663, "language_loss": 0.81435061, "learning_rate": 2.019473983451375e-06, "loss": 0.83635765, "num_input_tokens_seen": 91907495, "step": 4258, "time_per_iteration": 2.530698537826538 }, { "auxiliary_loss_clip": 0.01143955, "auxiliary_loss_mlp": 0.01033113, "balance_loss_clip": 1.04810452, "balance_loss_mlp": 1.02528667, "epoch": 0.5121144712318885, "flos": 21066747104640.0, "grad_norm": 1.7450303668897114, "language_loss": 0.71494269, "learning_rate": 2.0186950472743076e-06, "loss": 0.73671341, "num_input_tokens_seen": 91927400, "step": 4259, "time_per_iteration": 2.6229450702667236 }, { "auxiliary_loss_clip": 0.01179473, "auxiliary_loss_mlp": 0.01029154, "balance_loss_clip": 1.05014431, "balance_loss_mlp": 1.0214138, "epoch": 0.5122347141225275, "flos": 19860794568480.0, "grad_norm": 1.5940051514441052, "language_loss": 0.73616928, "learning_rate": 2.0179161082612162e-06, "loss": 0.75825554, "num_input_tokens_seen": 91946790, "step": 4260, "time_per_iteration": 2.5058846473693848 }, { "auxiliary_loss_clip": 0.01147976, "auxiliary_loss_mlp": 0.0102258, "balance_loss_clip": 1.04630542, "balance_loss_mlp": 1.01481354, "epoch": 0.5123549570131666, "flos": 22528491428640.0, "grad_norm": 3.0284066109139425, "language_loss": 0.72595286, "learning_rate": 2.017137166530266e-06, "loss": 0.74765843, "num_input_tokens_seen": 91966325, "step": 4261, "time_per_iteration": 2.5707757472991943 }, { "auxiliary_loss_clip": 0.01156574, "auxiliary_loss_mlp": 0.01029199, "balance_loss_clip": 1.04826891, "balance_loss_mlp": 1.02136111, "epoch": 0.5124751999038056, "flos": 20333378713920.0, "grad_norm": 1.8984480222855653, "language_loss": 0.79807913, "learning_rate": 2.0163582221996213e-06, "loss": 0.81993687, "num_input_tokens_seen": 91984700, "step": 4262, "time_per_iteration": 2.5551154613494873 }, { "auxiliary_loss_clip": 0.01156046, "auxiliary_loss_mlp": 0.01031801, "balance_loss_clip": 1.05128133, "balance_loss_mlp": 1.02387357, "epoch": 0.5125954427944448, "flos": 39785975469600.0, "grad_norm": 1.927183126491326, "language_loss": 0.68035018, "learning_rate": 2.015579275387446e-06, "loss": 0.70222867, "num_input_tokens_seen": 92010020, "step": 4263, "time_per_iteration": 2.707988977432251 }, { "auxiliary_loss_clip": 0.01144611, "auxiliary_loss_mlp": 0.01032351, "balance_loss_clip": 1.04984784, "balance_loss_mlp": 1.02484703, "epoch": 0.5127156856850839, "flos": 29205398655840.0, "grad_norm": 2.008834782707372, "language_loss": 0.68608034, "learning_rate": 2.0148003262119085e-06, "loss": 0.70784992, "num_input_tokens_seen": 92030990, "step": 4264, "time_per_iteration": 2.622547149658203 }, { "auxiliary_loss_clip": 0.01138086, "auxiliary_loss_mlp": 0.01032196, "balance_loss_clip": 1.0486331, "balance_loss_mlp": 1.02390456, "epoch": 0.5128359285757229, "flos": 13553696312160.0, "grad_norm": 1.7700112554102454, "language_loss": 0.76894861, "learning_rate": 2.0140213747911728e-06, "loss": 0.79065144, "num_input_tokens_seen": 92049525, "step": 4265, "time_per_iteration": 2.564765691757202 }, { "auxiliary_loss_clip": 0.01137042, "auxiliary_loss_mlp": 0.01030401, "balance_loss_clip": 1.04889452, "balance_loss_mlp": 1.02249169, "epoch": 0.5129561714663621, "flos": 25192093755360.0, "grad_norm": 8.00393825906871, "language_loss": 0.80407459, "learning_rate": 2.013242421243406e-06, "loss": 0.82574904, "num_input_tokens_seen": 92068430, "step": 4266, "time_per_iteration": 2.6074459552764893 }, { "auxiliary_loss_clip": 0.01129869, "auxiliary_loss_mlp": 0.01027767, "balance_loss_clip": 1.0502826, "balance_loss_mlp": 1.02008653, "epoch": 0.5130764143570011, "flos": 18150226346880.0, "grad_norm": 1.734913045760404, "language_loss": 0.78958935, "learning_rate": 2.012463465686774e-06, "loss": 0.81116569, "num_input_tokens_seen": 92088180, "step": 4267, "time_per_iteration": 2.614579677581787 }, { "auxiliary_loss_clip": 0.01032975, "auxiliary_loss_mlp": 0.01010047, "balance_loss_clip": 1.01909304, "balance_loss_mlp": 1.00871801, "epoch": 0.5131966572476402, "flos": 59794908333120.0, "grad_norm": 0.8440905277024637, "language_loss": 0.54785311, "learning_rate": 2.0116845082394446e-06, "loss": 0.56828332, "num_input_tokens_seen": 92153015, "step": 4268, "time_per_iteration": 3.2100374698638916 }, { "auxiliary_loss_clip": 0.01169391, "auxiliary_loss_mlp": 0.01024685, "balance_loss_clip": 1.04979444, "balance_loss_mlp": 1.01732421, "epoch": 0.5133169001382794, "flos": 18515222445120.0, "grad_norm": 1.8927505528591893, "language_loss": 0.78960693, "learning_rate": 2.0109055490195836e-06, "loss": 0.8115477, "num_input_tokens_seen": 92171470, "step": 4269, "time_per_iteration": 2.5420682430267334 }, { "auxiliary_loss_clip": 0.01116278, "auxiliary_loss_mlp": 0.01029347, "balance_loss_clip": 1.0404129, "balance_loss_mlp": 1.02158654, "epoch": 0.5134371430289184, "flos": 15523540266720.0, "grad_norm": 2.0668461126676863, "language_loss": 0.64304531, "learning_rate": 2.0101265881453605e-06, "loss": 0.66450155, "num_input_tokens_seen": 92189945, "step": 4270, "time_per_iteration": 2.64020037651062 }, { "auxiliary_loss_clip": 0.01149127, "auxiliary_loss_mlp": 0.01031385, "balance_loss_clip": 1.05204201, "balance_loss_mlp": 1.02328777, "epoch": 0.5135573859195575, "flos": 21433790469600.0, "grad_norm": 1.999578425482255, "language_loss": 0.77808607, "learning_rate": 2.009347625734941e-06, "loss": 0.79989117, "num_input_tokens_seen": 92209855, "step": 4271, "time_per_iteration": 2.5711421966552734 }, { "auxiliary_loss_clip": 0.01187258, "auxiliary_loss_mlp": 0.01027652, "balance_loss_clip": 1.05495477, "balance_loss_mlp": 1.01999891, "epoch": 0.5136776288101966, "flos": 17712654849120.0, "grad_norm": 2.5378062363440237, "language_loss": 0.74965131, "learning_rate": 2.0085686619064954e-06, "loss": 0.7718004, "num_input_tokens_seen": 92226295, "step": 4272, "time_per_iteration": 2.480548858642578 }, { "auxiliary_loss_clip": 0.01172144, "auxiliary_loss_mlp": 0.0103208, "balance_loss_clip": 1.05276608, "balance_loss_mlp": 1.02481127, "epoch": 0.5137978717008357, "flos": 16581684148800.0, "grad_norm": 2.2736692053820535, "language_loss": 0.82551664, "learning_rate": 2.00778969677819e-06, "loss": 0.84755886, "num_input_tokens_seen": 92243330, "step": 4273, "time_per_iteration": 2.4979779720306396 }, { "auxiliary_loss_clip": 0.01153027, "auxiliary_loss_mlp": 0.010255, "balance_loss_clip": 1.04905391, "balance_loss_mlp": 1.01838899, "epoch": 0.5139181145914747, "flos": 20668246871040.0, "grad_norm": 1.7290010172556742, "language_loss": 0.63894767, "learning_rate": 2.0070107304681934e-06, "loss": 0.66073292, "num_input_tokens_seen": 92262285, "step": 4274, "time_per_iteration": 2.583979606628418 }, { "auxiliary_loss_clip": 0.01139877, "auxiliary_loss_mlp": 0.01027449, "balance_loss_clip": 1.05024672, "balance_loss_mlp": 1.02044821, "epoch": 0.5140383574821139, "flos": 32927001196800.0, "grad_norm": 1.7105813106770567, "language_loss": 0.78202033, "learning_rate": 2.006231763094675e-06, "loss": 0.80369353, "num_input_tokens_seen": 92283305, "step": 4275, "time_per_iteration": 2.667637348175049 }, { "auxiliary_loss_clip": 0.01150512, "auxiliary_loss_mlp": 0.01024267, "balance_loss_clip": 1.05192018, "balance_loss_mlp": 1.01687646, "epoch": 0.514158600372753, "flos": 19537096585920.0, "grad_norm": 2.569340004538361, "language_loss": 0.87304324, "learning_rate": 2.0054527947758027e-06, "loss": 0.89479101, "num_input_tokens_seen": 92302105, "step": 4276, "time_per_iteration": 3.3294870853424072 }, { "auxiliary_loss_clip": 0.01063989, "auxiliary_loss_mlp": 0.01000137, "balance_loss_clip": 1.01418507, "balance_loss_mlp": 0.99886096, "epoch": 0.514278843263392, "flos": 62523866750400.0, "grad_norm": 0.7303419075994622, "language_loss": 0.55954689, "learning_rate": 2.004673825629746e-06, "loss": 0.58018816, "num_input_tokens_seen": 92362885, "step": 4277, "time_per_iteration": 3.9060966968536377 }, { "auxiliary_loss_clip": 0.01146925, "auxiliary_loss_mlp": 0.01027034, "balance_loss_clip": 1.04666781, "balance_loss_mlp": 1.01918983, "epoch": 0.5143990861540312, "flos": 25882332265440.0, "grad_norm": 1.5181779644111568, "language_loss": 0.72565281, "learning_rate": 2.0038948557746744e-06, "loss": 0.74739236, "num_input_tokens_seen": 92384740, "step": 4278, "time_per_iteration": 2.617494583129883 }, { "auxiliary_loss_clip": 0.01168185, "auxiliary_loss_mlp": 0.01028346, "balance_loss_clip": 1.05390799, "balance_loss_mlp": 1.02117872, "epoch": 0.5145193290446702, "flos": 23330663938080.0, "grad_norm": 1.6995259094468416, "language_loss": 0.75207806, "learning_rate": 2.0031158853287558e-06, "loss": 0.77404332, "num_input_tokens_seen": 92405175, "step": 4279, "time_per_iteration": 3.2967891693115234 }, { "auxiliary_loss_clip": 0.01150079, "auxiliary_loss_mlp": 0.01022927, "balance_loss_clip": 1.04975915, "balance_loss_mlp": 1.01587331, "epoch": 0.5146395719353093, "flos": 22856607197280.0, "grad_norm": 1.9933012049835963, "language_loss": 0.70425403, "learning_rate": 2.0023369144101593e-06, "loss": 0.7259841, "num_input_tokens_seen": 92423345, "step": 4280, "time_per_iteration": 2.5634348392486572 }, { "auxiliary_loss_clip": 0.01142939, "auxiliary_loss_mlp": 0.01023667, "balance_loss_clip": 1.04697299, "balance_loss_mlp": 1.01640439, "epoch": 0.5147598148259485, "flos": 26391581238720.0, "grad_norm": 1.8721299025977325, "language_loss": 0.76633012, "learning_rate": 2.0015579431370555e-06, "loss": 0.78799623, "num_input_tokens_seen": 92445025, "step": 4281, "time_per_iteration": 2.5984301567077637 }, { "auxiliary_loss_clip": 0.01164667, "auxiliary_loss_mlp": 0.01030421, "balance_loss_clip": 1.05088663, "balance_loss_mlp": 1.0230124, "epoch": 0.5148800577165875, "flos": 29965698378240.0, "grad_norm": 2.9635590428875247, "language_loss": 0.70092309, "learning_rate": 2.000778971627612e-06, "loss": 0.72287405, "num_input_tokens_seen": 92464490, "step": 4282, "time_per_iteration": 2.5600059032440186 }, { "auxiliary_loss_clip": 0.01145127, "auxiliary_loss_mlp": 0.01031531, "balance_loss_clip": 1.04557848, "balance_loss_mlp": 1.02381802, "epoch": 0.5150003006072266, "flos": 17931386722560.0, "grad_norm": 1.7470603900309525, "language_loss": 0.90465665, "learning_rate": 2e-06, "loss": 0.92642319, "num_input_tokens_seen": 92482085, "step": 4283, "time_per_iteration": 3.2736246585845947 }, { "auxiliary_loss_clip": 0.01179886, "auxiliary_loss_mlp": 0.01028927, "balance_loss_clip": 1.05261528, "balance_loss_mlp": 1.02133322, "epoch": 0.5151205434978657, "flos": 18478737202080.0, "grad_norm": 1.8355633822093362, "language_loss": 0.85463059, "learning_rate": 1.9992210283723878e-06, "loss": 0.87671864, "num_input_tokens_seen": 92499325, "step": 4284, "time_per_iteration": 2.467397928237915 }, { "auxiliary_loss_clip": 0.01179354, "auxiliary_loss_mlp": 0.0103052, "balance_loss_clip": 1.05222535, "balance_loss_mlp": 1.02303696, "epoch": 0.5152407863885048, "flos": 25341267253920.0, "grad_norm": 1.5621959186210572, "language_loss": 0.79242605, "learning_rate": 1.9984420568629448e-06, "loss": 0.81452477, "num_input_tokens_seen": 92522090, "step": 4285, "time_per_iteration": 2.53652024269104 }, { "auxiliary_loss_clip": 0.01171019, "auxiliary_loss_mlp": 0.01027738, "balance_loss_clip": 1.0522275, "balance_loss_mlp": 1.02049589, "epoch": 0.5153610292791438, "flos": 18329743288320.0, "grad_norm": 1.9080816077569298, "language_loss": 0.78208721, "learning_rate": 1.9976630855898405e-06, "loss": 0.80407476, "num_input_tokens_seen": 92539845, "step": 4286, "time_per_iteration": 2.5022623538970947 }, { "auxiliary_loss_clip": 0.01146242, "auxiliary_loss_mlp": 0.01030802, "balance_loss_clip": 1.04352331, "balance_loss_mlp": 1.0237689, "epoch": 0.515481272169783, "flos": 30409950430560.0, "grad_norm": 2.0591984748676806, "language_loss": 0.7446714, "learning_rate": 1.9968841146712445e-06, "loss": 0.76644182, "num_input_tokens_seen": 92559460, "step": 4287, "time_per_iteration": 2.6065165996551514 }, { "auxiliary_loss_clip": 0.01109081, "auxiliary_loss_mlp": 0.00762245, "balance_loss_clip": 1.0438571, "balance_loss_mlp": 1.00036263, "epoch": 0.5156015150604221, "flos": 23037309398400.0, "grad_norm": 1.6984138350745757, "language_loss": 0.71361017, "learning_rate": 1.996105144225326e-06, "loss": 0.73232347, "num_input_tokens_seen": 92579695, "step": 4288, "time_per_iteration": 2.7044570446014404 }, { "auxiliary_loss_clip": 0.01166529, "auxiliary_loss_mlp": 0.0102671, "balance_loss_clip": 1.05248618, "balance_loss_mlp": 1.01937854, "epoch": 0.5157217579510611, "flos": 17858559904320.0, "grad_norm": 1.7696288906471296, "language_loss": 0.78984892, "learning_rate": 1.995326174370254e-06, "loss": 0.81178129, "num_input_tokens_seen": 92598795, "step": 4289, "time_per_iteration": 2.6809003353118896 }, { "auxiliary_loss_clip": 0.01163934, "auxiliary_loss_mlp": 0.00761729, "balance_loss_clip": 1.04998577, "balance_loss_mlp": 1.00029707, "epoch": 0.5158420008417003, "flos": 19171489899360.0, "grad_norm": 1.6608578322385095, "language_loss": 0.72829723, "learning_rate": 1.994547205224197e-06, "loss": 0.74755388, "num_input_tokens_seen": 92617700, "step": 4290, "time_per_iteration": 2.53420090675354 }, { "auxiliary_loss_clip": 0.0114803, "auxiliary_loss_mlp": 0.01029373, "balance_loss_clip": 1.05007291, "balance_loss_mlp": 1.02170825, "epoch": 0.5159622437323393, "flos": 22419538536960.0, "grad_norm": 1.928529321795158, "language_loss": 0.67650759, "learning_rate": 1.993768236905325e-06, "loss": 0.69828165, "num_input_tokens_seen": 92638370, "step": 4291, "time_per_iteration": 2.529031991958618 }, { "auxiliary_loss_clip": 0.01151392, "auxiliary_loss_mlp": 0.01028537, "balance_loss_clip": 1.04849863, "balance_loss_mlp": 1.02073431, "epoch": 0.5160824866229784, "flos": 24603014156640.0, "grad_norm": 3.632317051427749, "language_loss": 0.65889406, "learning_rate": 1.992989269531807e-06, "loss": 0.68069339, "num_input_tokens_seen": 92657180, "step": 4292, "time_per_iteration": 2.626387119293213 }, { "auxiliary_loss_clip": 0.0115152, "auxiliary_loss_mlp": 0.01028412, "balance_loss_clip": 1.04821551, "balance_loss_mlp": 1.02118182, "epoch": 0.5162027295136175, "flos": 18002740945440.0, "grad_norm": 2.7332672715928035, "language_loss": 0.68023241, "learning_rate": 1.99221030322181e-06, "loss": 0.70203173, "num_input_tokens_seen": 92673985, "step": 4293, "time_per_iteration": 2.5230021476745605 }, { "auxiliary_loss_clip": 0.01158345, "auxiliary_loss_mlp": 0.0103102, "balance_loss_clip": 1.05110109, "balance_loss_mlp": 1.02345586, "epoch": 0.5163229724042566, "flos": 27344615343840.0, "grad_norm": 1.760402976779895, "language_loss": 0.80817127, "learning_rate": 1.991431338093505e-06, "loss": 0.83006489, "num_input_tokens_seen": 92696340, "step": 4294, "time_per_iteration": 2.6287946701049805 }, { "auxiliary_loss_clip": 0.01158522, "auxiliary_loss_mlp": 0.01024377, "balance_loss_clip": 1.0556848, "balance_loss_mlp": 1.017555, "epoch": 0.5164432152948957, "flos": 21762768245280.0, "grad_norm": 1.7284400214623128, "language_loss": 0.79225934, "learning_rate": 1.9906523742650587e-06, "loss": 0.81408834, "num_input_tokens_seen": 92715200, "step": 4295, "time_per_iteration": 2.536449909210205 }, { "auxiliary_loss_clip": 0.01182025, "auxiliary_loss_mlp": 0.01024946, "balance_loss_clip": 1.05089569, "balance_loss_mlp": 1.01757288, "epoch": 0.5165634581855347, "flos": 25550337465120.0, "grad_norm": 1.891193642945937, "language_loss": 0.77530336, "learning_rate": 1.9898734118546397e-06, "loss": 0.79737306, "num_input_tokens_seen": 92735150, "step": 4296, "time_per_iteration": 2.52587628364563 }, { "auxiliary_loss_clip": 0.01102309, "auxiliary_loss_mlp": 0.01028601, "balance_loss_clip": 1.04564655, "balance_loss_mlp": 1.02039313, "epoch": 0.5166837010761739, "flos": 19901194760160.0, "grad_norm": 1.5973807157694226, "language_loss": 0.80298746, "learning_rate": 1.989094450980416e-06, "loss": 0.82429659, "num_input_tokens_seen": 92755250, "step": 4297, "time_per_iteration": 2.7178280353546143 }, { "auxiliary_loss_clip": 0.01167105, "auxiliary_loss_mlp": 0.0102359, "balance_loss_clip": 1.05201542, "balance_loss_mlp": 1.01637149, "epoch": 0.516803943966813, "flos": 26646080016000.0, "grad_norm": 2.1108061081763014, "language_loss": 0.76170522, "learning_rate": 1.9883154917605556e-06, "loss": 0.78361213, "num_input_tokens_seen": 92774460, "step": 4298, "time_per_iteration": 2.573253870010376 }, { "auxiliary_loss_clip": 0.01177503, "auxiliary_loss_mlp": 0.01026598, "balance_loss_clip": 1.05000448, "balance_loss_mlp": 1.01925755, "epoch": 0.516924186857452, "flos": 19682857973280.0, "grad_norm": 1.9247530156425106, "language_loss": 0.83416545, "learning_rate": 1.9875365343132262e-06, "loss": 0.85620654, "num_input_tokens_seen": 92791580, "step": 4299, "time_per_iteration": 2.4652633666992188 }, { "auxiliary_loss_clip": 0.01168949, "auxiliary_loss_mlp": 0.00761895, "balance_loss_clip": 1.053509, "balance_loss_mlp": 1.00037134, "epoch": 0.5170444297480912, "flos": 15956586227520.0, "grad_norm": 4.402812375365344, "language_loss": 0.84980524, "learning_rate": 1.9867575787565946e-06, "loss": 0.86911368, "num_input_tokens_seen": 92806240, "step": 4300, "time_per_iteration": 2.4845516681671143 }, { "auxiliary_loss_clip": 0.01169433, "auxiliary_loss_mlp": 0.01025497, "balance_loss_clip": 1.05305254, "balance_loss_mlp": 1.01786137, "epoch": 0.5171646726387302, "flos": 14174160945600.0, "grad_norm": 8.050482324066607, "language_loss": 0.86100364, "learning_rate": 1.9859786252088275e-06, "loss": 0.88295293, "num_input_tokens_seen": 92823420, "step": 4301, "time_per_iteration": 2.475338935852051 }, { "auxiliary_loss_clip": 0.01142631, "auxiliary_loss_mlp": 0.01026129, "balance_loss_clip": 1.04912829, "balance_loss_mlp": 1.01787329, "epoch": 0.5172849155293693, "flos": 23578553994720.0, "grad_norm": 2.823246445030755, "language_loss": 0.6668824, "learning_rate": 1.9851996737880914e-06, "loss": 0.68857002, "num_input_tokens_seen": 92838605, "step": 4302, "time_per_iteration": 3.3237054347991943 }, { "auxiliary_loss_clip": 0.01175203, "auxiliary_loss_mlp": 0.01029374, "balance_loss_clip": 1.05385494, "balance_loss_mlp": 1.02092767, "epoch": 0.5174051584200084, "flos": 14283544840800.0, "grad_norm": 1.8424624682288173, "language_loss": 0.74883783, "learning_rate": 1.9844207246125537e-06, "loss": 0.77088362, "num_input_tokens_seen": 92855185, "step": 4303, "time_per_iteration": 3.2509493827819824 }, { "auxiliary_loss_clip": 0.01150493, "auxiliary_loss_mlp": 0.01031092, "balance_loss_clip": 1.04884434, "balance_loss_mlp": 1.02407646, "epoch": 0.5175254013106475, "flos": 37889389336800.0, "grad_norm": 1.700411068155176, "language_loss": 0.68417877, "learning_rate": 1.983641777800379e-06, "loss": 0.70599461, "num_input_tokens_seen": 92877830, "step": 4304, "time_per_iteration": 2.686309337615967 }, { "auxiliary_loss_clip": 0.01057458, "auxiliary_loss_mlp": 0.01001268, "balance_loss_clip": 1.01349974, "balance_loss_mlp": 0.99995041, "epoch": 0.5176456442012866, "flos": 68549750399520.0, "grad_norm": 0.7413487372527123, "language_loss": 0.58818018, "learning_rate": 1.9828628334697343e-06, "loss": 0.60876745, "num_input_tokens_seen": 92945040, "step": 4305, "time_per_iteration": 3.9933457374572754 }, { "auxiliary_loss_clip": 0.01059743, "auxiliary_loss_mlp": 0.01004257, "balance_loss_clip": 1.01515794, "balance_loss_mlp": 1.0029397, "epoch": 0.5177658870919257, "flos": 64084148047680.0, "grad_norm": 0.7686340712849001, "language_loss": 0.54718125, "learning_rate": 1.982083891738784e-06, "loss": 0.56782126, "num_input_tokens_seen": 93005910, "step": 4306, "time_per_iteration": 3.149439811706543 }, { "auxiliary_loss_clip": 0.01149755, "auxiliary_loss_mlp": 0.01026632, "balance_loss_clip": 1.05291224, "balance_loss_mlp": 1.01950884, "epoch": 0.5178861299825648, "flos": 26651252058240.0, "grad_norm": 1.4865942429204844, "language_loss": 0.82680678, "learning_rate": 1.9813049527256923e-06, "loss": 0.8485707, "num_input_tokens_seen": 93026305, "step": 4307, "time_per_iteration": 2.5993480682373047 }, { "auxiliary_loss_clip": 0.0113804, "auxiliary_loss_mlp": 0.01028265, "balance_loss_clip": 1.04710555, "balance_loss_mlp": 1.02065396, "epoch": 0.5180063728732038, "flos": 17931889560000.0, "grad_norm": 2.092448108888957, "language_loss": 0.81429005, "learning_rate": 1.9805260165486252e-06, "loss": 0.83595306, "num_input_tokens_seen": 93045675, "step": 4308, "time_per_iteration": 2.562340021133423 }, { "auxiliary_loss_clip": 0.01166555, "auxiliary_loss_mlp": 0.01027873, "balance_loss_clip": 1.05224705, "balance_loss_mlp": 1.02103281, "epoch": 0.518126615763843, "flos": 19500898678560.0, "grad_norm": 1.7773831408378273, "language_loss": 0.86548686, "learning_rate": 1.9797470833257457e-06, "loss": 0.88743114, "num_input_tokens_seen": 93065375, "step": 4309, "time_per_iteration": 3.4610531330108643 }, { "auxiliary_loss_clip": 0.0117366, "auxiliary_loss_mlp": 0.01031292, "balance_loss_clip": 1.05663419, "balance_loss_mlp": 1.02306974, "epoch": 0.5182468586544821, "flos": 20704085608800.0, "grad_norm": 2.152844176742528, "language_loss": 0.76881254, "learning_rate": 1.9789681531752177e-06, "loss": 0.79086208, "num_input_tokens_seen": 93085595, "step": 4310, "time_per_iteration": 2.551381826400757 }, { "auxiliary_loss_clip": 0.01121853, "auxiliary_loss_mlp": 0.01025004, "balance_loss_clip": 1.04856253, "balance_loss_mlp": 1.01804793, "epoch": 0.5183671015451211, "flos": 23112111649440.0, "grad_norm": 1.730858671885893, "language_loss": 0.72813576, "learning_rate": 1.978189226215204e-06, "loss": 0.74960434, "num_input_tokens_seen": 93106140, "step": 4311, "time_per_iteration": 2.6316781044006348 }, { "auxiliary_loss_clip": 0.01179459, "auxiliary_loss_mlp": 0.010258, "balance_loss_clip": 1.05070639, "balance_loss_mlp": 1.01787829, "epoch": 0.5184873444357603, "flos": 17597093236800.0, "grad_norm": 2.0121868768614917, "language_loss": 0.77025747, "learning_rate": 1.9774103025638675e-06, "loss": 0.79231, "num_input_tokens_seen": 93124265, "step": 4312, "time_per_iteration": 2.4795100688934326 }, { "auxiliary_loss_clip": 0.01132121, "auxiliary_loss_mlp": 0.01028608, "balance_loss_clip": 1.05346727, "balance_loss_mlp": 1.02025747, "epoch": 0.5186075873263993, "flos": 24936804804960.0, "grad_norm": 1.449008013557843, "language_loss": 0.76481533, "learning_rate": 1.9766313823393696e-06, "loss": 0.78642261, "num_input_tokens_seen": 93145130, "step": 4313, "time_per_iteration": 2.646343946456909 }, { "auxiliary_loss_clip": 0.01117301, "auxiliary_loss_mlp": 0.01024944, "balance_loss_clip": 1.04236507, "balance_loss_mlp": 1.01732671, "epoch": 0.5187278302170384, "flos": 15190108788000.0, "grad_norm": 2.5316828050232436, "language_loss": 0.69339895, "learning_rate": 1.975852465659873e-06, "loss": 0.7148214, "num_input_tokens_seen": 93161110, "step": 4314, "time_per_iteration": 2.6124422550201416 }, { "auxiliary_loss_clip": 0.01170204, "auxiliary_loss_mlp": 0.01025079, "balance_loss_clip": 1.05286753, "balance_loss_mlp": 1.01695228, "epoch": 0.5188480731076776, "flos": 25009416121440.0, "grad_norm": 2.459095808132988, "language_loss": 0.69812715, "learning_rate": 1.9750735526435377e-06, "loss": 0.72008002, "num_input_tokens_seen": 93178055, "step": 4315, "time_per_iteration": 2.5361380577087402 }, { "auxiliary_loss_clip": 0.01152108, "auxiliary_loss_mlp": 0.01028614, "balance_loss_clip": 1.05119562, "balance_loss_mlp": 1.02064538, "epoch": 0.5189683159983166, "flos": 24790145493600.0, "grad_norm": 2.464762059785151, "language_loss": 0.79473656, "learning_rate": 1.974294643408525e-06, "loss": 0.81654376, "num_input_tokens_seen": 93195850, "step": 4316, "time_per_iteration": 2.582254648208618 }, { "auxiliary_loss_clip": 0.0117271, "auxiliary_loss_mlp": 0.01029124, "balance_loss_clip": 1.05177522, "balance_loss_mlp": 1.02135801, "epoch": 0.5190885588889557, "flos": 24754270838880.0, "grad_norm": 1.8112922333072472, "language_loss": 0.66841817, "learning_rate": 1.9735157380729947e-06, "loss": 0.69043648, "num_input_tokens_seen": 93216260, "step": 4317, "time_per_iteration": 2.5762274265289307 }, { "auxiliary_loss_clip": 0.01154699, "auxiliary_loss_mlp": 0.01026934, "balance_loss_clip": 1.04968488, "balance_loss_mlp": 1.0190959, "epoch": 0.5192088017795948, "flos": 24712649470560.0, "grad_norm": 1.7428413582748108, "language_loss": 0.84314603, "learning_rate": 1.9727368367551053e-06, "loss": 0.86496234, "num_input_tokens_seen": 93234810, "step": 4318, "time_per_iteration": 2.590475559234619 }, { "auxiliary_loss_clip": 0.01142707, "auxiliary_loss_mlp": 0.01028127, "balance_loss_clip": 1.04809618, "balance_loss_mlp": 1.02065301, "epoch": 0.5193290446702339, "flos": 27229592485920.0, "grad_norm": 2.1218593285500678, "language_loss": 0.68338513, "learning_rate": 1.9719579395730164e-06, "loss": 0.7050935, "num_input_tokens_seen": 93254185, "step": 4319, "time_per_iteration": 2.6093714237213135 }, { "auxiliary_loss_clip": 0.01182962, "auxiliary_loss_mlp": 0.01029809, "balance_loss_clip": 1.05454028, "balance_loss_mlp": 1.02230728, "epoch": 0.5194492875608729, "flos": 11473355036640.0, "grad_norm": 2.082816970538392, "language_loss": 0.93587148, "learning_rate": 1.9711790466448854e-06, "loss": 0.95799923, "num_input_tokens_seen": 93268205, "step": 4320, "time_per_iteration": 2.4482600688934326 }, { "auxiliary_loss_clip": 0.01131546, "auxiliary_loss_mlp": 0.01033874, "balance_loss_clip": 1.04900038, "balance_loss_mlp": 1.02589869, "epoch": 0.5195695304515121, "flos": 20338909925760.0, "grad_norm": 2.280337273130573, "language_loss": 0.71308422, "learning_rate": 1.9704001580888704e-06, "loss": 0.73473841, "num_input_tokens_seen": 93286945, "step": 4321, "time_per_iteration": 2.6330838203430176 }, { "auxiliary_loss_clip": 0.01148432, "auxiliary_loss_mlp": 0.00762401, "balance_loss_clip": 1.04813898, "balance_loss_mlp": 1.0003866, "epoch": 0.5196897733421512, "flos": 20048320992000.0, "grad_norm": 1.764018121089085, "language_loss": 0.86609864, "learning_rate": 1.9696212740231283e-06, "loss": 0.88520694, "num_input_tokens_seen": 93305595, "step": 4322, "time_per_iteration": 2.529374122619629 }, { "auxiliary_loss_clip": 0.01172725, "auxiliary_loss_mlp": 0.01035728, "balance_loss_clip": 1.04996276, "balance_loss_mlp": 1.02763987, "epoch": 0.5198100162327902, "flos": 23805510852000.0, "grad_norm": 2.3867800298154065, "language_loss": 0.81873965, "learning_rate": 1.9688423945658146e-06, "loss": 0.84082413, "num_input_tokens_seen": 93326460, "step": 4323, "time_per_iteration": 2.5528557300567627 }, { "auxiliary_loss_clip": 0.01114096, "auxiliary_loss_mlp": 0.01028004, "balance_loss_clip": 1.04053736, "balance_loss_mlp": 1.01977813, "epoch": 0.5199302591234293, "flos": 24023955389760.0, "grad_norm": 1.8364964158532469, "language_loss": 0.71646571, "learning_rate": 1.9680635198350845e-06, "loss": 0.73788667, "num_input_tokens_seen": 93346170, "step": 4324, "time_per_iteration": 2.614895820617676 }, { "auxiliary_loss_clip": 0.01168489, "auxiliary_loss_mlp": 0.01032715, "balance_loss_clip": 1.05009949, "balance_loss_mlp": 1.02431083, "epoch": 0.5200505020140684, "flos": 26359370113920.0, "grad_norm": 2.689628976281599, "language_loss": 0.73242307, "learning_rate": 1.967284649949093e-06, "loss": 0.75443512, "num_input_tokens_seen": 93365380, "step": 4325, "time_per_iteration": 2.5670487880706787 }, { "auxiliary_loss_clip": 0.01136372, "auxiliary_loss_mlp": 0.01027675, "balance_loss_clip": 1.04613483, "balance_loss_mlp": 1.0200572, "epoch": 0.5201707449047075, "flos": 39604267593600.0, "grad_norm": 1.9130511791519462, "language_loss": 0.71988118, "learning_rate": 1.966505785025994e-06, "loss": 0.74152166, "num_input_tokens_seen": 93387285, "step": 4326, "time_per_iteration": 2.723332643508911 }, { "auxiliary_loss_clip": 0.01136234, "auxiliary_loss_mlp": 0.01031301, "balance_loss_clip": 1.04948628, "balance_loss_mlp": 1.02280116, "epoch": 0.5202909877953465, "flos": 53682794451360.0, "grad_norm": 1.7825008099235473, "language_loss": 0.75933617, "learning_rate": 1.965726925183941e-06, "loss": 0.78101152, "num_input_tokens_seen": 93410390, "step": 4327, "time_per_iteration": 2.8526856899261475 }, { "auxiliary_loss_clip": 0.01179247, "auxiliary_loss_mlp": 0.01023966, "balance_loss_clip": 1.05242705, "balance_loss_mlp": 1.01738298, "epoch": 0.5204112306859857, "flos": 19537024752000.0, "grad_norm": 1.7807816055674144, "language_loss": 0.84506893, "learning_rate": 1.964948070541087e-06, "loss": 0.86710107, "num_input_tokens_seen": 93429050, "step": 4328, "time_per_iteration": 4.085917949676514 }, { "auxiliary_loss_clip": 0.01156336, "auxiliary_loss_mlp": 0.01030422, "balance_loss_clip": 1.04612613, "balance_loss_mlp": 1.02313864, "epoch": 0.5205314735766248, "flos": 15304700642400.0, "grad_norm": 2.404954763156062, "language_loss": 0.69408822, "learning_rate": 1.9641692212155816e-06, "loss": 0.71595579, "num_input_tokens_seen": 93446815, "step": 4329, "time_per_iteration": 2.534522294998169 }, { "auxiliary_loss_clip": 0.01123522, "auxiliary_loss_mlp": 0.01033847, "balance_loss_clip": 1.0497334, "balance_loss_mlp": 1.02526999, "epoch": 0.5206517164672638, "flos": 59263707708960.0, "grad_norm": 1.8333433596630122, "language_loss": 0.72866011, "learning_rate": 1.9633903773255777e-06, "loss": 0.75023377, "num_input_tokens_seen": 93469130, "step": 4330, "time_per_iteration": 2.9390010833740234 }, { "auxiliary_loss_clip": 0.01178161, "auxiliary_loss_mlp": 0.01028687, "balance_loss_clip": 1.05018806, "balance_loss_mlp": 1.02124834, "epoch": 0.520771959357903, "flos": 26871133274400.0, "grad_norm": 1.7550552416264054, "language_loss": 0.7432245, "learning_rate": 1.9626115389892237e-06, "loss": 0.76529288, "num_input_tokens_seen": 93489920, "step": 4331, "time_per_iteration": 3.30470871925354 }, { "auxiliary_loss_clip": 0.01145344, "auxiliary_loss_mlp": 0.01032686, "balance_loss_clip": 1.05138779, "balance_loss_mlp": 1.02532458, "epoch": 0.520892202248542, "flos": 26907079763040.0, "grad_norm": 1.9899159110118643, "language_loss": 0.85384941, "learning_rate": 1.96183270632467e-06, "loss": 0.8756296, "num_input_tokens_seen": 93509770, "step": 4332, "time_per_iteration": 2.638207197189331 }, { "auxiliary_loss_clip": 0.01130881, "auxiliary_loss_mlp": 0.00762677, "balance_loss_clip": 1.04605389, "balance_loss_mlp": 1.00039482, "epoch": 0.5210124451391811, "flos": 25849438718400.0, "grad_norm": 13.400508270272685, "language_loss": 0.78783613, "learning_rate": 1.9610538794500644e-06, "loss": 0.80677164, "num_input_tokens_seen": 93529320, "step": 4333, "time_per_iteration": 2.602910280227661 }, { "auxiliary_loss_clip": 0.01046276, "auxiliary_loss_mlp": 0.01003062, "balance_loss_clip": 1.01392341, "balance_loss_mlp": 1.00183392, "epoch": 0.5211326880298203, "flos": 70553816828640.0, "grad_norm": 1.042255020320784, "language_loss": 0.59481245, "learning_rate": 1.9602750584835542e-06, "loss": 0.61530584, "num_input_tokens_seen": 93595255, "step": 4334, "time_per_iteration": 3.301297903060913 }, { "auxiliary_loss_clip": 0.01148514, "auxiliary_loss_mlp": 0.01028828, "balance_loss_clip": 1.04601276, "balance_loss_mlp": 1.02112126, "epoch": 0.5212529309204593, "flos": 15628901462400.0, "grad_norm": 2.1627189352108664, "language_loss": 0.82772064, "learning_rate": 1.959496243543286e-06, "loss": 0.84949398, "num_input_tokens_seen": 93613135, "step": 4335, "time_per_iteration": 3.264582395553589 }, { "auxiliary_loss_clip": 0.01172377, "auxiliary_loss_mlp": 0.01035829, "balance_loss_clip": 1.05567598, "balance_loss_mlp": 1.02787805, "epoch": 0.5213731738110984, "flos": 26242659158880.0, "grad_norm": 1.927305561014411, "language_loss": 0.79447913, "learning_rate": 1.9587174347474057e-06, "loss": 0.81656122, "num_input_tokens_seen": 93629645, "step": 4336, "time_per_iteration": 2.5260794162750244 }, { "auxiliary_loss_clip": 0.01110745, "auxiliary_loss_mlp": 0.01032114, "balance_loss_clip": 1.04296613, "balance_loss_mlp": 1.02384055, "epoch": 0.5214934167017375, "flos": 19418410198080.0, "grad_norm": 2.4737378909257064, "language_loss": 0.81884533, "learning_rate": 1.9579386322140574e-06, "loss": 0.84027398, "num_input_tokens_seen": 93645325, "step": 4337, "time_per_iteration": 2.5793254375457764 }, { "auxiliary_loss_clip": 0.01181746, "auxiliary_loss_mlp": 0.00762848, "balance_loss_clip": 1.05177855, "balance_loss_mlp": 1.0004313, "epoch": 0.5216136595923766, "flos": 30955792397760.0, "grad_norm": 1.8734303114909736, "language_loss": 0.8063221, "learning_rate": 1.9571598360613854e-06, "loss": 0.82576805, "num_input_tokens_seen": 93668200, "step": 4338, "time_per_iteration": 2.5575098991394043 }, { "auxiliary_loss_clip": 0.01138415, "auxiliary_loss_mlp": 0.01025474, "balance_loss_clip": 1.04498017, "balance_loss_mlp": 1.01795745, "epoch": 0.5217339024830157, "flos": 21945050792640.0, "grad_norm": 2.1054396042899555, "language_loss": 0.69978976, "learning_rate": 1.956381046407532e-06, "loss": 0.72142863, "num_input_tokens_seen": 93688495, "step": 4339, "time_per_iteration": 2.5283966064453125 }, { "auxiliary_loss_clip": 0.01137764, "auxiliary_loss_mlp": 0.0103136, "balance_loss_clip": 1.04825544, "balance_loss_mlp": 1.02393961, "epoch": 0.5218541453736548, "flos": 20923212568800.0, "grad_norm": 1.759193578018192, "language_loss": 0.86133569, "learning_rate": 1.9556022633706394e-06, "loss": 0.88302696, "num_input_tokens_seen": 93707285, "step": 4340, "time_per_iteration": 2.5818004608154297 }, { "auxiliary_loss_clip": 0.01144887, "auxiliary_loss_mlp": 0.01024888, "balance_loss_clip": 1.04813147, "balance_loss_mlp": 1.01701987, "epoch": 0.5219743882642939, "flos": 23951667325920.0, "grad_norm": 1.5710724898792854, "language_loss": 0.80147606, "learning_rate": 1.954823487068848e-06, "loss": 0.82317376, "num_input_tokens_seen": 93727495, "step": 4341, "time_per_iteration": 2.5698561668395996 }, { "auxiliary_loss_clip": 0.01165288, "auxiliary_loss_mlp": 0.01028332, "balance_loss_clip": 1.05184913, "balance_loss_mlp": 1.02095926, "epoch": 0.5220946311549329, "flos": 28799284026720.0, "grad_norm": 1.6638654591045896, "language_loss": 0.81317961, "learning_rate": 1.9540447176202976e-06, "loss": 0.83511579, "num_input_tokens_seen": 93748740, "step": 4342, "time_per_iteration": 2.6036038398742676 }, { "auxiliary_loss_clip": 0.01064738, "auxiliary_loss_mlp": 0.01001177, "balance_loss_clip": 1.01564109, "balance_loss_mlp": 0.999901, "epoch": 0.5222148740455721, "flos": 67189380488640.0, "grad_norm": 0.8746603350419844, "language_loss": 0.60736388, "learning_rate": 1.9532659551431272e-06, "loss": 0.62802309, "num_input_tokens_seen": 93815770, "step": 4343, "time_per_iteration": 3.2807562351226807 }, { "auxiliary_loss_clip": 0.01169163, "auxiliary_loss_mlp": 0.01028198, "balance_loss_clip": 1.05036044, "balance_loss_mlp": 1.02046728, "epoch": 0.5223351169362112, "flos": 61856171314560.0, "grad_norm": 1.612467490306125, "language_loss": 0.67254663, "learning_rate": 1.9524871997554744e-06, "loss": 0.69452024, "num_input_tokens_seen": 93843530, "step": 4344, "time_per_iteration": 2.875688076019287 }, { "auxiliary_loss_clip": 0.01167154, "auxiliary_loss_mlp": 0.0102549, "balance_loss_clip": 1.05033875, "balance_loss_mlp": 1.01839972, "epoch": 0.5224553598268502, "flos": 14647391596320.0, "grad_norm": 2.026520096973405, "language_loss": 0.8045435, "learning_rate": 1.951708451575475e-06, "loss": 0.82646996, "num_input_tokens_seen": 93860595, "step": 4345, "time_per_iteration": 2.4841549396514893 }, { "auxiliary_loss_clip": 0.01146306, "auxiliary_loss_mlp": 0.01021294, "balance_loss_clip": 1.04540873, "balance_loss_mlp": 1.01423728, "epoch": 0.5225756027174894, "flos": 14826046530720.0, "grad_norm": 2.3662538282136616, "language_loss": 0.82406056, "learning_rate": 1.9509297107212657e-06, "loss": 0.84573656, "num_input_tokens_seen": 93877365, "step": 4346, "time_per_iteration": 2.564983606338501 }, { "auxiliary_loss_clip": 0.01181243, "auxiliary_loss_mlp": 0.01028418, "balance_loss_clip": 1.05218863, "balance_loss_mlp": 1.02100861, "epoch": 0.5226958456081284, "flos": 23512012644480.0, "grad_norm": 1.5827213332650931, "language_loss": 0.78970307, "learning_rate": 1.95015097731098e-06, "loss": 0.81179971, "num_input_tokens_seen": 93896855, "step": 4347, "time_per_iteration": 2.5175647735595703 }, { "auxiliary_loss_clip": 0.01181093, "auxiliary_loss_mlp": 0.01030298, "balance_loss_clip": 1.05230188, "balance_loss_mlp": 1.02262378, "epoch": 0.5228160884987675, "flos": 19062932094240.0, "grad_norm": 1.9681112948893074, "language_loss": 0.81574214, "learning_rate": 1.949372251462751e-06, "loss": 0.83785599, "num_input_tokens_seen": 93914270, "step": 4348, "time_per_iteration": 2.45896053314209 }, { "auxiliary_loss_clip": 0.01140246, "auxiliary_loss_mlp": 0.00762652, "balance_loss_clip": 1.04947138, "balance_loss_mlp": 1.00041389, "epoch": 0.5229363313894067, "flos": 21063227242560.0, "grad_norm": 1.825665114251146, "language_loss": 0.8294192, "learning_rate": 1.9485935332947124e-06, "loss": 0.84844822, "num_input_tokens_seen": 93932180, "step": 4349, "time_per_iteration": 2.605397939682007 }, { "auxiliary_loss_clip": 0.01148351, "auxiliary_loss_mlp": 0.01024523, "balance_loss_clip": 1.04904413, "balance_loss_mlp": 1.01686358, "epoch": 0.5230565742800457, "flos": 14830392482880.0, "grad_norm": 2.0957970642051293, "language_loss": 0.82975453, "learning_rate": 1.947814822924993e-06, "loss": 0.85148323, "num_input_tokens_seen": 93949690, "step": 4350, "time_per_iteration": 2.4984629154205322 }, { "auxiliary_loss_clip": 0.01180355, "auxiliary_loss_mlp": 0.0103121, "balance_loss_clip": 1.0534451, "balance_loss_mlp": 1.02363372, "epoch": 0.5231768171706848, "flos": 25813025309280.0, "grad_norm": 1.7921617396078087, "language_loss": 0.83151448, "learning_rate": 1.9470361204717236e-06, "loss": 0.85363024, "num_input_tokens_seen": 93968830, "step": 4351, "time_per_iteration": 2.5274040699005127 }, { "auxiliary_loss_clip": 0.01140704, "auxiliary_loss_mlp": 0.00762655, "balance_loss_clip": 1.04739678, "balance_loss_mlp": 1.00039887, "epoch": 0.5232970600613239, "flos": 22743811190880.0, "grad_norm": 1.5965777141250963, "language_loss": 0.8058483, "learning_rate": 1.9462574260530326e-06, "loss": 0.82488191, "num_input_tokens_seen": 93989110, "step": 4352, "time_per_iteration": 2.584390878677368 }, { "auxiliary_loss_clip": 0.01155802, "auxiliary_loss_mlp": 0.0102696, "balance_loss_clip": 1.04713202, "balance_loss_mlp": 1.01933098, "epoch": 0.523417302951963, "flos": 17310706587360.0, "grad_norm": 1.6585666491280384, "language_loss": 0.80773449, "learning_rate": 1.9454787397870472e-06, "loss": 0.82956207, "num_input_tokens_seen": 94006430, "step": 4353, "time_per_iteration": 2.5026144981384277 }, { "auxiliary_loss_clip": 0.01100969, "auxiliary_loss_mlp": 0.010259, "balance_loss_clip": 1.04563224, "balance_loss_mlp": 1.01822913, "epoch": 0.523537545842602, "flos": 18551743605120.0, "grad_norm": 1.8533753148605197, "language_loss": 0.71756136, "learning_rate": 1.944700061791894e-06, "loss": 0.73883009, "num_input_tokens_seen": 94024825, "step": 4354, "time_per_iteration": 3.4319164752960205 }, { "auxiliary_loss_clip": 0.01164514, "auxiliary_loss_mlp": 0.01031332, "balance_loss_clip": 1.05092192, "balance_loss_mlp": 1.02384615, "epoch": 0.5236577887332411, "flos": 19719271382400.0, "grad_norm": 2.9374759215858073, "language_loss": 0.65505803, "learning_rate": 1.943921392185698e-06, "loss": 0.67701644, "num_input_tokens_seen": 94043450, "step": 4355, "time_per_iteration": 2.4950029850006104 }, { "auxiliary_loss_clip": 0.0115361, "auxiliary_loss_mlp": 0.01023773, "balance_loss_clip": 1.04916191, "balance_loss_mlp": 1.01654625, "epoch": 0.5237780316238803, "flos": 23550221901600.0, "grad_norm": 1.894947635656332, "language_loss": 0.76793718, "learning_rate": 1.9431427310865814e-06, "loss": 0.78971094, "num_input_tokens_seen": 94063055, "step": 4356, "time_per_iteration": 2.5821406841278076 }, { "auxiliary_loss_clip": 0.0111814, "auxiliary_loss_mlp": 0.01029636, "balance_loss_clip": 1.04359293, "balance_loss_mlp": 1.02190804, "epoch": 0.5238982745145193, "flos": 22491898434720.0, "grad_norm": 1.8241290737544524, "language_loss": 0.78194398, "learning_rate": 1.942364078612667e-06, "loss": 0.80342174, "num_input_tokens_seen": 94081785, "step": 4357, "time_per_iteration": 2.5512640476226807 }, { "auxiliary_loss_clip": 0.0114055, "auxiliary_loss_mlp": 0.01024719, "balance_loss_clip": 1.04631686, "balance_loss_mlp": 1.01700616, "epoch": 0.5240185174051584, "flos": 27088931306880.0, "grad_norm": 1.942367988550633, "language_loss": 0.75419885, "learning_rate": 1.9415854348820765e-06, "loss": 0.77585149, "num_input_tokens_seen": 94101635, "step": 4358, "time_per_iteration": 3.337209939956665 }, { "auxiliary_loss_clip": 0.0116858, "auxiliary_loss_mlp": 0.01025185, "balance_loss_clip": 1.04995036, "balance_loss_mlp": 1.01735878, "epoch": 0.5241387602957975, "flos": 22674683819520.0, "grad_norm": 4.4565234059039085, "language_loss": 0.68632448, "learning_rate": 1.940806800012929e-06, "loss": 0.70826215, "num_input_tokens_seen": 94121705, "step": 4359, "time_per_iteration": 2.4932973384857178 }, { "auxiliary_loss_clip": 0.01121267, "auxiliary_loss_mlp": 0.00763385, "balance_loss_clip": 1.0467335, "balance_loss_mlp": 1.00041533, "epoch": 0.5242590031864366, "flos": 40553458584000.0, "grad_norm": 1.524441598559505, "language_loss": 0.63529474, "learning_rate": 1.9400281741233432e-06, "loss": 0.65414125, "num_input_tokens_seen": 94146595, "step": 4360, "time_per_iteration": 2.7648115158081055 }, { "auxiliary_loss_clip": 0.0104226, "auxiliary_loss_mlp": 0.01003802, "balance_loss_clip": 1.01814198, "balance_loss_mlp": 1.00254464, "epoch": 0.5243792460770756, "flos": 66676324317600.0, "grad_norm": 0.6673643675278815, "language_loss": 0.52611959, "learning_rate": 1.939249557331435e-06, "loss": 0.5465802, "num_input_tokens_seen": 94212410, "step": 4361, "time_per_iteration": 4.08796763420105 }, { "auxiliary_loss_clip": 0.01147368, "auxiliary_loss_mlp": 0.01032295, "balance_loss_clip": 1.04889917, "balance_loss_mlp": 1.02498198, "epoch": 0.5244994889677148, "flos": 28183668182880.0, "grad_norm": 1.7361323250800624, "language_loss": 0.72854805, "learning_rate": 1.938470949755321e-06, "loss": 0.75034469, "num_input_tokens_seen": 94232290, "step": 4362, "time_per_iteration": 2.608452558517456 }, { "auxiliary_loss_clip": 0.01046106, "auxiliary_loss_mlp": 0.01009166, "balance_loss_clip": 1.01487517, "balance_loss_mlp": 1.0077889, "epoch": 0.5246197318583539, "flos": 65950390737600.0, "grad_norm": 0.8143372606020913, "language_loss": 0.55681795, "learning_rate": 1.937692351513115e-06, "loss": 0.57737064, "num_input_tokens_seen": 94291285, "step": 4363, "time_per_iteration": 3.108673334121704 }, { "auxiliary_loss_clip": 0.01172595, "auxiliary_loss_mlp": 0.01026972, "balance_loss_clip": 1.05268121, "balance_loss_mlp": 1.01955676, "epoch": 0.5247399747489929, "flos": 21033494388000.0, "grad_norm": 1.9334861869505877, "language_loss": 0.80626965, "learning_rate": 1.9369137627229297e-06, "loss": 0.82826531, "num_input_tokens_seen": 94309685, "step": 4364, "time_per_iteration": 2.5099470615386963 }, { "auxiliary_loss_clip": 0.01165532, "auxiliary_loss_mlp": 0.01029252, "balance_loss_clip": 1.05115867, "balance_loss_mlp": 1.02183771, "epoch": 0.5248602176396321, "flos": 19025944013760.0, "grad_norm": 4.354513202404929, "language_loss": 0.8827337, "learning_rate": 1.936135183502877e-06, "loss": 0.90468156, "num_input_tokens_seen": 94326985, "step": 4365, "time_per_iteration": 2.4985086917877197 }, { "auxiliary_loss_clip": 0.01139916, "auxiliary_loss_mlp": 0.0102554, "balance_loss_clip": 1.04719615, "balance_loss_mlp": 1.0178864, "epoch": 0.5249804605302711, "flos": 22200088324320.0, "grad_norm": 3.097777315804127, "language_loss": 0.8049221, "learning_rate": 1.935356613971066e-06, "loss": 0.82657665, "num_input_tokens_seen": 94347645, "step": 4366, "time_per_iteration": 2.5815346240997314 }, { "auxiliary_loss_clip": 0.01152291, "auxiliary_loss_mlp": 0.00762871, "balance_loss_clip": 1.05047441, "balance_loss_mlp": 1.00027847, "epoch": 0.5251007034209102, "flos": 23805690436800.0, "grad_norm": 1.7451214356947342, "language_loss": 0.76749545, "learning_rate": 1.9345780542456047e-06, "loss": 0.78664708, "num_input_tokens_seen": 94367020, "step": 4367, "time_per_iteration": 2.566049575805664 }, { "auxiliary_loss_clip": 0.01154308, "auxiliary_loss_mlp": 0.01030049, "balance_loss_clip": 1.04676652, "balance_loss_mlp": 1.02266407, "epoch": 0.5252209463115494, "flos": 23294609698560.0, "grad_norm": 2.539368467900155, "language_loss": 0.71612817, "learning_rate": 1.9337995044446007e-06, "loss": 0.73797172, "num_input_tokens_seen": 94385860, "step": 4368, "time_per_iteration": 2.5064399242401123 }, { "auxiliary_loss_clip": 0.01172477, "auxiliary_loss_mlp": 0.01029733, "balance_loss_clip": 1.05195975, "balance_loss_mlp": 1.02208567, "epoch": 0.5253411892021884, "flos": 19828691194560.0, "grad_norm": 2.119487344962156, "language_loss": 0.79590958, "learning_rate": 1.9330209646861596e-06, "loss": 0.81793165, "num_input_tokens_seen": 94405010, "step": 4369, "time_per_iteration": 2.5012383460998535 }, { "auxiliary_loss_clip": 0.01149447, "auxiliary_loss_mlp": 0.01021854, "balance_loss_clip": 1.04913247, "balance_loss_mlp": 1.01430821, "epoch": 0.5254614320928275, "flos": 24133734371520.0, "grad_norm": 1.5492319927620837, "language_loss": 0.77588511, "learning_rate": 1.9322424350883843e-06, "loss": 0.79759812, "num_input_tokens_seen": 94426845, "step": 4370, "time_per_iteration": 2.543083667755127 }, { "auxiliary_loss_clip": 0.01154701, "auxiliary_loss_mlp": 0.01030154, "balance_loss_clip": 1.04966474, "balance_loss_mlp": 1.02295935, "epoch": 0.5255816749834666, "flos": 24644958777600.0, "grad_norm": 1.5835690852099031, "language_loss": 0.7904796, "learning_rate": 1.931463915769379e-06, "loss": 0.8123281, "num_input_tokens_seen": 94446960, "step": 4371, "time_per_iteration": 2.5909414291381836 }, { "auxiliary_loss_clip": 0.01119151, "auxiliary_loss_mlp": 0.0102459, "balance_loss_clip": 1.04339218, "balance_loss_mlp": 1.01697254, "epoch": 0.5257019178741057, "flos": 14136598193760.0, "grad_norm": 2.065794894621055, "language_loss": 0.73973966, "learning_rate": 1.930685406847242e-06, "loss": 0.761177, "num_input_tokens_seen": 94461535, "step": 4372, "time_per_iteration": 2.576129198074341 }, { "auxiliary_loss_clip": 0.01151597, "auxiliary_loss_mlp": 0.01030679, "balance_loss_clip": 1.05069458, "balance_loss_mlp": 1.02356195, "epoch": 0.5258221607647448, "flos": 23548964808000.0, "grad_norm": 1.6180321551509378, "language_loss": 0.81690347, "learning_rate": 1.9299069084400734e-06, "loss": 0.83872616, "num_input_tokens_seen": 94482395, "step": 4373, "time_per_iteration": 2.5695531368255615 }, { "auxiliary_loss_clip": 0.01135375, "auxiliary_loss_mlp": 0.01024822, "balance_loss_clip": 1.05031157, "balance_loss_mlp": 1.01723456, "epoch": 0.5259424036553839, "flos": 24966106656000.0, "grad_norm": 1.960683832959697, "language_loss": 0.6979146, "learning_rate": 1.9291284206659717e-06, "loss": 0.71951658, "num_input_tokens_seen": 94500580, "step": 4374, "time_per_iteration": 2.576040267944336 }, { "auxiliary_loss_clip": 0.01182095, "auxiliary_loss_mlp": 0.0102957, "balance_loss_clip": 1.05309963, "balance_loss_mlp": 1.02211392, "epoch": 0.526062646546023, "flos": 28763912209440.0, "grad_norm": 1.9453742907939997, "language_loss": 0.71162331, "learning_rate": 1.928349943643032e-06, "loss": 0.73373997, "num_input_tokens_seen": 94519680, "step": 4375, "time_per_iteration": 2.529801368713379 }, { "auxiliary_loss_clip": 0.01166295, "auxiliary_loss_mlp": 0.01023196, "balance_loss_clip": 1.05459487, "balance_loss_mlp": 1.01567435, "epoch": 0.526182889436662, "flos": 22821379047840.0, "grad_norm": 1.7199050338528266, "language_loss": 0.81818551, "learning_rate": 1.9275714774893493e-06, "loss": 0.84008038, "num_input_tokens_seen": 94539135, "step": 4376, "time_per_iteration": 2.5035271644592285 }, { "auxiliary_loss_clip": 0.01129921, "auxiliary_loss_mlp": 0.01026224, "balance_loss_clip": 1.04597354, "balance_loss_mlp": 1.01840353, "epoch": 0.5263031323273012, "flos": 22929469932480.0, "grad_norm": 2.5083229192044705, "language_loss": 0.72846961, "learning_rate": 1.9267930223230154e-06, "loss": 0.75003099, "num_input_tokens_seen": 94557610, "step": 4377, "time_per_iteration": 2.5654830932617188 }, { "auxiliary_loss_clip": 0.01154612, "auxiliary_loss_mlp": 0.0103233, "balance_loss_clip": 1.05024529, "balance_loss_mlp": 1.02484703, "epoch": 0.5264233752179402, "flos": 17748637254720.0, "grad_norm": 18.34013681371961, "language_loss": 0.77988422, "learning_rate": 1.9260145782621224e-06, "loss": 0.80175364, "num_input_tokens_seen": 94575390, "step": 4378, "time_per_iteration": 2.492393732070923 }, { "auxiliary_loss_clip": 0.01152283, "auxiliary_loss_mlp": 0.01031524, "balance_loss_clip": 1.05390489, "balance_loss_mlp": 1.02435946, "epoch": 0.5265436181085793, "flos": 24421629533280.0, "grad_norm": 1.9155275107804242, "language_loss": 0.87989962, "learning_rate": 1.925236145424758e-06, "loss": 0.90173769, "num_input_tokens_seen": 94594210, "step": 4379, "time_per_iteration": 2.56430983543396 }, { "auxiliary_loss_clip": 0.01071363, "auxiliary_loss_mlp": 0.0100294, "balance_loss_clip": 1.01811242, "balance_loss_mlp": 1.00161076, "epoch": 0.5266638609992185, "flos": 69207310864320.0, "grad_norm": 0.694764303337086, "language_loss": 0.57559264, "learning_rate": 1.924457723929012e-06, "loss": 0.59633565, "num_input_tokens_seen": 94665020, "step": 4380, "time_per_iteration": 4.753666639328003 }, { "auxiliary_loss_clip": 0.01166079, "auxiliary_loss_mlp": 0.01019916, "balance_loss_clip": 1.05096173, "balance_loss_mlp": 1.01231098, "epoch": 0.5267841038898575, "flos": 20738703169920.0, "grad_norm": 1.446430196496651, "language_loss": 0.82792437, "learning_rate": 1.9236793138929685e-06, "loss": 0.84978426, "num_input_tokens_seen": 94684290, "step": 4381, "time_per_iteration": 2.4930408000946045 }, { "auxiliary_loss_clip": 0.01170918, "auxiliary_loss_mlp": 0.01027803, "balance_loss_clip": 1.05015409, "balance_loss_mlp": 1.02010226, "epoch": 0.5269043467804966, "flos": 17234395824000.0, "grad_norm": 4.508901117841244, "language_loss": 0.81669581, "learning_rate": 1.9229009154347133e-06, "loss": 0.83868307, "num_input_tokens_seen": 94701880, "step": 4382, "time_per_iteration": 2.5050671100616455 }, { "auxiliary_loss_clip": 0.01106572, "auxiliary_loss_mlp": 0.00762292, "balance_loss_clip": 1.04198194, "balance_loss_mlp": 1.00019002, "epoch": 0.5270245896711357, "flos": 18223160916000.0, "grad_norm": 2.2882136460565183, "language_loss": 0.8069526, "learning_rate": 1.922122528672327e-06, "loss": 0.82564127, "num_input_tokens_seen": 94720545, "step": 4383, "time_per_iteration": 2.565284490585327 }, { "auxiliary_loss_clip": 0.01177442, "auxiliary_loss_mlp": 0.01031723, "balance_loss_clip": 1.05223179, "balance_loss_mlp": 1.02461171, "epoch": 0.5271448325617748, "flos": 21287562161760.0, "grad_norm": 2.235319208612525, "language_loss": 0.7826941, "learning_rate": 1.9213441537238914e-06, "loss": 0.80478561, "num_input_tokens_seen": 94737420, "step": 4384, "time_per_iteration": 3.2055299282073975 }, { "auxiliary_loss_clip": 0.01026784, "auxiliary_loss_mlp": 0.01003284, "balance_loss_clip": 1.01326418, "balance_loss_mlp": 1.00200272, "epoch": 0.5272650754524139, "flos": 65495507265120.0, "grad_norm": 0.829727081199027, "language_loss": 0.57345259, "learning_rate": 1.920565790707485e-06, "loss": 0.59375334, "num_input_tokens_seen": 94802810, "step": 4385, "time_per_iteration": 3.409660816192627 }, { "auxiliary_loss_clip": 0.01128999, "auxiliary_loss_mlp": 0.01027381, "balance_loss_clip": 1.04679358, "balance_loss_mlp": 1.01963866, "epoch": 0.527385318343053, "flos": 19676428837440.0, "grad_norm": 2.1426490655716615, "language_loss": 0.65523791, "learning_rate": 1.9197874397411853e-06, "loss": 0.67680174, "num_input_tokens_seen": 94819440, "step": 4386, "time_per_iteration": 2.6807758808135986 }, { "auxiliary_loss_clip": 0.01137435, "auxiliary_loss_mlp": 0.01023872, "balance_loss_clip": 1.04313481, "balance_loss_mlp": 1.01678753, "epoch": 0.5275055612336921, "flos": 12712021535040.0, "grad_norm": 2.60896740803713, "language_loss": 0.67450857, "learning_rate": 1.919009100943067e-06, "loss": 0.69612163, "num_input_tokens_seen": 94835130, "step": 4387, "time_per_iteration": 3.2995054721832275 }, { "auxiliary_loss_clip": 0.01134783, "auxiliary_loss_mlp": 0.01027556, "balance_loss_clip": 1.0475173, "balance_loss_mlp": 1.01958668, "epoch": 0.5276258041243311, "flos": 17749032341280.0, "grad_norm": 2.167849000040118, "language_loss": 0.65644968, "learning_rate": 1.9182307744312043e-06, "loss": 0.67807305, "num_input_tokens_seen": 94852235, "step": 4388, "time_per_iteration": 2.567580461502075 }, { "auxiliary_loss_clip": 0.01154735, "auxiliary_loss_mlp": 0.01030812, "balance_loss_clip": 1.04814744, "balance_loss_mlp": 1.02300429, "epoch": 0.5277460470149702, "flos": 22710450723360.0, "grad_norm": 1.7664827884494285, "language_loss": 0.76524615, "learning_rate": 1.9174524603236676e-06, "loss": 0.78710163, "num_input_tokens_seen": 94871185, "step": 4389, "time_per_iteration": 2.55613374710083 }, { "auxiliary_loss_clip": 0.01155046, "auxiliary_loss_mlp": 0.01023609, "balance_loss_clip": 1.05125928, "balance_loss_mlp": 1.01606917, "epoch": 0.5278662899056094, "flos": 19902703272480.0, "grad_norm": 3.0689393656558157, "language_loss": 0.76163632, "learning_rate": 1.916674158738527e-06, "loss": 0.78342283, "num_input_tokens_seen": 94890090, "step": 4390, "time_per_iteration": 2.5223183631896973 }, { "auxiliary_loss_clip": 0.01134286, "auxiliary_loss_mlp": 0.00762618, "balance_loss_clip": 1.04986477, "balance_loss_mlp": 1.00030112, "epoch": 0.5279865327962484, "flos": 18005219215680.0, "grad_norm": 2.145251885964708, "language_loss": 0.59813571, "learning_rate": 1.9158958697938506e-06, "loss": 0.61710477, "num_input_tokens_seen": 94908470, "step": 4391, "time_per_iteration": 2.53839111328125 }, { "auxiliary_loss_clip": 0.01146189, "auxiliary_loss_mlp": 0.01028554, "balance_loss_clip": 1.04804063, "balance_loss_mlp": 1.02035868, "epoch": 0.5281067756868875, "flos": 15924446936640.0, "grad_norm": 3.4762573918903787, "language_loss": 0.85779727, "learning_rate": 1.9151175936077032e-06, "loss": 0.87954473, "num_input_tokens_seen": 94923440, "step": 4392, "time_per_iteration": 2.500943660736084 }, { "auxiliary_loss_clip": 0.01161744, "auxiliary_loss_mlp": 0.01028114, "balance_loss_clip": 1.04939938, "balance_loss_mlp": 1.0205617, "epoch": 0.5282270185775266, "flos": 19426491514080.0, "grad_norm": 1.7583768016612598, "language_loss": 0.79420847, "learning_rate": 1.9143393302981507e-06, "loss": 0.81610703, "num_input_tokens_seen": 94941125, "step": 4393, "time_per_iteration": 2.4812142848968506 }, { "auxiliary_loss_clip": 0.01156263, "auxiliary_loss_mlp": 0.01033006, "balance_loss_clip": 1.05045307, "balance_loss_mlp": 1.02507234, "epoch": 0.5283472614681657, "flos": 16399617103200.0, "grad_norm": 1.9006384020414302, "language_loss": 0.835446, "learning_rate": 1.913561079983252e-06, "loss": 0.85733879, "num_input_tokens_seen": 94959950, "step": 4394, "time_per_iteration": 2.505732297897339 }, { "auxiliary_loss_clip": 0.01153013, "auxiliary_loss_mlp": 0.0103016, "balance_loss_clip": 1.04671264, "balance_loss_mlp": 1.0223577, "epoch": 0.5284675043588047, "flos": 26760528202560.0, "grad_norm": 2.026407431629551, "language_loss": 0.74782848, "learning_rate": 1.9127828427810693e-06, "loss": 0.76966023, "num_input_tokens_seen": 94980515, "step": 4395, "time_per_iteration": 2.5552170276641846 }, { "auxiliary_loss_clip": 0.01145498, "auxiliary_loss_mlp": 0.01031115, "balance_loss_clip": 1.04819536, "balance_loss_mlp": 1.02287197, "epoch": 0.5285877472494439, "flos": 19899901749600.0, "grad_norm": 1.8368029031418889, "language_loss": 0.80946201, "learning_rate": 1.9120046188096607e-06, "loss": 0.83122814, "num_input_tokens_seen": 94998560, "step": 4396, "time_per_iteration": 2.549801826477051 }, { "auxiliary_loss_clip": 0.01152543, "auxiliary_loss_mlp": 0.01030307, "balance_loss_clip": 1.05353284, "balance_loss_mlp": 1.02232051, "epoch": 0.528707990140083, "flos": 20011261077600.0, "grad_norm": 1.8177112046913058, "language_loss": 0.74040318, "learning_rate": 1.9112264081870804e-06, "loss": 0.76223171, "num_input_tokens_seen": 95016950, "step": 4397, "time_per_iteration": 2.5176665782928467 }, { "auxiliary_loss_clip": 0.01133311, "auxiliary_loss_mlp": 0.01031561, "balance_loss_clip": 1.04735982, "balance_loss_mlp": 1.0231564, "epoch": 0.528828233030722, "flos": 20667959535360.0, "grad_norm": 2.382800392860685, "language_loss": 0.75363851, "learning_rate": 1.9104482110313843e-06, "loss": 0.77528721, "num_input_tokens_seen": 95036540, "step": 4398, "time_per_iteration": 2.549166440963745 }, { "auxiliary_loss_clip": 0.01165628, "auxiliary_loss_mlp": 0.01027514, "balance_loss_clip": 1.05159807, "balance_loss_mlp": 1.02022409, "epoch": 0.5289484759213612, "flos": 25192452924960.0, "grad_norm": 2.017270539928092, "language_loss": 0.73899651, "learning_rate": 1.909670027460623e-06, "loss": 0.76092798, "num_input_tokens_seen": 95053840, "step": 4399, "time_per_iteration": 2.5313267707824707 }, { "auxiliary_loss_clip": 0.01165952, "auxiliary_loss_mlp": 0.01024886, "balance_loss_clip": 1.05147886, "balance_loss_mlp": 1.01767707, "epoch": 0.5290687188120002, "flos": 31139260204800.0, "grad_norm": 1.8689930288766083, "language_loss": 0.71949863, "learning_rate": 1.908891857592847e-06, "loss": 0.74140704, "num_input_tokens_seen": 95074910, "step": 4400, "time_per_iteration": 2.5715370178222656 }, { "auxiliary_loss_clip": 0.01130862, "auxiliary_loss_mlp": 0.01028716, "balance_loss_clip": 1.04721594, "balance_loss_mlp": 1.02087259, "epoch": 0.5291889617026393, "flos": 20119854799680.0, "grad_norm": 2.1204519209943813, "language_loss": 0.89593029, "learning_rate": 1.9081137015461034e-06, "loss": 0.91752607, "num_input_tokens_seen": 95090985, "step": 4401, "time_per_iteration": 2.529270887374878 }, { "auxiliary_loss_clip": 0.01117695, "auxiliary_loss_mlp": 0.01027075, "balance_loss_clip": 1.04898739, "balance_loss_mlp": 1.01956451, "epoch": 0.5293092045932785, "flos": 19643750792160.0, "grad_norm": 1.741328340241747, "language_loss": 0.90535915, "learning_rate": 1.9073355594384383e-06, "loss": 0.92680681, "num_input_tokens_seen": 95109225, "step": 4402, "time_per_iteration": 2.558415174484253 }, { "auxiliary_loss_clip": 0.01130157, "auxiliary_loss_mlp": 0.01030899, "balance_loss_clip": 1.04730916, "balance_loss_mlp": 1.02342796, "epoch": 0.5294294474839175, "flos": 24317740932960.0, "grad_norm": 8.042123752529532, "language_loss": 0.80412173, "learning_rate": 1.906557431387895e-06, "loss": 0.82573235, "num_input_tokens_seen": 95128215, "step": 4403, "time_per_iteration": 2.5844004154205322 }, { "auxiliary_loss_clip": 0.0113745, "auxiliary_loss_mlp": 0.01028283, "balance_loss_clip": 1.05343008, "balance_loss_mlp": 1.02053404, "epoch": 0.5295496903745566, "flos": 18875944425120.0, "grad_norm": 2.027460532491256, "language_loss": 0.78983301, "learning_rate": 1.905779317512516e-06, "loss": 0.8114903, "num_input_tokens_seen": 95145760, "step": 4404, "time_per_iteration": 2.535022020339966 }, { "auxiliary_loss_clip": 0.01162504, "auxiliary_loss_mlp": 0.01026441, "balance_loss_clip": 1.04905987, "balance_loss_mlp": 1.01913023, "epoch": 0.5296699332651957, "flos": 20923104817920.0, "grad_norm": 1.8615360915125665, "language_loss": 0.80205214, "learning_rate": 1.9050012179303385e-06, "loss": 0.82394159, "num_input_tokens_seen": 95164270, "step": 4405, "time_per_iteration": 2.508223056793213 }, { "auxiliary_loss_clip": 0.01168918, "auxiliary_loss_mlp": 0.01024852, "balance_loss_clip": 1.05079257, "balance_loss_mlp": 1.01716304, "epoch": 0.5297901761558348, "flos": 22046748458400.0, "grad_norm": 2.299814833061545, "language_loss": 0.69306344, "learning_rate": 1.904223132759401e-06, "loss": 0.71500111, "num_input_tokens_seen": 95182870, "step": 4406, "time_per_iteration": 4.07892370223999 }, { "auxiliary_loss_clip": 0.01167065, "auxiliary_loss_mlp": 0.01028614, "balance_loss_clip": 1.05026388, "balance_loss_mlp": 1.0210979, "epoch": 0.5299104190464738, "flos": 21798499232160.0, "grad_norm": 2.261561242200187, "language_loss": 0.69205201, "learning_rate": 1.9034450621177383e-06, "loss": 0.71400881, "num_input_tokens_seen": 95201190, "step": 4407, "time_per_iteration": 2.4833242893218994 }, { "auxiliary_loss_clip": 0.0116692, "auxiliary_loss_mlp": 0.01031808, "balance_loss_clip": 1.0529182, "balance_loss_mlp": 1.0238744, "epoch": 0.530030661937113, "flos": 14720793085920.0, "grad_norm": 1.7770362419403682, "language_loss": 0.70307994, "learning_rate": 1.9026670061233824e-06, "loss": 0.7250672, "num_input_tokens_seen": 95218625, "step": 4408, "time_per_iteration": 2.483349084854126 }, { "auxiliary_loss_clip": 0.01148621, "auxiliary_loss_mlp": 0.01030573, "balance_loss_clip": 1.0518446, "balance_loss_mlp": 1.02346838, "epoch": 0.5301509048277521, "flos": 21251507922240.0, "grad_norm": 1.6143442446748593, "language_loss": 0.80733001, "learning_rate": 1.901888964894365e-06, "loss": 0.82912195, "num_input_tokens_seen": 95237665, "step": 4409, "time_per_iteration": 2.610455274581909 }, { "auxiliary_loss_clip": 0.01182694, "auxiliary_loss_mlp": 0.01026716, "balance_loss_clip": 1.05211115, "balance_loss_mlp": 1.01902127, "epoch": 0.5302711477183911, "flos": 25957062682560.0, "grad_norm": 2.1224931422161233, "language_loss": 0.67845351, "learning_rate": 1.9011109385487134e-06, "loss": 0.70054758, "num_input_tokens_seen": 95258915, "step": 4410, "time_per_iteration": 3.4122140407562256 }, { "auxiliary_loss_clip": 0.01182915, "auxiliary_loss_mlp": 0.01025468, "balance_loss_clip": 1.05382335, "balance_loss_mlp": 1.01706982, "epoch": 0.5303913906090303, "flos": 22273130644320.0, "grad_norm": 2.268623946621441, "language_loss": 0.66072059, "learning_rate": 1.900332927204454e-06, "loss": 0.68280435, "num_input_tokens_seen": 95277365, "step": 4411, "time_per_iteration": 2.552734375 }, { "auxiliary_loss_clip": 0.0115622, "auxiliary_loss_mlp": 0.01028826, "balance_loss_clip": 1.04999197, "balance_loss_mlp": 1.02100301, "epoch": 0.5305116334996693, "flos": 24936014631840.0, "grad_norm": 1.7269892721987645, "language_loss": 0.76480794, "learning_rate": 1.8995549309796097e-06, "loss": 0.78665841, "num_input_tokens_seen": 95296670, "step": 4412, "time_per_iteration": 3.3970329761505127 }, { "auxiliary_loss_clip": 0.0116997, "auxiliary_loss_mlp": 0.01034547, "balance_loss_clip": 1.05314028, "balance_loss_mlp": 1.02708495, "epoch": 0.5306318763903084, "flos": 20189341340640.0, "grad_norm": 2.1311352402905954, "language_loss": 0.75798273, "learning_rate": 1.8987769499922028e-06, "loss": 0.78002787, "num_input_tokens_seen": 95315640, "step": 4413, "time_per_iteration": 2.4893364906311035 }, { "auxiliary_loss_clip": 0.01166412, "auxiliary_loss_mlp": 0.00762154, "balance_loss_clip": 1.05168474, "balance_loss_mlp": 1.00034142, "epoch": 0.5307521192809476, "flos": 20266370443200.0, "grad_norm": 2.2683011588842596, "language_loss": 0.70802349, "learning_rate": 1.897998984360252e-06, "loss": 0.72730911, "num_input_tokens_seen": 95334610, "step": 4414, "time_per_iteration": 2.473365306854248 }, { "auxiliary_loss_clip": 0.01148306, "auxiliary_loss_mlp": 0.01035435, "balance_loss_clip": 1.04843831, "balance_loss_mlp": 1.02831864, "epoch": 0.5308723621715866, "flos": 28844281589280.0, "grad_norm": 1.3970698269089006, "language_loss": 0.78345454, "learning_rate": 1.897221034201775e-06, "loss": 0.80529189, "num_input_tokens_seen": 95358350, "step": 4415, "time_per_iteration": 2.6044163703918457 }, { "auxiliary_loss_clip": 0.01136743, "auxiliary_loss_mlp": 0.01023692, "balance_loss_clip": 1.04697061, "balance_loss_mlp": 1.01676893, "epoch": 0.5309926050622257, "flos": 27457770519840.0, "grad_norm": 1.4523453662294938, "language_loss": 0.66867661, "learning_rate": 1.8964430996347842e-06, "loss": 0.69028103, "num_input_tokens_seen": 95379900, "step": 4416, "time_per_iteration": 2.6198270320892334 }, { "auxiliary_loss_clip": 0.01152077, "auxiliary_loss_mlp": 0.01031079, "balance_loss_clip": 1.04905272, "balance_loss_mlp": 1.02279985, "epoch": 0.5311128479528648, "flos": 20514547835520.0, "grad_norm": 1.842874167155541, "language_loss": 0.82947659, "learning_rate": 1.8956651807772931e-06, "loss": 0.85130811, "num_input_tokens_seen": 95397935, "step": 4417, "time_per_iteration": 2.53528094291687 }, { "auxiliary_loss_clip": 0.01163785, "auxiliary_loss_mlp": 0.01028941, "balance_loss_clip": 1.05090106, "balance_loss_mlp": 1.02167845, "epoch": 0.5312330908435039, "flos": 21397664396160.0, "grad_norm": 1.5590419833876699, "language_loss": 0.83726192, "learning_rate": 1.8948872777473115e-06, "loss": 0.85918909, "num_input_tokens_seen": 95415890, "step": 4418, "time_per_iteration": 2.4960360527038574 }, { "auxiliary_loss_clip": 0.01152751, "auxiliary_loss_mlp": 0.01037988, "balance_loss_clip": 1.04981172, "balance_loss_mlp": 1.02972674, "epoch": 0.531353333734143, "flos": 24717354592320.0, "grad_norm": 1.7158031730678103, "language_loss": 0.63421869, "learning_rate": 1.8941093906628458e-06, "loss": 0.65612602, "num_input_tokens_seen": 95433675, "step": 4419, "time_per_iteration": 2.546546220779419 }, { "auxiliary_loss_clip": 0.01147255, "auxiliary_loss_mlp": 0.01032747, "balance_loss_clip": 1.04842198, "balance_loss_mlp": 1.02532005, "epoch": 0.531473576624782, "flos": 30480694065120.0, "grad_norm": 1.5911597946001315, "language_loss": 0.70833516, "learning_rate": 1.893331519641902e-06, "loss": 0.73013514, "num_input_tokens_seen": 95455820, "step": 4420, "time_per_iteration": 2.5878655910491943 }, { "auxiliary_loss_clip": 0.01124697, "auxiliary_loss_mlp": 0.01031091, "balance_loss_clip": 1.04229736, "balance_loss_mlp": 1.02316976, "epoch": 0.5315938195154212, "flos": 23002979172960.0, "grad_norm": 2.3036663698469133, "language_loss": 0.73998272, "learning_rate": 1.8925536648024815e-06, "loss": 0.76154059, "num_input_tokens_seen": 95473240, "step": 4421, "time_per_iteration": 2.5818796157836914 }, { "auxiliary_loss_clip": 0.01182429, "auxiliary_loss_mlp": 0.01025021, "balance_loss_clip": 1.0524137, "balance_loss_mlp": 1.01736212, "epoch": 0.5317140624060602, "flos": 22748588146560.0, "grad_norm": 1.8618662822583194, "language_loss": 0.75979215, "learning_rate": 1.8917758262625849e-06, "loss": 0.78186667, "num_input_tokens_seen": 95493480, "step": 4422, "time_per_iteration": 2.4602174758911133 }, { "auxiliary_loss_clip": 0.01145583, "auxiliary_loss_mlp": 0.01032093, "balance_loss_clip": 1.04962254, "balance_loss_mlp": 1.02488947, "epoch": 0.5318343052966993, "flos": 22821091712160.0, "grad_norm": 1.6806355533732282, "language_loss": 0.80924165, "learning_rate": 1.8909980041402089e-06, "loss": 0.83101845, "num_input_tokens_seen": 95512075, "step": 4423, "time_per_iteration": 2.5615642070770264 }, { "auxiliary_loss_clip": 0.01159584, "auxiliary_loss_mlp": 0.01027933, "balance_loss_clip": 1.04839039, "balance_loss_mlp": 1.02040195, "epoch": 0.5319545481873384, "flos": 13626092126880.0, "grad_norm": 2.087697602002662, "language_loss": 0.6568048, "learning_rate": 1.8902201985533494e-06, "loss": 0.67867994, "num_input_tokens_seen": 95529340, "step": 4424, "time_per_iteration": 2.4573988914489746 }, { "auxiliary_loss_clip": 0.01153195, "auxiliary_loss_mlp": 0.01028811, "balance_loss_clip": 1.05193973, "balance_loss_mlp": 1.02101445, "epoch": 0.5320747910779775, "flos": 22162525572480.0, "grad_norm": 2.0849710829557173, "language_loss": 0.74483812, "learning_rate": 1.8894424096199983e-06, "loss": 0.76665819, "num_input_tokens_seen": 95548545, "step": 4425, "time_per_iteration": 2.5466535091400146 }, { "auxiliary_loss_clip": 0.01167197, "auxiliary_loss_mlp": 0.01028727, "balance_loss_clip": 1.05254233, "balance_loss_mlp": 1.02108026, "epoch": 0.5321950339686166, "flos": 18588085180320.0, "grad_norm": 1.9074383122321865, "language_loss": 0.85704124, "learning_rate": 1.8886646374581463e-06, "loss": 0.87900043, "num_input_tokens_seen": 95567770, "step": 4426, "time_per_iteration": 2.51999568939209 }, { "auxiliary_loss_clip": 0.01164025, "auxiliary_loss_mlp": 0.0103005, "balance_loss_clip": 1.0489049, "balance_loss_mlp": 1.02231956, "epoch": 0.5323152768592557, "flos": 22856822699040.0, "grad_norm": 1.6136112105486435, "language_loss": 0.71113056, "learning_rate": 1.8878868821857795e-06, "loss": 0.73307133, "num_input_tokens_seen": 95587420, "step": 4427, "time_per_iteration": 2.5084800720214844 }, { "auxiliary_loss_clip": 0.01119098, "auxiliary_loss_mlp": 0.0102879, "balance_loss_clip": 1.04307437, "balance_loss_mlp": 1.02091074, "epoch": 0.5324355197498948, "flos": 33948695752800.0, "grad_norm": 2.6273859013819427, "language_loss": 0.75377697, "learning_rate": 1.8871091439208838e-06, "loss": 0.77525586, "num_input_tokens_seen": 95609030, "step": 4428, "time_per_iteration": 2.6853768825531006 }, { "auxiliary_loss_clip": 0.01120906, "auxiliary_loss_mlp": 0.01033399, "balance_loss_clip": 1.04625142, "balance_loss_mlp": 1.02553701, "epoch": 0.5325557626405338, "flos": 23256723694080.0, "grad_norm": 7.868983126400292, "language_loss": 0.76764345, "learning_rate": 1.8863314227814414e-06, "loss": 0.78918648, "num_input_tokens_seen": 95627340, "step": 4429, "time_per_iteration": 2.5927271842956543 }, { "auxiliary_loss_clip": 0.01172144, "auxiliary_loss_mlp": 0.01027976, "balance_loss_clip": 1.0522058, "balance_loss_mlp": 1.02039421, "epoch": 0.532676005531173, "flos": 26718691332480.0, "grad_norm": 3.0993748915889787, "language_loss": 0.4892351, "learning_rate": 1.8855537188854313e-06, "loss": 0.51123637, "num_input_tokens_seen": 95646315, "step": 4430, "time_per_iteration": 2.5373682975769043 }, { "auxiliary_loss_clip": 0.01172089, "auxiliary_loss_mlp": 0.0103071, "balance_loss_clip": 1.05128741, "balance_loss_mlp": 1.02297962, "epoch": 0.5327962484218121, "flos": 17894614143840.0, "grad_norm": 1.8804659843668308, "language_loss": 0.78030604, "learning_rate": 1.8847760323508315e-06, "loss": 0.80233407, "num_input_tokens_seen": 95665220, "step": 4431, "time_per_iteration": 2.5360002517700195 }, { "auxiliary_loss_clip": 0.01148663, "auxiliary_loss_mlp": 0.01028988, "balance_loss_clip": 1.05086875, "balance_loss_mlp": 1.02178478, "epoch": 0.5329164913124511, "flos": 17925388590240.0, "grad_norm": 1.9196326523549299, "language_loss": 0.75422728, "learning_rate": 1.883998363295616e-06, "loss": 0.77600384, "num_input_tokens_seen": 95682700, "step": 4432, "time_per_iteration": 3.349290370941162 }, { "auxiliary_loss_clip": 0.01055709, "auxiliary_loss_mlp": 0.01001795, "balance_loss_clip": 1.01605844, "balance_loss_mlp": 1.00051904, "epoch": 0.5330367342030903, "flos": 57254188290240.0, "grad_norm": 0.875410332554564, "language_loss": 0.62602299, "learning_rate": 1.8832207118377565e-06, "loss": 0.64659804, "num_input_tokens_seen": 95738070, "step": 4433, "time_per_iteration": 3.045703411102295 }, { "auxiliary_loss_clip": 0.01180718, "auxiliary_loss_mlp": 0.0103019, "balance_loss_clip": 1.05340755, "balance_loss_mlp": 1.02314508, "epoch": 0.5331569770937293, "flos": 17420521486080.0, "grad_norm": 5.47990908066652, "language_loss": 0.69369596, "learning_rate": 1.882443078095222e-06, "loss": 0.71580499, "num_input_tokens_seen": 95756950, "step": 4434, "time_per_iteration": 2.455066204071045 }, { "auxiliary_loss_clip": 0.01042738, "auxiliary_loss_mlp": 0.01001566, "balance_loss_clip": 1.01918793, "balance_loss_mlp": 1.00026655, "epoch": 0.5332772199843684, "flos": 56750793781440.0, "grad_norm": 0.8701299662460252, "language_loss": 0.66808534, "learning_rate": 1.8816654621859794e-06, "loss": 0.68852836, "num_input_tokens_seen": 95816615, "step": 4435, "time_per_iteration": 3.050875425338745 }, { "auxiliary_loss_clip": 0.01179198, "auxiliary_loss_mlp": 0.01030485, "balance_loss_clip": 1.05230701, "balance_loss_mlp": 1.0226593, "epoch": 0.5333974628750076, "flos": 18697756411200.0, "grad_norm": 2.448877561688296, "language_loss": 0.72199619, "learning_rate": 1.8808878642279915e-06, "loss": 0.744093, "num_input_tokens_seen": 95832020, "step": 4436, "time_per_iteration": 3.191826820373535 }, { "auxiliary_loss_clip": 0.01140087, "auxiliary_loss_mlp": 0.01031831, "balance_loss_clip": 1.04443204, "balance_loss_mlp": 1.02333784, "epoch": 0.5335177057656466, "flos": 23805510852000.0, "grad_norm": 2.028104632396337, "language_loss": 0.64900625, "learning_rate": 1.8801102843392209e-06, "loss": 0.67072546, "num_input_tokens_seen": 95851425, "step": 4437, "time_per_iteration": 2.578589677810669 }, { "auxiliary_loss_clip": 0.01137862, "auxiliary_loss_mlp": 0.01030285, "balance_loss_clip": 1.04639387, "balance_loss_mlp": 1.02220869, "epoch": 0.5336379486562857, "flos": 25078687160640.0, "grad_norm": 1.5465712270606171, "language_loss": 0.85208458, "learning_rate": 1.8793327226376238e-06, "loss": 0.87376606, "num_input_tokens_seen": 95870745, "step": 4438, "time_per_iteration": 2.592703342437744 }, { "auxiliary_loss_clip": 0.011601, "auxiliary_loss_mlp": 0.0103832, "balance_loss_clip": 1.05246174, "balance_loss_mlp": 1.03040433, "epoch": 0.5337581915469248, "flos": 21396694638240.0, "grad_norm": 1.7241769170354893, "language_loss": 0.80307353, "learning_rate": 1.8785551792411569e-06, "loss": 0.82505769, "num_input_tokens_seen": 95889755, "step": 4439, "time_per_iteration": 3.2784945964813232 }, { "auxiliary_loss_clip": 0.01153102, "auxiliary_loss_mlp": 0.01024479, "balance_loss_clip": 1.05073404, "balance_loss_mlp": 1.01748741, "epoch": 0.5338784344375639, "flos": 14865907968000.0, "grad_norm": 1.9158273628971725, "language_loss": 0.82469356, "learning_rate": 1.8777776542677733e-06, "loss": 0.84646928, "num_input_tokens_seen": 95907805, "step": 4440, "time_per_iteration": 2.5092251300811768 }, { "auxiliary_loss_clip": 0.0113939, "auxiliary_loss_mlp": 0.01022866, "balance_loss_clip": 1.04602623, "balance_loss_mlp": 1.01560664, "epoch": 0.5339986773282029, "flos": 20813505420960.0, "grad_norm": 2.1428733890438507, "language_loss": 0.72980285, "learning_rate": 1.8770001478354216e-06, "loss": 0.75142539, "num_input_tokens_seen": 95927480, "step": 4441, "time_per_iteration": 2.5767526626586914 }, { "auxiliary_loss_clip": 0.01165248, "auxiliary_loss_mlp": 0.01025251, "balance_loss_clip": 1.05123329, "balance_loss_mlp": 1.01733601, "epoch": 0.5341189202188421, "flos": 17969093142240.0, "grad_norm": 2.140188206744242, "language_loss": 0.84105897, "learning_rate": 1.8762226600620504e-06, "loss": 0.86296391, "num_input_tokens_seen": 95946095, "step": 4442, "time_per_iteration": 2.476121664047241 }, { "auxiliary_loss_clip": 0.01162683, "auxiliary_loss_mlp": 0.01027443, "balance_loss_clip": 1.05073571, "balance_loss_mlp": 1.01924157, "epoch": 0.5342391631094812, "flos": 11031868590240.0, "grad_norm": 2.6937365275458274, "language_loss": 0.58643252, "learning_rate": 1.8754451910656031e-06, "loss": 0.60833383, "num_input_tokens_seen": 95959995, "step": 4443, "time_per_iteration": 2.5115911960601807 }, { "auxiliary_loss_clip": 0.01132349, "auxiliary_loss_mlp": 0.0102559, "balance_loss_clip": 1.04790258, "balance_loss_mlp": 1.01805043, "epoch": 0.5343594060001202, "flos": 15339138618720.0, "grad_norm": 1.8215398977998534, "language_loss": 0.82471341, "learning_rate": 1.8746677409640212e-06, "loss": 0.84629279, "num_input_tokens_seen": 95977095, "step": 4444, "time_per_iteration": 2.5727782249450684 }, { "auxiliary_loss_clip": 0.01173735, "auxiliary_loss_mlp": 0.01032457, "balance_loss_clip": 1.05459654, "balance_loss_mlp": 1.02501249, "epoch": 0.5344796488907594, "flos": 26900902045920.0, "grad_norm": 1.8504053605642417, "language_loss": 0.84506112, "learning_rate": 1.8738903098752432e-06, "loss": 0.86712301, "num_input_tokens_seen": 95996225, "step": 4445, "time_per_iteration": 2.5756003856658936 }, { "auxiliary_loss_clip": 0.0115566, "auxiliary_loss_mlp": 0.01025274, "balance_loss_clip": 1.05055606, "balance_loss_mlp": 1.01819348, "epoch": 0.5345998917813984, "flos": 25411220715360.0, "grad_norm": 2.9806193041250912, "language_loss": 0.72937763, "learning_rate": 1.8731128979172052e-06, "loss": 0.75118697, "num_input_tokens_seen": 96015425, "step": 4446, "time_per_iteration": 2.5926973819732666 }, { "auxiliary_loss_clip": 0.01149534, "auxiliary_loss_mlp": 0.01030263, "balance_loss_clip": 1.04890966, "balance_loss_mlp": 1.02277112, "epoch": 0.5347201346720375, "flos": 32853384205440.0, "grad_norm": 2.3184018112215115, "language_loss": 0.67263514, "learning_rate": 1.8723355052078394e-06, "loss": 0.69443309, "num_input_tokens_seen": 96035460, "step": 4447, "time_per_iteration": 2.631833076477051 }, { "auxiliary_loss_clip": 0.01163336, "auxiliary_loss_mlp": 0.0102373, "balance_loss_clip": 1.04933882, "balance_loss_mlp": 1.01561141, "epoch": 0.5348403775626767, "flos": 17967943799520.0, "grad_norm": 2.2728956199709502, "language_loss": 0.77449656, "learning_rate": 1.8715581318650765e-06, "loss": 0.79636729, "num_input_tokens_seen": 96054515, "step": 4448, "time_per_iteration": 2.4791722297668457 }, { "auxiliary_loss_clip": 0.01149134, "auxiliary_loss_mlp": 0.01036317, "balance_loss_clip": 1.04965878, "balance_loss_mlp": 1.02844334, "epoch": 0.5349606204533157, "flos": 17603342787840.0, "grad_norm": 2.2034681290079816, "language_loss": 0.81758261, "learning_rate": 1.8707807780068422e-06, "loss": 0.83943713, "num_input_tokens_seen": 96072330, "step": 4449, "time_per_iteration": 2.5297300815582275 }, { "auxiliary_loss_clip": 0.01153195, "auxiliary_loss_mlp": 0.01031394, "balance_loss_clip": 1.05025756, "balance_loss_mlp": 1.02363384, "epoch": 0.5350808633439548, "flos": 29167835904000.0, "grad_norm": 2.041306722984649, "language_loss": 0.66607416, "learning_rate": 1.8700034437510611e-06, "loss": 0.68792009, "num_input_tokens_seen": 96092425, "step": 4450, "time_per_iteration": 2.559838056564331 }, { "auxiliary_loss_clip": 0.01131037, "auxiliary_loss_mlp": 0.0103698, "balance_loss_clip": 1.04648089, "balance_loss_mlp": 1.02881455, "epoch": 0.5352011062345938, "flos": 19500000754560.0, "grad_norm": 2.1593037538274245, "language_loss": 0.81523573, "learning_rate": 1.8692261292156549e-06, "loss": 0.83691591, "num_input_tokens_seen": 96111660, "step": 4451, "time_per_iteration": 2.5490548610687256 }, { "auxiliary_loss_clip": 0.01182273, "auxiliary_loss_mlp": 0.01030467, "balance_loss_clip": 1.05520284, "balance_loss_mlp": 1.02313578, "epoch": 0.535321349125233, "flos": 23477646502080.0, "grad_norm": 1.9003868309267589, "language_loss": 0.8088572, "learning_rate": 1.8684488345185401e-06, "loss": 0.83098459, "num_input_tokens_seen": 96131835, "step": 4452, "time_per_iteration": 2.471238136291504 }, { "auxiliary_loss_clip": 0.01184705, "auxiliary_loss_mlp": 0.01033358, "balance_loss_clip": 1.05493796, "balance_loss_mlp": 1.02614617, "epoch": 0.535441592015872, "flos": 20478062592480.0, "grad_norm": 2.4275291194827404, "language_loss": 0.79095125, "learning_rate": 1.8676715597776332e-06, "loss": 0.81313187, "num_input_tokens_seen": 96150180, "step": 4453, "time_per_iteration": 2.445166826248169 }, { "auxiliary_loss_clip": 0.01113938, "auxiliary_loss_mlp": 0.01028211, "balance_loss_clip": 1.04541755, "balance_loss_mlp": 1.02070665, "epoch": 0.5355618349065111, "flos": 19573150825440.0, "grad_norm": 1.7464563782103881, "language_loss": 0.75924158, "learning_rate": 1.8668943051108455e-06, "loss": 0.78066301, "num_input_tokens_seen": 96167485, "step": 4454, "time_per_iteration": 2.5586471557617188 }, { "auxiliary_loss_clip": 0.01151639, "auxiliary_loss_mlp": 0.01032623, "balance_loss_clip": 1.04892683, "balance_loss_mlp": 1.02459967, "epoch": 0.5356820777971503, "flos": 24024637812000.0, "grad_norm": 2.1981737370781524, "language_loss": 0.76420474, "learning_rate": 1.8661170706360856e-06, "loss": 0.78604734, "num_input_tokens_seen": 96186650, "step": 4455, "time_per_iteration": 2.535245895385742 }, { "auxiliary_loss_clip": 0.01168192, "auxiliary_loss_mlp": 0.01026651, "balance_loss_clip": 1.05441248, "balance_loss_mlp": 1.01911747, "epoch": 0.5358023206877893, "flos": 20884680059040.0, "grad_norm": 1.485297792518803, "language_loss": 0.81413269, "learning_rate": 1.8653398564712594e-06, "loss": 0.83608115, "num_input_tokens_seen": 96205595, "step": 4456, "time_per_iteration": 2.4929521083831787 }, { "auxiliary_loss_clip": 0.01166248, "auxiliary_loss_mlp": 0.01025179, "balance_loss_clip": 1.05233932, "balance_loss_mlp": 1.01796675, "epoch": 0.5359225635784284, "flos": 22418999782560.0, "grad_norm": 1.5607718889845816, "language_loss": 0.82432407, "learning_rate": 1.8645626627342704e-06, "loss": 0.84623832, "num_input_tokens_seen": 96226360, "step": 4457, "time_per_iteration": 2.509904146194458 }, { "auxiliary_loss_clip": 0.01170981, "auxiliary_loss_mlp": 0.01023915, "balance_loss_clip": 1.0513382, "balance_loss_mlp": 1.0163033, "epoch": 0.5360428064690675, "flos": 24097787882880.0, "grad_norm": 2.0398102300388974, "language_loss": 0.80876982, "learning_rate": 1.8637854895430172e-06, "loss": 0.83071876, "num_input_tokens_seen": 96245625, "step": 4458, "time_per_iteration": 4.041447877883911 }, { "auxiliary_loss_clip": 0.01129622, "auxiliary_loss_mlp": 0.01029891, "balance_loss_clip": 1.04463339, "balance_loss_mlp": 1.02128386, "epoch": 0.5361630493597066, "flos": 21434508808800.0, "grad_norm": 2.9560774804609946, "language_loss": 0.6953823, "learning_rate": 1.8630083370153978e-06, "loss": 0.71697748, "num_input_tokens_seen": 96265265, "step": 4459, "time_per_iteration": 2.552347183227539 }, { "auxiliary_loss_clip": 0.01026337, "auxiliary_loss_mlp": 0.01002314, "balance_loss_clip": 1.01404476, "balance_loss_mlp": 1.00106192, "epoch": 0.5362832922503457, "flos": 68888706701760.0, "grad_norm": 0.7488331012306003, "language_loss": 0.5540691, "learning_rate": 1.8622312052693041e-06, "loss": 0.57435554, "num_input_tokens_seen": 96326445, "step": 4460, "time_per_iteration": 3.280698776245117 }, { "auxiliary_loss_clip": 0.0115549, "auxiliary_loss_mlp": 0.01029311, "balance_loss_clip": 1.04594958, "balance_loss_mlp": 1.02193761, "epoch": 0.5364035351409848, "flos": 9793704929280.0, "grad_norm": 2.0756046637175505, "language_loss": 0.71788174, "learning_rate": 1.8614540944226267e-06, "loss": 0.73972976, "num_input_tokens_seen": 96343115, "step": 4461, "time_per_iteration": 2.455687999725342 }, { "auxiliary_loss_clip": 0.01150697, "auxiliary_loss_mlp": 0.01023904, "balance_loss_clip": 1.05287957, "balance_loss_mlp": 1.01720762, "epoch": 0.5365237780316239, "flos": 23290084161600.0, "grad_norm": 1.9044927736149022, "language_loss": 0.67970395, "learning_rate": 1.8606770045932537e-06, "loss": 0.70144993, "num_input_tokens_seen": 96362230, "step": 4462, "time_per_iteration": 3.3082761764526367 }, { "auxiliary_loss_clip": 0.01132322, "auxiliary_loss_mlp": 0.01028142, "balance_loss_clip": 1.04379964, "balance_loss_mlp": 1.01968968, "epoch": 0.5366440209222629, "flos": 26578137904320.0, "grad_norm": 1.8048264643383831, "language_loss": 0.81464565, "learning_rate": 1.859899935899068e-06, "loss": 0.83625031, "num_input_tokens_seen": 96382085, "step": 4463, "time_per_iteration": 2.6179988384246826 }, { "auxiliary_loss_clip": 0.01151818, "auxiliary_loss_mlp": 0.01027935, "balance_loss_clip": 1.0517143, "balance_loss_mlp": 1.02015054, "epoch": 0.5367642638129021, "flos": 19608055722240.0, "grad_norm": 1.9173903517033555, "language_loss": 0.7903496, "learning_rate": 1.8591228884579506e-06, "loss": 0.81214714, "num_input_tokens_seen": 96400580, "step": 4464, "time_per_iteration": 3.236088514328003 }, { "auxiliary_loss_clip": 0.01142542, "auxiliary_loss_mlp": 0.01029016, "balance_loss_clip": 1.04800797, "balance_loss_mlp": 1.02049232, "epoch": 0.5368845067035412, "flos": 23915217999840.0, "grad_norm": 1.9438653431739015, "language_loss": 0.82186651, "learning_rate": 1.8583458623877795e-06, "loss": 0.84358209, "num_input_tokens_seen": 96419680, "step": 4465, "time_per_iteration": 2.556576728820801 }, { "auxiliary_loss_clip": 0.01170771, "auxiliary_loss_mlp": 0.01025472, "balance_loss_clip": 1.05269504, "balance_loss_mlp": 1.01740789, "epoch": 0.5370047495941802, "flos": 16873135089600.0, "grad_norm": 1.8914426734117873, "language_loss": 0.74285388, "learning_rate": 1.8575688578064281e-06, "loss": 0.76481628, "num_input_tokens_seen": 96437805, "step": 4466, "time_per_iteration": 2.4849202632904053 }, { "auxiliary_loss_clip": 0.01171847, "auxiliary_loss_mlp": 0.01027629, "balance_loss_clip": 1.05456161, "balance_loss_mlp": 1.02044654, "epoch": 0.5371249924848194, "flos": 20740930021440.0, "grad_norm": 1.7108019603836448, "language_loss": 0.76465684, "learning_rate": 1.8567918748317674e-06, "loss": 0.78665161, "num_input_tokens_seen": 96457155, "step": 4467, "time_per_iteration": 2.5107898712158203 }, { "auxiliary_loss_clip": 0.0113894, "auxiliary_loss_mlp": 0.01026375, "balance_loss_clip": 1.04597056, "balance_loss_mlp": 1.01876926, "epoch": 0.5372452353754584, "flos": 17968123384320.0, "grad_norm": 2.2119857581544355, "language_loss": 0.8283878, "learning_rate": 1.8560149135816659e-06, "loss": 0.85004091, "num_input_tokens_seen": 96473990, "step": 4468, "time_per_iteration": 2.5311617851257324 }, { "auxiliary_loss_clip": 0.0116168, "auxiliary_loss_mlp": 0.01031926, "balance_loss_clip": 1.04730868, "balance_loss_mlp": 1.02468967, "epoch": 0.5373654782660975, "flos": 15377024623200.0, "grad_norm": 2.0537498006134203, "language_loss": 0.84077656, "learning_rate": 1.8552379741739873e-06, "loss": 0.86271262, "num_input_tokens_seen": 96491335, "step": 4469, "time_per_iteration": 2.4708502292633057 }, { "auxiliary_loss_clip": 0.01044688, "auxiliary_loss_mlp": 0.0075302, "balance_loss_clip": 1.01559901, "balance_loss_mlp": 0.99983215, "epoch": 0.5374857211567367, "flos": 69000101946720.0, "grad_norm": 0.9012597188028867, "language_loss": 0.55688059, "learning_rate": 1.8544610567265935e-06, "loss": 0.57485771, "num_input_tokens_seen": 96545275, "step": 4470, "time_per_iteration": 3.1267526149749756 }, { "auxiliary_loss_clip": 0.01155003, "auxiliary_loss_mlp": 0.00762783, "balance_loss_clip": 1.05368614, "balance_loss_mlp": 1.00036156, "epoch": 0.5376059640473757, "flos": 15085358180640.0, "grad_norm": 1.8224564870296625, "language_loss": 0.83337647, "learning_rate": 1.853684161357341e-06, "loss": 0.85255432, "num_input_tokens_seen": 96562935, "step": 4471, "time_per_iteration": 2.5121047496795654 }, { "auxiliary_loss_clip": 0.01164671, "auxiliary_loss_mlp": 0.00762595, "balance_loss_clip": 1.0513587, "balance_loss_mlp": 1.00026023, "epoch": 0.5377262069380148, "flos": 19792601038080.0, "grad_norm": 1.6703114942887323, "language_loss": 0.76740313, "learning_rate": 1.852907288184085e-06, "loss": 0.78667581, "num_input_tokens_seen": 96581820, "step": 4472, "time_per_iteration": 2.479194402694702 }, { "auxiliary_loss_clip": 0.01129404, "auxiliary_loss_mlp": 0.01027068, "balance_loss_clip": 1.04803622, "balance_loss_mlp": 1.01908159, "epoch": 0.5378464498286539, "flos": 30003081545280.0, "grad_norm": 2.2155807855123615, "language_loss": 0.69932276, "learning_rate": 1.8521304373246762e-06, "loss": 0.72088748, "num_input_tokens_seen": 96602865, "step": 4473, "time_per_iteration": 2.6594223976135254 }, { "auxiliary_loss_clip": 0.01171022, "auxiliary_loss_mlp": 0.01034036, "balance_loss_clip": 1.05215001, "balance_loss_mlp": 1.02545333, "epoch": 0.537966692719293, "flos": 21251220586560.0, "grad_norm": 2.862776246688931, "language_loss": 0.88376796, "learning_rate": 1.8513536088969626e-06, "loss": 0.90581858, "num_input_tokens_seen": 96620530, "step": 4474, "time_per_iteration": 2.479975938796997 }, { "auxiliary_loss_clip": 0.01173811, "auxiliary_loss_mlp": 0.01028698, "balance_loss_clip": 1.05662394, "balance_loss_mlp": 1.02120566, "epoch": 0.538086935609932, "flos": 21543174364800.0, "grad_norm": 1.819604712648578, "language_loss": 0.80426043, "learning_rate": 1.8505768030187884e-06, "loss": 0.82628548, "num_input_tokens_seen": 96640660, "step": 4475, "time_per_iteration": 2.4890153408050537 }, { "auxiliary_loss_clip": 0.01154504, "auxiliary_loss_mlp": 0.01028765, "balance_loss_clip": 1.05424452, "balance_loss_mlp": 1.02119517, "epoch": 0.5382071785005712, "flos": 22747223302080.0, "grad_norm": 1.5113235451721474, "language_loss": 0.80235386, "learning_rate": 1.849800019807995e-06, "loss": 0.82418656, "num_input_tokens_seen": 96661885, "step": 4476, "time_per_iteration": 2.531810998916626 }, { "auxiliary_loss_clip": 0.01140355, "auxiliary_loss_mlp": 0.01030096, "balance_loss_clip": 1.05136609, "balance_loss_mlp": 1.02298522, "epoch": 0.5383274213912103, "flos": 24934577953440.0, "grad_norm": 2.0267725415137914, "language_loss": 0.70576018, "learning_rate": 1.8490232593824186e-06, "loss": 0.72746468, "num_input_tokens_seen": 96678340, "step": 4477, "time_per_iteration": 2.560688018798828 }, { "auxiliary_loss_clip": 0.01150904, "auxiliary_loss_mlp": 0.01023523, "balance_loss_clip": 1.05139637, "balance_loss_mlp": 1.01634645, "epoch": 0.5384476642818493, "flos": 22310190558720.0, "grad_norm": 1.5968810064116328, "language_loss": 0.85044718, "learning_rate": 1.8482465218598935e-06, "loss": 0.87219143, "num_input_tokens_seen": 96698285, "step": 4478, "time_per_iteration": 2.5161447525024414 }, { "auxiliary_loss_clip": 0.01140362, "auxiliary_loss_mlp": 0.01036065, "balance_loss_clip": 1.04966211, "balance_loss_mlp": 1.02836967, "epoch": 0.5385679071724885, "flos": 22711025394720.0, "grad_norm": 1.6990872582435756, "language_loss": 0.83042067, "learning_rate": 1.8474698073582508e-06, "loss": 0.85218501, "num_input_tokens_seen": 96719655, "step": 4479, "time_per_iteration": 2.6061758995056152 }, { "auxiliary_loss_clip": 0.01143971, "auxiliary_loss_mlp": 0.01032676, "balance_loss_clip": 1.04755759, "balance_loss_mlp": 1.02433455, "epoch": 0.5386881500631275, "flos": 15953748787680.0, "grad_norm": 2.288399348075206, "language_loss": 0.87137687, "learning_rate": 1.8466931159953166e-06, "loss": 0.8931433, "num_input_tokens_seen": 96736290, "step": 4480, "time_per_iteration": 2.547316074371338 }, { "auxiliary_loss_clip": 0.01158788, "auxiliary_loss_mlp": 0.01031836, "balance_loss_clip": 1.05382717, "balance_loss_mlp": 1.02382565, "epoch": 0.5388083929537666, "flos": 24060045546240.0, "grad_norm": 1.6743944239367483, "language_loss": 0.84044397, "learning_rate": 1.8459164478889158e-06, "loss": 0.86235029, "num_input_tokens_seen": 96757685, "step": 4481, "time_per_iteration": 2.5812387466430664 }, { "auxiliary_loss_clip": 0.01127772, "auxiliary_loss_mlp": 0.01028948, "balance_loss_clip": 1.04343212, "balance_loss_mlp": 1.02133608, "epoch": 0.5389286358444056, "flos": 22236896820000.0, "grad_norm": 1.7876856811740105, "language_loss": 0.75891519, "learning_rate": 1.8451398031568663e-06, "loss": 0.78048241, "num_input_tokens_seen": 96777310, "step": 4482, "time_per_iteration": 2.5908989906311035 }, { "auxiliary_loss_clip": 0.01141432, "auxiliary_loss_mlp": 0.01028147, "balance_loss_clip": 1.05113304, "balance_loss_mlp": 1.01950419, "epoch": 0.5390488787350448, "flos": 24281722610400.0, "grad_norm": 1.6346658995542769, "language_loss": 0.74502188, "learning_rate": 1.844363181916986e-06, "loss": 0.76671761, "num_input_tokens_seen": 96798035, "step": 4483, "time_per_iteration": 3.368490219116211 }, { "auxiliary_loss_clip": 0.01164745, "auxiliary_loss_mlp": 0.01027737, "balance_loss_clip": 1.0501523, "balance_loss_mlp": 1.01949406, "epoch": 0.5391691216256839, "flos": 16581396813120.0, "grad_norm": 2.10097818459996, "language_loss": 0.83091593, "learning_rate": 1.8435865842870868e-06, "loss": 0.85284072, "num_input_tokens_seen": 96815975, "step": 4484, "time_per_iteration": 3.227917194366455 }, { "auxiliary_loss_clip": 0.01143933, "auxiliary_loss_mlp": 0.00762596, "balance_loss_clip": 1.04528058, "balance_loss_mlp": 1.00037789, "epoch": 0.5392893645163229, "flos": 23330053349760.0, "grad_norm": 1.8957000596057645, "language_loss": 0.71713603, "learning_rate": 1.8428100103849787e-06, "loss": 0.73620141, "num_input_tokens_seen": 96835770, "step": 4485, "time_per_iteration": 2.551851749420166 }, { "auxiliary_loss_clip": 0.01153959, "auxiliary_loss_mlp": 0.01031619, "balance_loss_clip": 1.05328178, "balance_loss_mlp": 1.02390671, "epoch": 0.5394096074069621, "flos": 15669804491520.0, "grad_norm": 2.1502277648265076, "language_loss": 0.72821307, "learning_rate": 1.842033460328467e-06, "loss": 0.75006878, "num_input_tokens_seen": 96854490, "step": 4486, "time_per_iteration": 2.5069868564605713 }, { "auxiliary_loss_clip": 0.01156665, "auxiliary_loss_mlp": 0.00762571, "balance_loss_clip": 1.04925704, "balance_loss_mlp": 1.00031805, "epoch": 0.5395298502976011, "flos": 22893451609920.0, "grad_norm": 1.9738344410757764, "language_loss": 0.75043035, "learning_rate": 1.8412569342353541e-06, "loss": 0.76962268, "num_input_tokens_seen": 96874645, "step": 4487, "time_per_iteration": 2.5525474548339844 }, { "auxiliary_loss_clip": 0.01160758, "auxiliary_loss_mlp": 0.01031037, "balance_loss_clip": 1.05414414, "balance_loss_mlp": 1.02267468, "epoch": 0.5396500931882402, "flos": 23842139762880.0, "grad_norm": 1.9576247804807647, "language_loss": 0.84635687, "learning_rate": 1.840480432223438e-06, "loss": 0.86827481, "num_input_tokens_seen": 96893650, "step": 4488, "time_per_iteration": 2.5451931953430176 }, { "auxiliary_loss_clip": 0.01155386, "auxiliary_loss_mlp": 0.01026243, "balance_loss_clip": 1.04886496, "balance_loss_mlp": 1.01819646, "epoch": 0.5397703360788794, "flos": 26322992621760.0, "grad_norm": 2.099891869796701, "language_loss": 0.78063428, "learning_rate": 1.8397039544105131e-06, "loss": 0.80245066, "num_input_tokens_seen": 96912735, "step": 4489, "time_per_iteration": 3.506901264190674 }, { "auxiliary_loss_clip": 0.01146397, "auxiliary_loss_mlp": 0.01033766, "balance_loss_clip": 1.0452944, "balance_loss_mlp": 1.02589238, "epoch": 0.5398905789695184, "flos": 21214591675680.0, "grad_norm": 1.6409253235779193, "language_loss": 0.69548965, "learning_rate": 1.8389275009143711e-06, "loss": 0.71729124, "num_input_tokens_seen": 96932475, "step": 4490, "time_per_iteration": 2.5461764335632324 }, { "auxiliary_loss_clip": 0.01179022, "auxiliary_loss_mlp": 0.01030338, "balance_loss_clip": 1.05143619, "balance_loss_mlp": 1.02319789, "epoch": 0.5400108218601575, "flos": 25080339340800.0, "grad_norm": 2.7671946453261755, "language_loss": 0.73578727, "learning_rate": 1.8381510718527988e-06, "loss": 0.75788087, "num_input_tokens_seen": 96952085, "step": 4491, "time_per_iteration": 3.1670947074890137 }, { "auxiliary_loss_clip": 0.01155984, "auxiliary_loss_mlp": 0.01025124, "balance_loss_clip": 1.04852414, "balance_loss_mlp": 1.01754498, "epoch": 0.5401310647507966, "flos": 26357502432000.0, "grad_norm": 1.8121501131867186, "language_loss": 0.63477898, "learning_rate": 1.8373746673435812e-06, "loss": 0.6565901, "num_input_tokens_seen": 96973110, "step": 4492, "time_per_iteration": 2.549323558807373 }, { "auxiliary_loss_clip": 0.01186094, "auxiliary_loss_mlp": 0.01031896, "balance_loss_clip": 1.05668545, "balance_loss_mlp": 1.02427208, "epoch": 0.5402513076414357, "flos": 27855336912480.0, "grad_norm": 1.6793685145490196, "language_loss": 0.79199922, "learning_rate": 1.8365982875044964e-06, "loss": 0.81417912, "num_input_tokens_seen": 96993420, "step": 4493, "time_per_iteration": 2.5017194747924805 }, { "auxiliary_loss_clip": 0.01172741, "auxiliary_loss_mlp": 0.00763522, "balance_loss_clip": 1.05241323, "balance_loss_mlp": 1.00033164, "epoch": 0.5403715505320748, "flos": 22893774862560.0, "grad_norm": 1.9430863649714007, "language_loss": 0.76312411, "learning_rate": 1.8358219324533217e-06, "loss": 0.7824868, "num_input_tokens_seen": 97013685, "step": 4494, "time_per_iteration": 2.4984166622161865 }, { "auxiliary_loss_clip": 0.0114851, "auxiliary_loss_mlp": 0.01024629, "balance_loss_clip": 1.04830122, "balance_loss_mlp": 1.01782203, "epoch": 0.5404917934227139, "flos": 30224148021120.0, "grad_norm": 1.59996317792238, "language_loss": 0.70252335, "learning_rate": 1.8350456023078292e-06, "loss": 0.72425479, "num_input_tokens_seen": 97036060, "step": 4495, "time_per_iteration": 2.5826103687286377 }, { "auxiliary_loss_clip": 0.01186906, "auxiliary_loss_mlp": 0.01029432, "balance_loss_clip": 1.05381131, "balance_loss_mlp": 1.02083683, "epoch": 0.540612036313353, "flos": 19938506093280.0, "grad_norm": 2.4415925277065926, "language_loss": 0.78364003, "learning_rate": 1.8342692971857874e-06, "loss": 0.80580342, "num_input_tokens_seen": 97055260, "step": 4496, "time_per_iteration": 2.456369400024414 }, { "auxiliary_loss_clip": 0.01153664, "auxiliary_loss_mlp": 0.0103037, "balance_loss_clip": 1.0518012, "balance_loss_mlp": 1.02240682, "epoch": 0.540732279203992, "flos": 24279603509760.0, "grad_norm": 2.4165786889871024, "language_loss": 0.71372038, "learning_rate": 1.833493017204962e-06, "loss": 0.73556072, "num_input_tokens_seen": 97075365, "step": 4497, "time_per_iteration": 2.5310826301574707 }, { "auxiliary_loss_clip": 0.0117967, "auxiliary_loss_mlp": 0.01031432, "balance_loss_clip": 1.05159962, "balance_loss_mlp": 1.02320695, "epoch": 0.5408525220946312, "flos": 20193220372320.0, "grad_norm": 1.7369930185957934, "language_loss": 0.7823298, "learning_rate": 1.8327167624831134e-06, "loss": 0.80444086, "num_input_tokens_seen": 97093095, "step": 4498, "time_per_iteration": 2.4674322605133057 }, { "auxiliary_loss_clip": 0.01180207, "auxiliary_loss_mlp": 0.01027753, "balance_loss_clip": 1.05333638, "balance_loss_mlp": 1.02009392, "epoch": 0.5409727649852702, "flos": 24134452710720.0, "grad_norm": 1.6528234405072557, "language_loss": 0.70704764, "learning_rate": 1.831940533137999e-06, "loss": 0.72912717, "num_input_tokens_seen": 97112000, "step": 4499, "time_per_iteration": 2.474181652069092 }, { "auxiliary_loss_clip": 0.01166403, "auxiliary_loss_mlp": 0.01029638, "balance_loss_clip": 1.05324948, "balance_loss_mlp": 1.02180028, "epoch": 0.5410930078759093, "flos": 23912703812640.0, "grad_norm": 1.6517740707569153, "language_loss": 0.72533536, "learning_rate": 1.8311643292873718e-06, "loss": 0.7472958, "num_input_tokens_seen": 97130820, "step": 4500, "time_per_iteration": 2.5099613666534424 }, { "auxiliary_loss_clip": 0.01163061, "auxiliary_loss_mlp": 0.01028168, "balance_loss_clip": 1.05018878, "balance_loss_mlp": 1.02116466, "epoch": 0.5412132507665485, "flos": 21105135946560.0, "grad_norm": 2.3810275978443913, "language_loss": 0.88106406, "learning_rate": 1.8303881510489818e-06, "loss": 0.90297639, "num_input_tokens_seen": 97149210, "step": 4501, "time_per_iteration": 2.483950614929199 }, { "auxiliary_loss_clip": 0.01155563, "auxiliary_loss_mlp": 0.01028837, "balance_loss_clip": 1.05144477, "balance_loss_mlp": 1.02114153, "epoch": 0.5413334936571875, "flos": 30227344630560.0, "grad_norm": 2.096227142042472, "language_loss": 0.69195724, "learning_rate": 1.829611998540574e-06, "loss": 0.71380121, "num_input_tokens_seen": 97170415, "step": 4502, "time_per_iteration": 2.5858943462371826 }, { "auxiliary_loss_clip": 0.01168918, "auxiliary_loss_mlp": 0.00762989, "balance_loss_clip": 1.05093384, "balance_loss_mlp": 1.00032055, "epoch": 0.5414537365478266, "flos": 24279639426720.0, "grad_norm": 1.6657380597185838, "language_loss": 0.79437649, "learning_rate": 1.8288358718798914e-06, "loss": 0.81369555, "num_input_tokens_seen": 97189605, "step": 4503, "time_per_iteration": 2.5018670558929443 }, { "auxiliary_loss_clip": 0.01165851, "auxiliary_loss_mlp": 0.0076269, "balance_loss_clip": 1.05459213, "balance_loss_mlp": 1.0003742, "epoch": 0.5415739794384657, "flos": 16654546884000.0, "grad_norm": 1.6166771426625932, "language_loss": 0.72526741, "learning_rate": 1.8280597711846703e-06, "loss": 0.74455273, "num_input_tokens_seen": 97207845, "step": 4504, "time_per_iteration": 2.484806537628174 }, { "auxiliary_loss_clip": 0.01166643, "auxiliary_loss_mlp": 0.01032127, "balance_loss_clip": 1.05274916, "balance_loss_mlp": 1.02468276, "epoch": 0.5416942223291048, "flos": 23185728640800.0, "grad_norm": 1.8980658475421754, "language_loss": 0.83355147, "learning_rate": 1.8272836965726455e-06, "loss": 0.8555392, "num_input_tokens_seen": 97226780, "step": 4505, "time_per_iteration": 2.493708372116089 }, { "auxiliary_loss_clip": 0.01108533, "auxiliary_loss_mlp": 0.0103283, "balance_loss_clip": 1.04274368, "balance_loss_mlp": 1.02449751, "epoch": 0.5418144652197439, "flos": 20303250772800.0, "grad_norm": 1.6462317751812328, "language_loss": 0.78334123, "learning_rate": 1.8265076481615461e-06, "loss": 0.80475485, "num_input_tokens_seen": 97246695, "step": 4506, "time_per_iteration": 2.6407880783081055 }, { "auxiliary_loss_clip": 0.01153491, "auxiliary_loss_mlp": 0.01028921, "balance_loss_clip": 1.05325866, "balance_loss_mlp": 1.02074897, "epoch": 0.541934708110383, "flos": 12458636183520.0, "grad_norm": 3.0514145260126058, "language_loss": 0.87148565, "learning_rate": 1.8257316260690987e-06, "loss": 0.89330983, "num_input_tokens_seen": 97264480, "step": 4507, "time_per_iteration": 2.49899959564209 }, { "auxiliary_loss_clip": 0.01165599, "auxiliary_loss_mlp": 0.01031374, "balance_loss_clip": 1.05210614, "balance_loss_mlp": 1.02388763, "epoch": 0.5420549510010221, "flos": 21253806607680.0, "grad_norm": 1.5003419585975646, "language_loss": 0.76069653, "learning_rate": 1.8249556304130254e-06, "loss": 0.78266621, "num_input_tokens_seen": 97285760, "step": 4508, "time_per_iteration": 2.5251686573028564 }, { "auxiliary_loss_clip": 0.01142926, "auxiliary_loss_mlp": 0.0102992, "balance_loss_clip": 1.04689574, "balance_loss_mlp": 1.02191222, "epoch": 0.5421751938916611, "flos": 29490528211680.0, "grad_norm": 2.0069208504379876, "language_loss": 0.69042802, "learning_rate": 1.824179661311044e-06, "loss": 0.71215653, "num_input_tokens_seen": 97304510, "step": 4509, "time_per_iteration": 3.339434862136841 }, { "auxiliary_loss_clip": 0.0112597, "auxiliary_loss_mlp": 0.01027699, "balance_loss_clip": 1.04428315, "balance_loss_mlp": 1.01974225, "epoch": 0.5422954367823003, "flos": 18734241654240.0, "grad_norm": 1.9126665267667449, "language_loss": 0.80052149, "learning_rate": 1.823403718880868e-06, "loss": 0.8220582, "num_input_tokens_seen": 97323270, "step": 4510, "time_per_iteration": 3.3283324241638184 }, { "auxiliary_loss_clip": 0.01153621, "auxiliary_loss_mlp": 0.01029157, "balance_loss_clip": 1.04716897, "balance_loss_mlp": 1.0212059, "epoch": 0.5424156796729394, "flos": 39969012273120.0, "grad_norm": 1.6138148258739187, "language_loss": 0.66465437, "learning_rate": 1.822627803240207e-06, "loss": 0.68648213, "num_input_tokens_seen": 97345600, "step": 4511, "time_per_iteration": 2.6767802238464355 }, { "auxiliary_loss_clip": 0.01140944, "auxiliary_loss_mlp": 0.01025984, "balance_loss_clip": 1.04779387, "balance_loss_mlp": 1.01819336, "epoch": 0.5425359225635784, "flos": 11546541024480.0, "grad_norm": 2.1658114938050943, "language_loss": 0.8493197, "learning_rate": 1.8218519145067675e-06, "loss": 0.87098897, "num_input_tokens_seen": 97361220, "step": 4512, "time_per_iteration": 2.506682872772217 }, { "auxiliary_loss_clip": 0.01132956, "auxiliary_loss_mlp": 0.01031256, "balance_loss_clip": 1.04648781, "balance_loss_mlp": 1.02397847, "epoch": 0.5426561654542175, "flos": 20229705615360.0, "grad_norm": 1.7508909212106323, "language_loss": 0.89367008, "learning_rate": 1.8210760527982508e-06, "loss": 0.91531217, "num_input_tokens_seen": 97381505, "step": 4513, "time_per_iteration": 2.5502448081970215 }, { "auxiliary_loss_clip": 0.01157067, "auxiliary_loss_mlp": 0.00761886, "balance_loss_clip": 1.05452478, "balance_loss_mlp": 1.00027895, "epoch": 0.5427764083448566, "flos": 21871685220000.0, "grad_norm": 1.86123082797642, "language_loss": 0.75346935, "learning_rate": 1.8203002182323552e-06, "loss": 0.77265882, "num_input_tokens_seen": 97399060, "step": 4514, "time_per_iteration": 2.529646396636963 }, { "auxiliary_loss_clip": 0.01157118, "auxiliary_loss_mlp": 0.01028269, "balance_loss_clip": 1.0523138, "balance_loss_mlp": 1.020401, "epoch": 0.5428966512354957, "flos": 19640949269280.0, "grad_norm": 1.9366403365040081, "language_loss": 0.75383413, "learning_rate": 1.819524410926773e-06, "loss": 0.77568805, "num_input_tokens_seen": 97416740, "step": 4515, "time_per_iteration": 3.2036292552948 }, { "auxiliary_loss_clip": 0.01108143, "auxiliary_loss_mlp": 0.01029061, "balance_loss_clip": 1.04606855, "balance_loss_mlp": 1.02099609, "epoch": 0.5430168941261347, "flos": 22382191286880.0, "grad_norm": 1.4077946447279752, "language_loss": 0.76886755, "learning_rate": 1.8187486309991944e-06, "loss": 0.79023957, "num_input_tokens_seen": 97437620, "step": 4516, "time_per_iteration": 2.631969451904297 }, { "auxiliary_loss_clip": 0.01173681, "auxiliary_loss_mlp": 0.01033879, "balance_loss_clip": 1.05513608, "balance_loss_mlp": 1.0261215, "epoch": 0.5431371370167739, "flos": 18764190010560.0, "grad_norm": 2.4886148773419645, "language_loss": 0.77517033, "learning_rate": 1.817972878567304e-06, "loss": 0.79724598, "num_input_tokens_seen": 97456275, "step": 4517, "time_per_iteration": 2.469712972640991 }, { "auxiliary_loss_clip": 0.01158724, "auxiliary_loss_mlp": 0.01035626, "balance_loss_clip": 1.04975486, "balance_loss_mlp": 1.02793074, "epoch": 0.543257379907413, "flos": 18806026880640.0, "grad_norm": 1.7065706157010194, "language_loss": 0.75998491, "learning_rate": 1.8171971537487834e-06, "loss": 0.78192842, "num_input_tokens_seen": 97474925, "step": 4518, "time_per_iteration": 3.279888391494751 }, { "auxiliary_loss_clip": 0.01184062, "auxiliary_loss_mlp": 0.01027625, "balance_loss_clip": 1.05403912, "balance_loss_mlp": 1.01996565, "epoch": 0.543377622798052, "flos": 17493384221280.0, "grad_norm": 1.7282555239602322, "language_loss": 0.80736768, "learning_rate": 1.8164214566613093e-06, "loss": 0.82948446, "num_input_tokens_seen": 97493550, "step": 4519, "time_per_iteration": 2.468061923980713 }, { "auxiliary_loss_clip": 0.01181326, "auxiliary_loss_mlp": 0.01030649, "balance_loss_clip": 1.05438173, "balance_loss_mlp": 1.02290332, "epoch": 0.5434978656886912, "flos": 18989315102880.0, "grad_norm": 2.9282429755532324, "language_loss": 0.65663886, "learning_rate": 1.8156457874225547e-06, "loss": 0.67875856, "num_input_tokens_seen": 97512010, "step": 4520, "time_per_iteration": 2.4591846466064453 }, { "auxiliary_loss_clip": 0.01149493, "auxiliary_loss_mlp": 0.01031438, "balance_loss_clip": 1.05268395, "balance_loss_mlp": 1.02376711, "epoch": 0.5436181085793302, "flos": 17274939683520.0, "grad_norm": 1.703305487045463, "language_loss": 0.80560184, "learning_rate": 1.814870146150187e-06, "loss": 0.82741117, "num_input_tokens_seen": 97530120, "step": 4521, "time_per_iteration": 2.4832561016082764 }, { "auxiliary_loss_clip": 0.01156482, "auxiliary_loss_mlp": 0.010249, "balance_loss_clip": 1.04787683, "balance_loss_mlp": 1.01670456, "epoch": 0.5437383514699693, "flos": 19098591247200.0, "grad_norm": 1.9531658106233405, "language_loss": 0.78550333, "learning_rate": 1.814094532961871e-06, "loss": 0.80731714, "num_input_tokens_seen": 97548695, "step": 4522, "time_per_iteration": 2.513141632080078 }, { "auxiliary_loss_clip": 0.01126161, "auxiliary_loss_mlp": 0.01029259, "balance_loss_clip": 1.04407835, "balance_loss_mlp": 1.02133131, "epoch": 0.5438585943606085, "flos": 22602719008320.0, "grad_norm": 1.7378841518395132, "language_loss": 0.83214235, "learning_rate": 1.8133189479752666e-06, "loss": 0.85369658, "num_input_tokens_seen": 97567625, "step": 4523, "time_per_iteration": 2.5923190116882324 }, { "auxiliary_loss_clip": 0.01182649, "auxiliary_loss_mlp": 0.01029656, "balance_loss_clip": 1.05491257, "balance_loss_mlp": 1.02242327, "epoch": 0.5439788372512475, "flos": 21798499232160.0, "grad_norm": 2.0068677161108344, "language_loss": 0.81768662, "learning_rate": 1.8125433913080292e-06, "loss": 0.83980966, "num_input_tokens_seen": 97585325, "step": 4524, "time_per_iteration": 2.501343011856079 }, { "auxiliary_loss_clip": 0.01061939, "auxiliary_loss_mlp": 0.01027234, "balance_loss_clip": 1.0401125, "balance_loss_mlp": 1.01977468, "epoch": 0.5440990801418866, "flos": 16399365684480.0, "grad_norm": 1.9514409230800407, "language_loss": 0.82569087, "learning_rate": 1.811767863077811e-06, "loss": 0.84658259, "num_input_tokens_seen": 97604275, "step": 4525, "time_per_iteration": 2.842409610748291 }, { "auxiliary_loss_clip": 0.01106337, "auxiliary_loss_mlp": 0.01034623, "balance_loss_clip": 1.04848135, "balance_loss_mlp": 1.02747655, "epoch": 0.5442193230325257, "flos": 21615642013440.0, "grad_norm": 1.5660917101336196, "language_loss": 0.78150338, "learning_rate": 1.8109923634022577e-06, "loss": 0.80291295, "num_input_tokens_seen": 97624300, "step": 4526, "time_per_iteration": 2.8210620880126953 }, { "auxiliary_loss_clip": 0.01184086, "auxiliary_loss_mlp": 0.01029945, "balance_loss_clip": 1.05370152, "balance_loss_mlp": 1.02222586, "epoch": 0.5443395659231648, "flos": 15481200559200.0, "grad_norm": 2.1303434322163666, "language_loss": 0.86503923, "learning_rate": 1.8102168923990128e-06, "loss": 0.88717949, "num_input_tokens_seen": 97637845, "step": 4527, "time_per_iteration": 2.41377854347229 }, { "auxiliary_loss_clip": 0.0117346, "auxiliary_loss_mlp": 0.00762141, "balance_loss_clip": 1.05485153, "balance_loss_mlp": 1.0002985, "epoch": 0.5444598088138038, "flos": 18770439561600.0, "grad_norm": 2.188456938922454, "language_loss": 0.79859364, "learning_rate": 1.809441450185714e-06, "loss": 0.81794971, "num_input_tokens_seen": 97656330, "step": 4528, "time_per_iteration": 2.5006253719329834 }, { "auxiliary_loss_clip": 0.01158083, "auxiliary_loss_mlp": 0.01029742, "balance_loss_clip": 1.04749632, "balance_loss_mlp": 1.02213597, "epoch": 0.544580051704443, "flos": 21142339528800.0, "grad_norm": 2.0405304689199957, "language_loss": 0.73556912, "learning_rate": 1.8086660368799958e-06, "loss": 0.75744736, "num_input_tokens_seen": 97674380, "step": 4529, "time_per_iteration": 2.52693510055542 }, { "auxiliary_loss_clip": 0.01157216, "auxiliary_loss_mlp": 0.01035029, "balance_loss_clip": 1.05303693, "balance_loss_mlp": 1.02666688, "epoch": 0.5447002945950821, "flos": 32491512882720.0, "grad_norm": 2.1596519584383014, "language_loss": 0.77316928, "learning_rate": 1.807890652599488e-06, "loss": 0.79509175, "num_input_tokens_seen": 97698765, "step": 4530, "time_per_iteration": 2.6087968349456787 }, { "auxiliary_loss_clip": 0.01178917, "auxiliary_loss_mlp": 0.01029332, "balance_loss_clip": 1.05411935, "balance_loss_mlp": 1.02227521, "epoch": 0.5448205374857211, "flos": 11798310112800.0, "grad_norm": 1.9903570571501108, "language_loss": 0.82982326, "learning_rate": 1.8071152974618156e-06, "loss": 0.85190582, "num_input_tokens_seen": 97716565, "step": 4531, "time_per_iteration": 2.4319210052490234 }, { "auxiliary_loss_clip": 0.0114152, "auxiliary_loss_mlp": 0.00761973, "balance_loss_clip": 1.0470655, "balance_loss_mlp": 1.00022471, "epoch": 0.5449407803763603, "flos": 24133770288480.0, "grad_norm": 2.013158114532229, "language_loss": 0.78223681, "learning_rate": 1.806339971584599e-06, "loss": 0.8012718, "num_input_tokens_seen": 97733225, "step": 4532, "time_per_iteration": 2.578399658203125 }, { "auxiliary_loss_clip": 0.01181646, "auxiliary_loss_mlp": 0.01023309, "balance_loss_clip": 1.0533185, "balance_loss_mlp": 1.01582909, "epoch": 0.5450610232669993, "flos": 23258555459040.0, "grad_norm": 1.6338854968884535, "language_loss": 0.85304344, "learning_rate": 1.8055646750854546e-06, "loss": 0.87509298, "num_input_tokens_seen": 97752735, "step": 4533, "time_per_iteration": 2.466252326965332 }, { "auxiliary_loss_clip": 0.01159589, "auxiliary_loss_mlp": 0.01026135, "balance_loss_clip": 1.05270207, "balance_loss_mlp": 1.01860082, "epoch": 0.5451812661576384, "flos": 17785086580800.0, "grad_norm": 2.631348210790543, "language_loss": 0.81756818, "learning_rate": 1.8047894080819945e-06, "loss": 0.83942544, "num_input_tokens_seen": 97769985, "step": 4534, "time_per_iteration": 2.4948573112487793 }, { "auxiliary_loss_clip": 0.01079102, "auxiliary_loss_mlp": 0.01001675, "balance_loss_clip": 1.01760507, "balance_loss_mlp": 1.00050056, "epoch": 0.5453015090482776, "flos": 71062598881440.0, "grad_norm": 0.7504844093021976, "language_loss": 0.63259584, "learning_rate": 1.8040141706918258e-06, "loss": 0.65340358, "num_input_tokens_seen": 97831225, "step": 4535, "time_per_iteration": 4.148670673370361 }, { "auxiliary_loss_clip": 0.01154703, "auxiliary_loss_mlp": 0.01030439, "balance_loss_clip": 1.05283654, "balance_loss_mlp": 1.02276182, "epoch": 0.5454217519389166, "flos": 25552205147040.0, "grad_norm": 1.655900762166184, "language_loss": 0.77018452, "learning_rate": 1.8032389630325525e-06, "loss": 0.79203594, "num_input_tokens_seen": 97849975, "step": 4536, "time_per_iteration": 3.4979844093322754 }, { "auxiliary_loss_clip": 0.01154019, "auxiliary_loss_mlp": 0.01029045, "balance_loss_clip": 1.04920042, "balance_loss_mlp": 1.02167809, "epoch": 0.5455419948295557, "flos": 23658348703200.0, "grad_norm": 1.6953478198590652, "language_loss": 0.75834203, "learning_rate": 1.8024637852217707e-06, "loss": 0.78017271, "num_input_tokens_seen": 97869700, "step": 4537, "time_per_iteration": 2.5622661113739014 }, { "auxiliary_loss_clip": 0.0115538, "auxiliary_loss_mlp": 0.01030873, "balance_loss_clip": 1.05351377, "balance_loss_mlp": 1.0236398, "epoch": 0.5456622377201948, "flos": 23403993593760.0, "grad_norm": 1.7788372609630276, "language_loss": 0.84546804, "learning_rate": 1.8016886373770766e-06, "loss": 0.86733055, "num_input_tokens_seen": 97888215, "step": 4538, "time_per_iteration": 2.5261292457580566 }, { "auxiliary_loss_clip": 0.01150889, "auxiliary_loss_mlp": 0.01032971, "balance_loss_clip": 1.04936314, "balance_loss_mlp": 1.0252521, "epoch": 0.5457824806108339, "flos": 23988044818080.0, "grad_norm": 2.442995765039039, "language_loss": 0.78593081, "learning_rate": 1.8009135196160579e-06, "loss": 0.80776942, "num_input_tokens_seen": 97907090, "step": 4539, "time_per_iteration": 2.5362846851348877 }, { "auxiliary_loss_clip": 0.0113693, "auxiliary_loss_mlp": 0.0102718, "balance_loss_clip": 1.04899561, "balance_loss_mlp": 1.01959813, "epoch": 0.545902723501473, "flos": 22565874595680.0, "grad_norm": 1.7353331630643059, "language_loss": 0.84039974, "learning_rate": 1.8001384320563e-06, "loss": 0.86204088, "num_input_tokens_seen": 97927345, "step": 4540, "time_per_iteration": 3.3294026851654053 }, { "auxiliary_loss_clip": 0.01079039, "auxiliary_loss_mlp": 0.01001379, "balance_loss_clip": 1.01780844, "balance_loss_mlp": 1.00013924, "epoch": 0.5460229663921121, "flos": 55198407827040.0, "grad_norm": 0.7735939925833387, "language_loss": 0.57784337, "learning_rate": 1.7993633748153833e-06, "loss": 0.59864753, "num_input_tokens_seen": 97981950, "step": 4541, "time_per_iteration": 2.966094732284546 }, { "auxiliary_loss_clip": 0.01171148, "auxiliary_loss_mlp": 0.01034736, "balance_loss_clip": 1.0516715, "balance_loss_mlp": 1.02719653, "epoch": 0.5461432092827512, "flos": 15413869035840.0, "grad_norm": 1.6997384192659923, "language_loss": 0.72595882, "learning_rate": 1.7985883480108834e-06, "loss": 0.74801767, "num_input_tokens_seen": 97999585, "step": 4542, "time_per_iteration": 2.5137646198272705 }, { "auxiliary_loss_clip": 0.01163873, "auxiliary_loss_mlp": 0.01025024, "balance_loss_clip": 1.05194044, "balance_loss_mlp": 1.0174005, "epoch": 0.5462634521733902, "flos": 24024925147680.0, "grad_norm": 1.532560365127204, "language_loss": 0.72411036, "learning_rate": 1.797813351760371e-06, "loss": 0.74599934, "num_input_tokens_seen": 98021290, "step": 4543, "time_per_iteration": 3.2686662673950195 }, { "auxiliary_loss_clip": 0.01185017, "auxiliary_loss_mlp": 0.01036667, "balance_loss_clip": 1.0551188, "balance_loss_mlp": 1.02915072, "epoch": 0.5463836950640293, "flos": 22820948044320.0, "grad_norm": 1.601897264004088, "language_loss": 0.7809217, "learning_rate": 1.7970383861814116e-06, "loss": 0.80313849, "num_input_tokens_seen": 98041060, "step": 4544, "time_per_iteration": 2.4774701595306396 }, { "auxiliary_loss_clip": 0.01171395, "auxiliary_loss_mlp": 0.01025324, "balance_loss_clip": 1.0550257, "balance_loss_mlp": 1.01742065, "epoch": 0.5465039379546685, "flos": 20448293820960.0, "grad_norm": 2.185797919405552, "language_loss": 0.73975098, "learning_rate": 1.7962634513915684e-06, "loss": 0.76171815, "num_input_tokens_seen": 98058410, "step": 4545, "time_per_iteration": 2.4716708660125732 }, { "auxiliary_loss_clip": 0.01180888, "auxiliary_loss_mlp": 0.01034258, "balance_loss_clip": 1.0537107, "balance_loss_mlp": 1.0272423, "epoch": 0.5466241808453075, "flos": 17343312798720.0, "grad_norm": 1.7868847298836332, "language_loss": 0.79374659, "learning_rate": 1.7954885475083969e-06, "loss": 0.815898, "num_input_tokens_seen": 98076080, "step": 4546, "time_per_iteration": 2.441497325897217 }, { "auxiliary_loss_clip": 0.01187298, "auxiliary_loss_mlp": 0.01029374, "balance_loss_clip": 1.05576384, "balance_loss_mlp": 1.02170253, "epoch": 0.5467444237359466, "flos": 21617042774880.0, "grad_norm": 1.9671670857288241, "language_loss": 0.72586811, "learning_rate": 1.7947136746494513e-06, "loss": 0.74803483, "num_input_tokens_seen": 98096995, "step": 4547, "time_per_iteration": 2.4693338871002197 }, { "auxiliary_loss_clip": 0.01167489, "auxiliary_loss_mlp": 0.01025806, "balance_loss_clip": 1.05318439, "balance_loss_mlp": 1.01803994, "epoch": 0.5468646666265857, "flos": 24170471033280.0, "grad_norm": 1.7745285852576222, "language_loss": 0.87748808, "learning_rate": 1.793938832932277e-06, "loss": 0.89942098, "num_input_tokens_seen": 98115105, "step": 4548, "time_per_iteration": 2.5283422470092773 }, { "auxiliary_loss_clip": 0.01179814, "auxiliary_loss_mlp": 0.01028518, "balance_loss_clip": 1.05163658, "balance_loss_mlp": 1.02091205, "epoch": 0.5469849095172248, "flos": 27527005642080.0, "grad_norm": 1.9326132183823848, "language_loss": 0.7037524, "learning_rate": 1.7931640224744185e-06, "loss": 0.72583568, "num_input_tokens_seen": 98135655, "step": 4549, "time_per_iteration": 2.4904673099517822 }, { "auxiliary_loss_clip": 0.01125668, "auxiliary_loss_mlp": 0.01032272, "balance_loss_clip": 1.04239154, "balance_loss_mlp": 1.02449989, "epoch": 0.5471051524078638, "flos": 27964684890720.0, "grad_norm": 1.5338805811301632, "language_loss": 0.73347068, "learning_rate": 1.7923892433934127e-06, "loss": 0.75505006, "num_input_tokens_seen": 98156730, "step": 4550, "time_per_iteration": 2.609875202178955 }, { "auxiliary_loss_clip": 0.01156647, "auxiliary_loss_mlp": 0.00762961, "balance_loss_clip": 1.05186772, "balance_loss_mlp": 1.00033557, "epoch": 0.547225395298503, "flos": 18150513682560.0, "grad_norm": 1.6941786993869092, "language_loss": 0.78732121, "learning_rate": 1.7916144958067939e-06, "loss": 0.80651724, "num_input_tokens_seen": 98174590, "step": 4551, "time_per_iteration": 2.501072406768799 }, { "auxiliary_loss_clip": 0.01170098, "auxiliary_loss_mlp": 0.01030862, "balance_loss_clip": 1.05388546, "balance_loss_mlp": 1.02246976, "epoch": 0.5473456381891421, "flos": 21361502405760.0, "grad_norm": 1.7850305143419478, "language_loss": 0.78982317, "learning_rate": 1.7908397798320905e-06, "loss": 0.81183279, "num_input_tokens_seen": 98194325, "step": 4552, "time_per_iteration": 2.501448154449463 }, { "auxiliary_loss_clip": 0.01170075, "auxiliary_loss_mlp": 0.00762693, "balance_loss_clip": 1.05236292, "balance_loss_mlp": 1.00027168, "epoch": 0.5474658810797811, "flos": 19932148791360.0, "grad_norm": 1.7895718020952465, "language_loss": 0.7480467, "learning_rate": 1.7900650955868265e-06, "loss": 0.76737434, "num_input_tokens_seen": 98213970, "step": 4553, "time_per_iteration": 2.4860212802886963 }, { "auxiliary_loss_clip": 0.01168681, "auxiliary_loss_mlp": 0.00762618, "balance_loss_clip": 1.05471992, "balance_loss_mlp": 1.00028718, "epoch": 0.5475861239704203, "flos": 50476223514240.0, "grad_norm": 1.4968426813209952, "language_loss": 0.76717949, "learning_rate": 1.7892904431885202e-06, "loss": 0.78649253, "num_input_tokens_seen": 98241145, "step": 4554, "time_per_iteration": 2.7391035556793213 }, { "auxiliary_loss_clip": 0.01123326, "auxiliary_loss_mlp": 0.01031638, "balance_loss_clip": 1.04376912, "balance_loss_mlp": 1.02426803, "epoch": 0.5477063668610593, "flos": 20705127200640.0, "grad_norm": 1.5995407083471982, "language_loss": 0.75124383, "learning_rate": 1.788515822754686e-06, "loss": 0.77279341, "num_input_tokens_seen": 98261565, "step": 4555, "time_per_iteration": 2.5830941200256348 }, { "auxiliary_loss_clip": 0.01137214, "auxiliary_loss_mlp": 0.01030394, "balance_loss_clip": 1.04467928, "balance_loss_mlp": 1.0219835, "epoch": 0.5478266097516984, "flos": 19609743819360.0, "grad_norm": 1.8934804045716727, "language_loss": 0.78104335, "learning_rate": 1.7877412344028335e-06, "loss": 0.80271941, "num_input_tokens_seen": 98281370, "step": 4556, "time_per_iteration": 2.570585012435913 }, { "auxiliary_loss_clip": 0.01170102, "auxiliary_loss_mlp": 0.01026302, "balance_loss_clip": 1.0503118, "balance_loss_mlp": 1.01849365, "epoch": 0.5479468526423376, "flos": 12896602767840.0, "grad_norm": 2.449149125392658, "language_loss": 0.77420294, "learning_rate": 1.7869666782504668e-06, "loss": 0.79616702, "num_input_tokens_seen": 98297950, "step": 4557, "time_per_iteration": 2.4682838916778564 }, { "auxiliary_loss_clip": 0.01141094, "auxiliary_loss_mlp": 0.01026735, "balance_loss_clip": 1.04637611, "balance_loss_mlp": 1.01958823, "epoch": 0.5480670955329766, "flos": 18588803519520.0, "grad_norm": 1.8227723301754377, "language_loss": 0.68598419, "learning_rate": 1.7861921544150867e-06, "loss": 0.70766246, "num_input_tokens_seen": 98316800, "step": 4558, "time_per_iteration": 2.5309336185455322 }, { "auxiliary_loss_clip": 0.01100452, "auxiliary_loss_mlp": 0.00762369, "balance_loss_clip": 1.04644299, "balance_loss_mlp": 1.00027955, "epoch": 0.5481873384236157, "flos": 15954610794720.0, "grad_norm": 1.832996551021055, "language_loss": 0.7643857, "learning_rate": 1.7854176630141856e-06, "loss": 0.78301382, "num_input_tokens_seen": 98333935, "step": 4559, "time_per_iteration": 2.6166388988494873 }, { "auxiliary_loss_clip": 0.01183803, "auxiliary_loss_mlp": 0.01029095, "balance_loss_clip": 1.05342889, "balance_loss_mlp": 1.02142072, "epoch": 0.5483075813142548, "flos": 22783816296000.0, "grad_norm": 2.243140800547559, "language_loss": 0.84195501, "learning_rate": 1.784643204165255e-06, "loss": 0.86408401, "num_input_tokens_seen": 98353255, "step": 4560, "time_per_iteration": 2.4909732341766357 }, { "auxiliary_loss_clip": 0.01163965, "auxiliary_loss_mlp": 0.0102571, "balance_loss_clip": 1.05364084, "balance_loss_mlp": 1.01805115, "epoch": 0.5484278242048939, "flos": 19317215369760.0, "grad_norm": 2.013880972794062, "language_loss": 0.77052754, "learning_rate": 1.7838687779857783e-06, "loss": 0.79242432, "num_input_tokens_seen": 98371130, "step": 4561, "time_per_iteration": 3.2372946739196777 }, { "auxiliary_loss_clip": 0.0114287, "auxiliary_loss_mlp": 0.01025386, "balance_loss_clip": 1.04671121, "balance_loss_mlp": 1.0180012, "epoch": 0.5485480670955329, "flos": 22816027420800.0, "grad_norm": 2.4574704270093983, "language_loss": 0.64021736, "learning_rate": 1.7830943845932366e-06, "loss": 0.66189992, "num_input_tokens_seen": 98390455, "step": 4562, "time_per_iteration": 3.294511318206787 }, { "auxiliary_loss_clip": 0.01158803, "auxiliary_loss_mlp": 0.010254, "balance_loss_clip": 1.05227959, "balance_loss_mlp": 1.01789224, "epoch": 0.5486683099861721, "flos": 22671307625280.0, "grad_norm": 1.6831326052692803, "language_loss": 0.75405514, "learning_rate": 1.7823200241051044e-06, "loss": 0.77589715, "num_input_tokens_seen": 98409370, "step": 4563, "time_per_iteration": 2.5374772548675537 }, { "auxiliary_loss_clip": 0.01183195, "auxiliary_loss_mlp": 0.01031233, "balance_loss_clip": 1.05497503, "balance_loss_mlp": 1.0239197, "epoch": 0.5487885528768112, "flos": 23149387065600.0, "grad_norm": 1.8758093585498061, "language_loss": 0.80663478, "learning_rate": 1.7815456966388513e-06, "loss": 0.82877898, "num_input_tokens_seen": 98428465, "step": 4564, "time_per_iteration": 2.456374168395996 }, { "auxiliary_loss_clip": 0.01136615, "auxiliary_loss_mlp": 0.01024145, "balance_loss_clip": 1.04730511, "balance_loss_mlp": 1.01581192, "epoch": 0.5489087957674502, "flos": 22053931850400.0, "grad_norm": 2.1224422993963636, "language_loss": 0.80646253, "learning_rate": 1.780771402311943e-06, "loss": 0.8280701, "num_input_tokens_seen": 98447300, "step": 4565, "time_per_iteration": 2.5562398433685303 }, { "auxiliary_loss_clip": 0.0115703, "auxiliary_loss_mlp": 0.01029343, "balance_loss_clip": 1.05536973, "balance_loss_mlp": 1.02109981, "epoch": 0.5490290386580894, "flos": 24315981001920.0, "grad_norm": 1.8498680601464759, "language_loss": 0.78733933, "learning_rate": 1.7799971412418374e-06, "loss": 0.80920303, "num_input_tokens_seen": 98468695, "step": 4566, "time_per_iteration": 3.3034801483154297 }, { "auxiliary_loss_clip": 0.01137905, "auxiliary_loss_mlp": 0.01027387, "balance_loss_clip": 1.0498426, "balance_loss_mlp": 1.01888108, "epoch": 0.5491492815487284, "flos": 18294946142400.0, "grad_norm": 2.626745488346446, "language_loss": 0.73992163, "learning_rate": 1.7792229135459918e-06, "loss": 0.76157457, "num_input_tokens_seen": 98485345, "step": 4567, "time_per_iteration": 2.5278403759002686 }, { "auxiliary_loss_clip": 0.01038805, "auxiliary_loss_mlp": 0.01008936, "balance_loss_clip": 1.02827811, "balance_loss_mlp": 1.00751173, "epoch": 0.5492695244393675, "flos": 64550267140320.0, "grad_norm": 0.7342323895977921, "language_loss": 0.61697328, "learning_rate": 1.7784487193418538e-06, "loss": 0.6374507, "num_input_tokens_seen": 98543195, "step": 4568, "time_per_iteration": 3.4124176502227783 }, { "auxiliary_loss_clip": 0.01122052, "auxiliary_loss_mlp": 0.01030316, "balance_loss_clip": 1.04288554, "balance_loss_mlp": 1.02244806, "epoch": 0.5493897673300067, "flos": 17379582540000.0, "grad_norm": 2.6401876864283733, "language_loss": 0.61277783, "learning_rate": 1.7776745587468698e-06, "loss": 0.63430154, "num_input_tokens_seen": 98560620, "step": 4569, "time_per_iteration": 3.28023624420166 }, { "auxiliary_loss_clip": 0.01182293, "auxiliary_loss_mlp": 0.01028777, "balance_loss_clip": 1.05207455, "balance_loss_mlp": 1.0205518, "epoch": 0.5495100102206457, "flos": 19901769431520.0, "grad_norm": 2.3296437552311238, "language_loss": 0.82019806, "learning_rate": 1.7769004318784776e-06, "loss": 0.8423087, "num_input_tokens_seen": 98578265, "step": 4570, "time_per_iteration": 2.47818922996521 }, { "auxiliary_loss_clip": 0.01168717, "auxiliary_loss_mlp": 0.01034801, "balance_loss_clip": 1.05287123, "balance_loss_mlp": 1.02718902, "epoch": 0.5496302531112848, "flos": 16727194117440.0, "grad_norm": 1.7607649297381358, "language_loss": 0.80647624, "learning_rate": 1.776126338854113e-06, "loss": 0.82851142, "num_input_tokens_seen": 98596055, "step": 4571, "time_per_iteration": 2.450101852416992 }, { "auxiliary_loss_clip": 0.011644, "auxiliary_loss_mlp": 0.01026116, "balance_loss_clip": 1.0530436, "balance_loss_mlp": 1.01834989, "epoch": 0.5497504960019239, "flos": 24572347461120.0, "grad_norm": 1.6111609210507272, "language_loss": 0.84406346, "learning_rate": 1.7753522797912044e-06, "loss": 0.86596859, "num_input_tokens_seen": 98616140, "step": 4572, "time_per_iteration": 2.543290376663208 }, { "auxiliary_loss_clip": 0.01159324, "auxiliary_loss_mlp": 0.01030711, "balance_loss_clip": 1.05006957, "balance_loss_mlp": 1.02315342, "epoch": 0.549870738892563, "flos": 15450497946720.0, "grad_norm": 2.641612562956146, "language_loss": 0.69910151, "learning_rate": 1.7745782548071765e-06, "loss": 0.72100186, "num_input_tokens_seen": 98633035, "step": 4573, "time_per_iteration": 2.488131523132324 }, { "auxiliary_loss_clip": 0.01136701, "auxiliary_loss_mlp": 0.01026244, "balance_loss_clip": 1.05249202, "balance_loss_mlp": 1.01930881, "epoch": 0.549990981783202, "flos": 21069117624000.0, "grad_norm": 1.6581709227825676, "language_loss": 0.73957455, "learning_rate": 1.7738042640194482e-06, "loss": 0.761204, "num_input_tokens_seen": 98652700, "step": 4574, "time_per_iteration": 2.558712959289551 }, { "auxiliary_loss_clip": 0.01180551, "auxiliary_loss_mlp": 0.01024479, "balance_loss_clip": 1.05243492, "balance_loss_mlp": 1.01686144, "epoch": 0.5501112246738411, "flos": 21395904465120.0, "grad_norm": 1.628529426630771, "language_loss": 0.70500451, "learning_rate": 1.7730303075454335e-06, "loss": 0.72705483, "num_input_tokens_seen": 98671590, "step": 4575, "time_per_iteration": 2.4553470611572266 }, { "auxiliary_loss_clip": 0.01142483, "auxiliary_loss_mlp": 0.01034322, "balance_loss_clip": 1.04694939, "balance_loss_mlp": 1.02657926, "epoch": 0.5502314675644803, "flos": 17456934895200.0, "grad_norm": 1.9281726605436766, "language_loss": 0.84696925, "learning_rate": 1.7722563855025402e-06, "loss": 0.86873728, "num_input_tokens_seen": 98689620, "step": 4576, "time_per_iteration": 2.542964458465576 }, { "auxiliary_loss_clip": 0.01153592, "auxiliary_loss_mlp": 0.01030794, "balance_loss_clip": 1.04674625, "balance_loss_mlp": 1.02315903, "epoch": 0.5503517104551193, "flos": 24310413873120.0, "grad_norm": 2.605787483472324, "language_loss": 0.70587492, "learning_rate": 1.7714824980081721e-06, "loss": 0.72771871, "num_input_tokens_seen": 98708915, "step": 4577, "time_per_iteration": 2.5685102939605713 }, { "auxiliary_loss_clip": 0.01164368, "auxiliary_loss_mlp": 0.01028392, "balance_loss_clip": 1.05386078, "balance_loss_mlp": 1.02085781, "epoch": 0.5504719533457584, "flos": 22419430786080.0, "grad_norm": 2.2426873606537625, "language_loss": 0.73637486, "learning_rate": 1.7707086451797276e-06, "loss": 0.75830245, "num_input_tokens_seen": 98729790, "step": 4578, "time_per_iteration": 2.545152425765991 }, { "auxiliary_loss_clip": 0.01049595, "auxiliary_loss_mlp": 0.01001017, "balance_loss_clip": 1.02047276, "balance_loss_mlp": 0.99973601, "epoch": 0.5505921962363975, "flos": 67294167012960.0, "grad_norm": 0.7010915505264341, "language_loss": 0.52349591, "learning_rate": 1.7699348271345993e-06, "loss": 0.54400206, "num_input_tokens_seen": 98792415, "step": 4579, "time_per_iteration": 3.1119425296783447 }, { "auxiliary_loss_clip": 0.01045997, "auxiliary_loss_mlp": 0.01002182, "balance_loss_clip": 1.02132034, "balance_loss_mlp": 1.00086486, "epoch": 0.5507124391270366, "flos": 45685133720160.0, "grad_norm": 0.7065087495343857, "language_loss": 0.54473293, "learning_rate": 1.7691610439901753e-06, "loss": 0.56521463, "num_input_tokens_seen": 98855350, "step": 4580, "time_per_iteration": 3.226086378097534 }, { "auxiliary_loss_clip": 0.01169338, "auxiliary_loss_mlp": 0.01024236, "balance_loss_clip": 1.05332017, "balance_loss_mlp": 1.01696718, "epoch": 0.5508326820176757, "flos": 22273848983520.0, "grad_norm": 1.7710869310391326, "language_loss": 0.75766027, "learning_rate": 1.7683872958638367e-06, "loss": 0.77959597, "num_input_tokens_seen": 98874230, "step": 4581, "time_per_iteration": 2.522345781326294 }, { "auxiliary_loss_clip": 0.01150336, "auxiliary_loss_mlp": 0.01029547, "balance_loss_clip": 1.04826164, "balance_loss_mlp": 1.0218457, "epoch": 0.5509529249083148, "flos": 20012446337280.0, "grad_norm": 2.110688569755603, "language_loss": 0.84532851, "learning_rate": 1.7676135828729614e-06, "loss": 0.8671273, "num_input_tokens_seen": 98893940, "step": 4582, "time_per_iteration": 2.547192335128784 }, { "auxiliary_loss_clip": 0.01167009, "auxiliary_loss_mlp": 0.01028356, "balance_loss_clip": 1.05374217, "balance_loss_mlp": 1.02041078, "epoch": 0.5510731677989539, "flos": 21834804890400.0, "grad_norm": 1.8754450842721864, "language_loss": 0.82860196, "learning_rate": 1.7668399051349205e-06, "loss": 0.85055566, "num_input_tokens_seen": 98913620, "step": 4583, "time_per_iteration": 2.5187606811523438 }, { "auxiliary_loss_clip": 0.01138014, "auxiliary_loss_mlp": 0.01029121, "balance_loss_clip": 1.04927564, "balance_loss_mlp": 1.0213604, "epoch": 0.5511934106895929, "flos": 21467905193280.0, "grad_norm": 1.7924341735221392, "language_loss": 0.82752115, "learning_rate": 1.766066262767081e-06, "loss": 0.84919244, "num_input_tokens_seen": 98931460, "step": 4584, "time_per_iteration": 2.5392379760742188 }, { "auxiliary_loss_clip": 0.01146717, "auxiliary_loss_mlp": 0.01028003, "balance_loss_clip": 1.04987025, "balance_loss_mlp": 1.02072811, "epoch": 0.5513136535802321, "flos": 21068938039200.0, "grad_norm": 2.1742256086064495, "language_loss": 0.7711612, "learning_rate": 1.765292655886803e-06, "loss": 0.79290831, "num_input_tokens_seen": 98950105, "step": 4585, "time_per_iteration": 2.524495840072632 }, { "auxiliary_loss_clip": 0.01147597, "auxiliary_loss_mlp": 0.01031191, "balance_loss_clip": 1.04987144, "balance_loss_mlp": 1.023839, "epoch": 0.5514338964708712, "flos": 27815008554720.0, "grad_norm": 2.322263208326984, "language_loss": 0.70731091, "learning_rate": 1.764519084611443e-06, "loss": 0.7290988, "num_input_tokens_seen": 98970560, "step": 4586, "time_per_iteration": 2.595081090927124 }, { "auxiliary_loss_clip": 0.01153412, "auxiliary_loss_mlp": 0.01027564, "balance_loss_clip": 1.05004573, "balance_loss_mlp": 1.02001178, "epoch": 0.5515541393615102, "flos": 21908529632640.0, "grad_norm": 1.7696927633791215, "language_loss": 0.7798273, "learning_rate": 1.7637455490583505e-06, "loss": 0.80163705, "num_input_tokens_seen": 98989885, "step": 4587, "time_per_iteration": 3.291539192199707 }, { "auxiliary_loss_clip": 0.01167547, "auxiliary_loss_mlp": 0.01035784, "balance_loss_clip": 1.05282784, "balance_loss_mlp": 1.02872658, "epoch": 0.5516743822521494, "flos": 20485425569280.0, "grad_norm": 1.9878543213521298, "language_loss": 0.76932448, "learning_rate": 1.7629720493448701e-06, "loss": 0.79135776, "num_input_tokens_seen": 99007180, "step": 4588, "time_per_iteration": 3.2492239475250244 }, { "auxiliary_loss_clip": 0.01159555, "auxiliary_loss_mlp": 0.01031486, "balance_loss_clip": 1.04974556, "balance_loss_mlp": 1.02381182, "epoch": 0.5517946251427884, "flos": 14940386966400.0, "grad_norm": 1.6400846543289014, "language_loss": 0.85042119, "learning_rate": 1.7621985855883418e-06, "loss": 0.87233162, "num_input_tokens_seen": 99023880, "step": 4589, "time_per_iteration": 2.5019426345825195 }, { "auxiliary_loss_clip": 0.01147521, "auxiliary_loss_mlp": 0.01025271, "balance_loss_clip": 1.04918921, "balance_loss_mlp": 1.01789212, "epoch": 0.5519148680334275, "flos": 18404868792000.0, "grad_norm": 2.070359683003653, "language_loss": 0.7271502, "learning_rate": 1.7614251579060983e-06, "loss": 0.74887812, "num_input_tokens_seen": 99042475, "step": 4590, "time_per_iteration": 2.4869439601898193 }, { "auxiliary_loss_clip": 0.01139736, "auxiliary_loss_mlp": 0.01027655, "balance_loss_clip": 1.04999554, "balance_loss_mlp": 1.01984727, "epoch": 0.5520351109240667, "flos": 25113340638720.0, "grad_norm": 1.8017582970278545, "language_loss": 0.84361428, "learning_rate": 1.76065176641547e-06, "loss": 0.8652882, "num_input_tokens_seen": 99065185, "step": 4591, "time_per_iteration": 2.6125736236572266 }, { "auxiliary_loss_clip": 0.01169031, "auxiliary_loss_mlp": 0.01028706, "balance_loss_clip": 1.05147123, "balance_loss_mlp": 1.02212834, "epoch": 0.5521553538147057, "flos": 21069548627520.0, "grad_norm": 1.992899783483276, "language_loss": 0.77545893, "learning_rate": 1.759878411233777e-06, "loss": 0.79743624, "num_input_tokens_seen": 99083645, "step": 4592, "time_per_iteration": 2.485550880432129 }, { "auxiliary_loss_clip": 0.01169118, "auxiliary_loss_mlp": 0.01028518, "balance_loss_clip": 1.05390692, "balance_loss_mlp": 1.02100444, "epoch": 0.5522755967053448, "flos": 18880003041600.0, "grad_norm": 2.4227484012834584, "language_loss": 0.75625759, "learning_rate": 1.7591050924783388e-06, "loss": 0.77823389, "num_input_tokens_seen": 99100835, "step": 4593, "time_per_iteration": 3.286653995513916 }, { "auxiliary_loss_clip": 0.01039471, "auxiliary_loss_mlp": 0.01001444, "balance_loss_clip": 1.02030063, "balance_loss_mlp": 1.00031698, "epoch": 0.5523958395959839, "flos": 64675634082720.0, "grad_norm": 0.8346652953439568, "language_loss": 0.57990396, "learning_rate": 1.7583318102664661e-06, "loss": 0.60031307, "num_input_tokens_seen": 99168400, "step": 4594, "time_per_iteration": 3.226811170578003 }, { "auxiliary_loss_clip": 0.01170944, "auxiliary_loss_mlp": 0.01025472, "balance_loss_clip": 1.04987621, "balance_loss_mlp": 1.01765764, "epoch": 0.552516082486623, "flos": 10889734815840.0, "grad_norm": 2.321033057498115, "language_loss": 0.78672385, "learning_rate": 1.757558564715466e-06, "loss": 0.80868793, "num_input_tokens_seen": 99186475, "step": 4595, "time_per_iteration": 3.2033987045288086 }, { "auxiliary_loss_clip": 0.01168905, "auxiliary_loss_mlp": 0.01027516, "balance_loss_clip": 1.05191851, "balance_loss_mlp": 1.01951456, "epoch": 0.552636325377262, "flos": 22199800988640.0, "grad_norm": 2.44448217659781, "language_loss": 0.74090993, "learning_rate": 1.7567853559426386e-06, "loss": 0.76287413, "num_input_tokens_seen": 99203525, "step": 4596, "time_per_iteration": 2.488661289215088 }, { "auxiliary_loss_clip": 0.01168962, "auxiliary_loss_mlp": 0.01029613, "balance_loss_clip": 1.05189037, "balance_loss_mlp": 1.0224576, "epoch": 0.5527565682679012, "flos": 23988188485920.0, "grad_norm": 1.8134326799593892, "language_loss": 0.75085539, "learning_rate": 1.7560121840652797e-06, "loss": 0.7728411, "num_input_tokens_seen": 99222910, "step": 4597, "time_per_iteration": 2.5199434757232666 }, { "auxiliary_loss_clip": 0.01125561, "auxiliary_loss_mlp": 0.01029507, "balance_loss_clip": 1.04561138, "balance_loss_mlp": 1.02182436, "epoch": 0.5528768111585403, "flos": 19719271382400.0, "grad_norm": 1.793154978073295, "language_loss": 0.69169855, "learning_rate": 1.7552390492006782e-06, "loss": 0.71324921, "num_input_tokens_seen": 99241230, "step": 4598, "time_per_iteration": 2.5465593338012695 }, { "auxiliary_loss_clip": 0.0113246, "auxiliary_loss_mlp": 0.00762745, "balance_loss_clip": 1.04645085, "balance_loss_mlp": 1.00032854, "epoch": 0.5529970540491793, "flos": 26215979245920.0, "grad_norm": 1.8083692966523477, "language_loss": 0.65241086, "learning_rate": 1.7544659514661184e-06, "loss": 0.67136288, "num_input_tokens_seen": 99264320, "step": 4599, "time_per_iteration": 2.6798739433288574 }, { "auxiliary_loss_clip": 0.01147745, "auxiliary_loss_mlp": 0.01026097, "balance_loss_clip": 1.04806423, "balance_loss_mlp": 1.01845527, "epoch": 0.5531172969398185, "flos": 24425975485440.0, "grad_norm": 1.9024197187585243, "language_loss": 0.79533237, "learning_rate": 1.7536928909788786e-06, "loss": 0.81707078, "num_input_tokens_seen": 99283625, "step": 4600, "time_per_iteration": 2.596637010574341 }, { "auxiliary_loss_clip": 0.0103664, "auxiliary_loss_mlp": 0.01002487, "balance_loss_clip": 1.01651597, "balance_loss_mlp": 1.00122356, "epoch": 0.5532375398304575, "flos": 64907325590400.0, "grad_norm": 0.8791730945884983, "language_loss": 0.61954385, "learning_rate": 1.752919867856231e-06, "loss": 0.63993508, "num_input_tokens_seen": 99335270, "step": 4601, "time_per_iteration": 3.0153841972351074 }, { "auxiliary_loss_clip": 0.0114162, "auxiliary_loss_mlp": 0.01027458, "balance_loss_clip": 1.04528308, "balance_loss_mlp": 1.01898265, "epoch": 0.5533577827210966, "flos": 19683109392000.0, "grad_norm": 1.6823600688204015, "language_loss": 0.78882086, "learning_rate": 1.7521468822154436e-06, "loss": 0.81051159, "num_input_tokens_seen": 99354185, "step": 4602, "time_per_iteration": 2.5211129188537598 }, { "auxiliary_loss_clip": 0.01147212, "auxiliary_loss_mlp": 0.01028709, "balance_loss_clip": 1.05094814, "balance_loss_mlp": 1.02138019, "epoch": 0.5534780256117358, "flos": 32306500646400.0, "grad_norm": 1.8645282535092698, "language_loss": 0.7520771, "learning_rate": 1.751373934173777e-06, "loss": 0.77383637, "num_input_tokens_seen": 99376930, "step": 4603, "time_per_iteration": 2.633695602416992 }, { "auxiliary_loss_clip": 0.01183637, "auxiliary_loss_mlp": 0.0102629, "balance_loss_clip": 1.05361855, "balance_loss_mlp": 1.01841068, "epoch": 0.5535982685023748, "flos": 23222429385600.0, "grad_norm": 1.7601963596200794, "language_loss": 0.73001647, "learning_rate": 1.750601023848487e-06, "loss": 0.75211573, "num_input_tokens_seen": 99397655, "step": 4604, "time_per_iteration": 2.4933791160583496 }, { "auxiliary_loss_clip": 0.01177378, "auxiliary_loss_mlp": 0.00762089, "balance_loss_clip": 1.05124617, "balance_loss_mlp": 1.00025618, "epoch": 0.5537185113930139, "flos": 24352538078880.0, "grad_norm": 2.0352001983548074, "language_loss": 0.73933303, "learning_rate": 1.749828151356823e-06, "loss": 0.75872773, "num_input_tokens_seen": 99417850, "step": 4605, "time_per_iteration": 2.5133845806121826 }, { "auxiliary_loss_clip": 0.01152228, "auxiliary_loss_mlp": 0.01026387, "balance_loss_clip": 1.04991579, "balance_loss_mlp": 1.01847124, "epoch": 0.553838754283653, "flos": 23549072558880.0, "grad_norm": 1.6733311078798252, "language_loss": 0.75699651, "learning_rate": 1.7490553168160297e-06, "loss": 0.77878261, "num_input_tokens_seen": 99438920, "step": 4606, "time_per_iteration": 2.5642964839935303 }, { "auxiliary_loss_clip": 0.01150945, "auxiliary_loss_mlp": 0.01029169, "balance_loss_clip": 1.04920435, "balance_loss_mlp": 1.02136636, "epoch": 0.5539589971742921, "flos": 17275047434400.0, "grad_norm": 2.0722238433793194, "language_loss": 0.76415431, "learning_rate": 1.748282520343345e-06, "loss": 0.78595543, "num_input_tokens_seen": 99457950, "step": 4607, "time_per_iteration": 2.5275027751922607 }, { "auxiliary_loss_clip": 0.01174679, "auxiliary_loss_mlp": 0.01030663, "balance_loss_clip": 1.05249405, "balance_loss_mlp": 1.02317691, "epoch": 0.5540792400649311, "flos": 27564173307360.0, "grad_norm": 1.6417205111758642, "language_loss": 0.78762734, "learning_rate": 1.7475097620560023e-06, "loss": 0.8096807, "num_input_tokens_seen": 99478015, "step": 4608, "time_per_iteration": 2.5603256225585938 }, { "auxiliary_loss_clip": 0.01179934, "auxiliary_loss_mlp": 0.01028759, "balance_loss_clip": 1.05292916, "balance_loss_mlp": 1.02160692, "epoch": 0.5541994829555702, "flos": 23878660922880.0, "grad_norm": 1.6605143205833293, "language_loss": 0.71124607, "learning_rate": 1.746737042071228e-06, "loss": 0.73333299, "num_input_tokens_seen": 99496520, "step": 4609, "time_per_iteration": 2.5010476112365723 }, { "auxiliary_loss_clip": 0.01147932, "auxiliary_loss_mlp": 0.01026578, "balance_loss_clip": 1.0500586, "balance_loss_mlp": 1.01921976, "epoch": 0.5543197258462094, "flos": 20115724349280.0, "grad_norm": 1.8831693302819061, "language_loss": 0.79109824, "learning_rate": 1.7459643605062424e-06, "loss": 0.81284326, "num_input_tokens_seen": 99513780, "step": 4610, "time_per_iteration": 2.506307363510132 }, { "auxiliary_loss_clip": 0.01123631, "auxiliary_loss_mlp": 0.01026055, "balance_loss_clip": 1.04940772, "balance_loss_mlp": 1.01850891, "epoch": 0.5544399687368484, "flos": 20916567931200.0, "grad_norm": 1.6717590851795467, "language_loss": 0.80972683, "learning_rate": 1.745191717478262e-06, "loss": 0.83122367, "num_input_tokens_seen": 99532360, "step": 4611, "time_per_iteration": 2.599390745162964 }, { "auxiliary_loss_clip": 0.01146056, "auxiliary_loss_mlp": 0.01023636, "balance_loss_clip": 1.04896021, "balance_loss_mlp": 1.01608372, "epoch": 0.5545602116274875, "flos": 25518665094720.0, "grad_norm": 1.6708212372080866, "language_loss": 0.79299116, "learning_rate": 1.7444191131044948e-06, "loss": 0.81468809, "num_input_tokens_seen": 99552635, "step": 4612, "time_per_iteration": 2.590104818344116 }, { "auxiliary_loss_clip": 0.01154343, "auxiliary_loss_mlp": 0.01031556, "balance_loss_clip": 1.05304646, "balance_loss_mlp": 1.0242722, "epoch": 0.5546804545181266, "flos": 20995572466560.0, "grad_norm": 1.696013434523613, "language_loss": 0.7268573, "learning_rate": 1.7436465475021456e-06, "loss": 0.74871629, "num_input_tokens_seen": 99572685, "step": 4613, "time_per_iteration": 2.540205955505371 }, { "auxiliary_loss_clip": 0.01127594, "auxiliary_loss_mlp": 0.01023968, "balance_loss_clip": 1.04750991, "balance_loss_mlp": 1.01627588, "epoch": 0.5548006974087657, "flos": 26833642356480.0, "grad_norm": 2.5006461680496406, "language_loss": 0.71411312, "learning_rate": 1.7428740207884111e-06, "loss": 0.73562872, "num_input_tokens_seen": 99593565, "step": 4614, "time_per_iteration": 3.546694278717041 }, { "auxiliary_loss_clip": 0.01126939, "auxiliary_loss_mlp": 0.01025146, "balance_loss_clip": 1.0510143, "balance_loss_mlp": 1.01747501, "epoch": 0.5549209402994048, "flos": 33656418721920.0, "grad_norm": 3.3649561097363283, "language_loss": 0.61179554, "learning_rate": 1.7421015330804833e-06, "loss": 0.63331646, "num_input_tokens_seen": 99613485, "step": 4615, "time_per_iteration": 3.619595766067505 }, { "auxiliary_loss_clip": 0.01178712, "auxiliary_loss_mlp": 0.01027605, "balance_loss_clip": 1.05174255, "balance_loss_mlp": 1.02025533, "epoch": 0.5550411831900439, "flos": 23769528446400.0, "grad_norm": 1.9791505101047118, "language_loss": 0.72662467, "learning_rate": 1.7413290844955475e-06, "loss": 0.74868786, "num_input_tokens_seen": 99633515, "step": 4616, "time_per_iteration": 2.5401973724365234 }, { "auxiliary_loss_clip": 0.01158569, "auxiliary_loss_mlp": 0.01039261, "balance_loss_clip": 1.05092716, "balance_loss_mlp": 1.03087497, "epoch": 0.555161426080683, "flos": 21651193415520.0, "grad_norm": 3.868063291569357, "language_loss": 0.78452921, "learning_rate": 1.7405566751507843e-06, "loss": 0.80650753, "num_input_tokens_seen": 99651560, "step": 4617, "time_per_iteration": 2.4831154346466064 }, { "auxiliary_loss_clip": 0.01136877, "auxiliary_loss_mlp": 0.01028094, "balance_loss_clip": 1.04844117, "balance_loss_mlp": 1.02078605, "epoch": 0.555281668971322, "flos": 49563122680320.0, "grad_norm": 1.4442443088828971, "language_loss": 0.67472601, "learning_rate": 1.7397843051633668e-06, "loss": 0.69637573, "num_input_tokens_seen": 99674255, "step": 4618, "time_per_iteration": 3.601215362548828 }, { "auxiliary_loss_clip": 0.0116405, "auxiliary_loss_mlp": 0.01026611, "balance_loss_clip": 1.05315113, "balance_loss_mlp": 1.01930368, "epoch": 0.5554019118619612, "flos": 20741612443680.0, "grad_norm": 1.6232205564313376, "language_loss": 0.7135613, "learning_rate": 1.739011974650464e-06, "loss": 0.73546785, "num_input_tokens_seen": 99693585, "step": 4619, "time_per_iteration": 2.513585090637207 }, { "auxiliary_loss_clip": 0.0112709, "auxiliary_loss_mlp": 0.01027808, "balance_loss_clip": 1.04729676, "balance_loss_mlp": 1.01946354, "epoch": 0.5555221547526003, "flos": 25483221443520.0, "grad_norm": 1.9035863682364191, "language_loss": 0.76489758, "learning_rate": 1.7382396837292365e-06, "loss": 0.78644663, "num_input_tokens_seen": 99714045, "step": 4620, "time_per_iteration": 2.6348772048950195 }, { "auxiliary_loss_clip": 0.01185223, "auxiliary_loss_mlp": 0.01028053, "balance_loss_clip": 1.05526006, "balance_loss_mlp": 1.02022374, "epoch": 0.5556423976432393, "flos": 21762516826560.0, "grad_norm": 5.574669562863935, "language_loss": 0.73454404, "learning_rate": 1.737467432516841e-06, "loss": 0.75667673, "num_input_tokens_seen": 99734145, "step": 4621, "time_per_iteration": 3.166785717010498 }, { "auxiliary_loss_clip": 0.01152019, "auxiliary_loss_mlp": 0.01024513, "balance_loss_clip": 1.04852271, "balance_loss_mlp": 1.01708055, "epoch": 0.5557626405338785, "flos": 24900175894080.0, "grad_norm": 2.4440120367810834, "language_loss": 0.74056745, "learning_rate": 1.7366952211304274e-06, "loss": 0.76233268, "num_input_tokens_seen": 99751990, "step": 4622, "time_per_iteration": 2.5730512142181396 }, { "auxiliary_loss_clip": 0.01141987, "auxiliary_loss_mlp": 0.01030104, "balance_loss_clip": 1.04817307, "balance_loss_mlp": 1.02324665, "epoch": 0.5558828834245175, "flos": 18697504992480.0, "grad_norm": 2.0639918051374013, "language_loss": 0.83551157, "learning_rate": 1.735923049687139e-06, "loss": 0.85723245, "num_input_tokens_seen": 99768565, "step": 4623, "time_per_iteration": 2.5073304176330566 }, { "auxiliary_loss_clip": 0.01145026, "auxiliary_loss_mlp": 0.01026325, "balance_loss_clip": 1.04653716, "balance_loss_mlp": 1.01904714, "epoch": 0.5560031263151566, "flos": 27272183612160.0, "grad_norm": 1.4226596694448752, "language_loss": 0.73811817, "learning_rate": 1.7351509183041144e-06, "loss": 0.75983167, "num_input_tokens_seen": 99788895, "step": 4624, "time_per_iteration": 2.598424196243286 }, { "auxiliary_loss_clip": 0.01182691, "auxiliary_loss_mlp": 0.01029964, "balance_loss_clip": 1.054456, "balance_loss_mlp": 1.0219357, "epoch": 0.5561233692057957, "flos": 23403742175040.0, "grad_norm": 1.6100850704866825, "language_loss": 0.7177639, "learning_rate": 1.7343788270984852e-06, "loss": 0.73989046, "num_input_tokens_seen": 99808035, "step": 4625, "time_per_iteration": 2.5035383701324463 }, { "auxiliary_loss_clip": 0.01151855, "auxiliary_loss_mlp": 0.01028354, "balance_loss_clip": 1.05265045, "balance_loss_mlp": 1.02012825, "epoch": 0.5562436120964348, "flos": 37670872965120.0, "grad_norm": 1.7594376553229845, "language_loss": 0.74956095, "learning_rate": 1.7336067761873764e-06, "loss": 0.77136302, "num_input_tokens_seen": 99830460, "step": 4626, "time_per_iteration": 2.669654369354248 }, { "auxiliary_loss_clip": 0.01170644, "auxiliary_loss_mlp": 0.01035899, "balance_loss_clip": 1.05115986, "balance_loss_mlp": 1.02820396, "epoch": 0.5563638549870739, "flos": 25155249342720.0, "grad_norm": 1.8084772758406684, "language_loss": 0.76577407, "learning_rate": 1.7328347656879076e-06, "loss": 0.78783953, "num_input_tokens_seen": 99850320, "step": 4627, "time_per_iteration": 2.6216530799865723 }, { "auxiliary_loss_clip": 0.01136345, "auxiliary_loss_mlp": 0.01028951, "balance_loss_clip": 1.04627621, "balance_loss_mlp": 1.02043962, "epoch": 0.556484097877713, "flos": 13581812903520.0, "grad_norm": 2.304049008472985, "language_loss": 0.68757439, "learning_rate": 1.7320627957171927e-06, "loss": 0.70922738, "num_input_tokens_seen": 99864980, "step": 4628, "time_per_iteration": 2.536886215209961 }, { "auxiliary_loss_clip": 0.01182275, "auxiliary_loss_mlp": 0.01029351, "balance_loss_clip": 1.05578709, "balance_loss_mlp": 1.02143574, "epoch": 0.5566043407683521, "flos": 24681372186720.0, "grad_norm": 1.766551978785845, "language_loss": 0.81656259, "learning_rate": 1.7312908663923382e-06, "loss": 0.83867884, "num_input_tokens_seen": 99881155, "step": 4629, "time_per_iteration": 2.5364909172058105 }, { "auxiliary_loss_clip": 0.01158425, "auxiliary_loss_mlp": 0.0103022, "balance_loss_clip": 1.04858005, "balance_loss_mlp": 1.0225606, "epoch": 0.5567245836589911, "flos": 20588164826880.0, "grad_norm": 2.117830900192486, "language_loss": 0.67430967, "learning_rate": 1.7305189778304463e-06, "loss": 0.69619608, "num_input_tokens_seen": 99899330, "step": 4630, "time_per_iteration": 2.489550828933716 }, { "auxiliary_loss_clip": 0.0115422, "auxiliary_loss_mlp": 0.01031371, "balance_loss_clip": 1.0536623, "balance_loss_mlp": 1.02420115, "epoch": 0.5568448265496303, "flos": 20704193359680.0, "grad_norm": 1.7219151764006755, "language_loss": 0.7970733, "learning_rate": 1.729747130148611e-06, "loss": 0.81892926, "num_input_tokens_seen": 99918525, "step": 4631, "time_per_iteration": 2.559868335723877 }, { "auxiliary_loss_clip": 0.01146387, "auxiliary_loss_mlp": 0.01029824, "balance_loss_clip": 1.0522536, "balance_loss_mlp": 1.02161694, "epoch": 0.5569650694402694, "flos": 25302914328960.0, "grad_norm": 1.9906360466550752, "language_loss": 0.77002174, "learning_rate": 1.7289753234639208e-06, "loss": 0.79178381, "num_input_tokens_seen": 99937500, "step": 4632, "time_per_iteration": 2.6512787342071533 }, { "auxiliary_loss_clip": 0.01173039, "auxiliary_loss_mlp": 0.0102451, "balance_loss_clip": 1.05285573, "balance_loss_mlp": 1.01727378, "epoch": 0.5570853123309084, "flos": 19712626744800.0, "grad_norm": 1.8960682045660164, "language_loss": 0.76505864, "learning_rate": 1.7282035578934592e-06, "loss": 0.78703415, "num_input_tokens_seen": 99955665, "step": 4633, "time_per_iteration": 2.5246453285217285 }, { "auxiliary_loss_clip": 0.01146973, "auxiliary_loss_mlp": 0.0102906, "balance_loss_clip": 1.0538919, "balance_loss_mlp": 1.02042973, "epoch": 0.5572055552215476, "flos": 16108094328480.0, "grad_norm": 2.6947840211787693, "language_loss": 0.79047763, "learning_rate": 1.727431833554301e-06, "loss": 0.81223798, "num_input_tokens_seen": 99974140, "step": 4634, "time_per_iteration": 2.509178400039673 }, { "auxiliary_loss_clip": 0.01114229, "auxiliary_loss_mlp": 0.01025049, "balance_loss_clip": 1.04698849, "balance_loss_mlp": 1.01780462, "epoch": 0.5573257981121866, "flos": 17128819126560.0, "grad_norm": 1.7904392829911988, "language_loss": 0.77368498, "learning_rate": 1.7266601505635175e-06, "loss": 0.79507768, "num_input_tokens_seen": 99991480, "step": 4635, "time_per_iteration": 2.6248199939727783 }, { "auxiliary_loss_clip": 0.01167858, "auxiliary_loss_mlp": 0.01028815, "balance_loss_clip": 1.05330324, "balance_loss_mlp": 1.02038121, "epoch": 0.5574460410028257, "flos": 18807032555520.0, "grad_norm": 2.8414797378326098, "language_loss": 0.75351518, "learning_rate": 1.7258885090381717e-06, "loss": 0.77548194, "num_input_tokens_seen": 100009520, "step": 4636, "time_per_iteration": 2.4988725185394287 }, { "auxiliary_loss_clip": 0.01155119, "auxiliary_loss_mlp": 0.01022918, "balance_loss_clip": 1.05057609, "balance_loss_mlp": 1.01615572, "epoch": 0.5575662838934649, "flos": 29642682817920.0, "grad_norm": 1.8169671990316751, "language_loss": 0.78814125, "learning_rate": 1.7251169090953213e-06, "loss": 0.80992162, "num_input_tokens_seen": 100029995, "step": 4637, "time_per_iteration": 2.6067161560058594 }, { "auxiliary_loss_clip": 0.01165012, "auxiliary_loss_mlp": 0.0102603, "balance_loss_clip": 1.05120671, "balance_loss_mlp": 1.0185082, "epoch": 0.5576865267841039, "flos": 22054470604800.0, "grad_norm": 3.0915996617384636, "language_loss": 0.765724, "learning_rate": 1.7243453508520168e-06, "loss": 0.78763443, "num_input_tokens_seen": 100046980, "step": 4638, "time_per_iteration": 2.4826924800872803 }, { "auxiliary_loss_clip": 0.01152913, "auxiliary_loss_mlp": 0.01037062, "balance_loss_clip": 1.05062008, "balance_loss_mlp": 1.02916121, "epoch": 0.557806769674743, "flos": 17196042899040.0, "grad_norm": 2.002077647754594, "language_loss": 0.84710813, "learning_rate": 1.7235738344253038e-06, "loss": 0.86900783, "num_input_tokens_seen": 100060610, "step": 4639, "time_per_iteration": 2.498401403427124 }, { "auxiliary_loss_clip": 0.01165355, "auxiliary_loss_mlp": 0.01031405, "balance_loss_clip": 1.0531106, "balance_loss_mlp": 1.023525, "epoch": 0.557927012565382, "flos": 24712721304480.0, "grad_norm": 1.981570330296746, "language_loss": 0.82517886, "learning_rate": 1.72280235993222e-06, "loss": 0.84714639, "num_input_tokens_seen": 100078915, "step": 4640, "time_per_iteration": 3.2847959995269775 }, { "auxiliary_loss_clip": 0.01165916, "auxiliary_loss_mlp": 0.00762846, "balance_loss_clip": 1.0525589, "balance_loss_mlp": 1.00035143, "epoch": 0.5580472554560212, "flos": 16983093656160.0, "grad_norm": 2.047215309021577, "language_loss": 0.69446194, "learning_rate": 1.722030927489798e-06, "loss": 0.71374953, "num_input_tokens_seen": 100096195, "step": 4641, "time_per_iteration": 3.204664468765259 }, { "auxiliary_loss_clip": 0.01139349, "auxiliary_loss_mlp": 0.01028881, "balance_loss_clip": 1.0503335, "balance_loss_mlp": 1.02128434, "epoch": 0.5581674983466602, "flos": 23509103370720.0, "grad_norm": 1.9430501623224297, "language_loss": 0.74129671, "learning_rate": 1.7212595372150634e-06, "loss": 0.76297903, "num_input_tokens_seen": 100116175, "step": 4642, "time_per_iteration": 2.5903873443603516 }, { "auxiliary_loss_clip": 0.01181885, "auxiliary_loss_mlp": 0.01026932, "balance_loss_clip": 1.05529308, "balance_loss_mlp": 1.01943326, "epoch": 0.5582877412372993, "flos": 13480294822560.0, "grad_norm": 2.1093382056714787, "language_loss": 0.73129779, "learning_rate": 1.720488189225035e-06, "loss": 0.7533859, "num_input_tokens_seen": 100133875, "step": 4643, "time_per_iteration": 2.427985906600952 }, { "auxiliary_loss_clip": 0.01170847, "auxiliary_loss_mlp": 0.01025053, "balance_loss_clip": 1.05294561, "balance_loss_mlp": 1.0168519, "epoch": 0.5584079841279385, "flos": 21903609009120.0, "grad_norm": 2.2809611572020474, "language_loss": 0.7924242, "learning_rate": 1.7197168836367265e-06, "loss": 0.81438327, "num_input_tokens_seen": 100150685, "step": 4644, "time_per_iteration": 2.4785544872283936 }, { "auxiliary_loss_clip": 0.01164363, "auxiliary_loss_mlp": 0.00763068, "balance_loss_clip": 1.05115128, "balance_loss_mlp": 1.00029826, "epoch": 0.5585282270185775, "flos": 18843553715520.0, "grad_norm": 1.9256155203746832, "language_loss": 0.82029343, "learning_rate": 1.7189456205671433e-06, "loss": 0.83956766, "num_input_tokens_seen": 100169530, "step": 4645, "time_per_iteration": 3.2568323612213135 }, { "auxiliary_loss_clip": 0.01171594, "auxiliary_loss_mlp": 0.01028053, "balance_loss_clip": 1.05373597, "balance_loss_mlp": 1.02072787, "epoch": 0.5586484699092166, "flos": 21868452693600.0, "grad_norm": 1.7497744686927457, "language_loss": 0.8195504, "learning_rate": 1.7181744001332866e-06, "loss": 0.84154689, "num_input_tokens_seen": 100188140, "step": 4646, "time_per_iteration": 2.5084187984466553 }, { "auxiliary_loss_clip": 0.0117978, "auxiliary_loss_mlp": 0.01028581, "balance_loss_clip": 1.05531073, "balance_loss_mlp": 1.02189946, "epoch": 0.5587687127998557, "flos": 22893236108160.0, "grad_norm": 1.7961886651112466, "language_loss": 0.63142347, "learning_rate": 1.7174032224521493e-06, "loss": 0.65350711, "num_input_tokens_seen": 100206850, "step": 4647, "time_per_iteration": 2.4671664237976074 }, { "auxiliary_loss_clip": 0.01167447, "auxiliary_loss_mlp": 0.01022963, "balance_loss_clip": 1.05313075, "balance_loss_mlp": 1.01574767, "epoch": 0.5588889556904948, "flos": 20303071188000.0, "grad_norm": 1.5677348354673284, "language_loss": 0.69867688, "learning_rate": 1.7166320876407184e-06, "loss": 0.720581, "num_input_tokens_seen": 100226270, "step": 4648, "time_per_iteration": 3.258004665374756 }, { "auxiliary_loss_clip": 0.01181471, "auxiliary_loss_mlp": 0.00763066, "balance_loss_clip": 1.0548718, "balance_loss_mlp": 1.00035834, "epoch": 0.5590091985811338, "flos": 16472156585760.0, "grad_norm": 1.806880508193415, "language_loss": 0.68034184, "learning_rate": 1.7158609958159742e-06, "loss": 0.6997872, "num_input_tokens_seen": 100243675, "step": 4649, "time_per_iteration": 2.4284780025482178 }, { "auxiliary_loss_clip": 0.01118536, "auxiliary_loss_mlp": 0.01032065, "balance_loss_clip": 1.04776096, "balance_loss_mlp": 1.02399468, "epoch": 0.559129441471773, "flos": 14532189153600.0, "grad_norm": 1.9358422738561323, "language_loss": 0.78304708, "learning_rate": 1.7150899470948911e-06, "loss": 0.80455315, "num_input_tokens_seen": 100258940, "step": 4650, "time_per_iteration": 2.685290575027466 }, { "auxiliary_loss_clip": 0.01052395, "auxiliary_loss_mlp": 0.0100329, "balance_loss_clip": 1.01465988, "balance_loss_mlp": 1.00194252, "epoch": 0.5592496843624121, "flos": 60521021497920.0, "grad_norm": 0.7981272011323937, "language_loss": 0.56625283, "learning_rate": 1.7143189415944365e-06, "loss": 0.58680969, "num_input_tokens_seen": 100323400, "step": 4651, "time_per_iteration": 3.164799928665161 }, { "auxiliary_loss_clip": 0.01165141, "auxiliary_loss_mlp": 0.01028073, "balance_loss_clip": 1.05289006, "balance_loss_mlp": 1.02019894, "epoch": 0.5593699272530511, "flos": 20886260405280.0, "grad_norm": 1.570580125794093, "language_loss": 0.76063758, "learning_rate": 1.7135479794315714e-06, "loss": 0.78256977, "num_input_tokens_seen": 100340355, "step": 4652, "time_per_iteration": 2.492568016052246 }, { "auxiliary_loss_clip": 0.01136886, "auxiliary_loss_mlp": 0.01025663, "balance_loss_clip": 1.05010176, "balance_loss_mlp": 1.01870155, "epoch": 0.5594901701436903, "flos": 12896746435680.0, "grad_norm": 2.8492364542250113, "language_loss": 0.78907204, "learning_rate": 1.7127770607232502e-06, "loss": 0.81069756, "num_input_tokens_seen": 100358900, "step": 4653, "time_per_iteration": 2.550506591796875 }, { "auxiliary_loss_clip": 0.01143194, "auxiliary_loss_mlp": 0.01026345, "balance_loss_clip": 1.04888797, "balance_loss_mlp": 1.01942492, "epoch": 0.5596104130343293, "flos": 23112111649440.0, "grad_norm": 1.8096611590323424, "language_loss": 0.79704452, "learning_rate": 1.7120061855864204e-06, "loss": 0.81873989, "num_input_tokens_seen": 100378910, "step": 4654, "time_per_iteration": 2.586460828781128 }, { "auxiliary_loss_clip": 0.01167718, "auxiliary_loss_mlp": 0.01025026, "balance_loss_clip": 1.05433726, "balance_loss_mlp": 1.01717591, "epoch": 0.5597306559249684, "flos": 25957816938720.0, "grad_norm": 1.8913860179172033, "language_loss": 0.70451891, "learning_rate": 1.7112353541380233e-06, "loss": 0.72644639, "num_input_tokens_seen": 100398770, "step": 4655, "time_per_iteration": 2.532477617263794 }, { "auxiliary_loss_clip": 0.01153666, "auxiliary_loss_mlp": 0.01036701, "balance_loss_clip": 1.05270672, "balance_loss_mlp": 1.02849984, "epoch": 0.5598508988156076, "flos": 22492293521280.0, "grad_norm": 1.4561367645998067, "language_loss": 0.72236884, "learning_rate": 1.7104645664949931e-06, "loss": 0.74427253, "num_input_tokens_seen": 100421240, "step": 4656, "time_per_iteration": 2.633747100830078 }, { "auxiliary_loss_clip": 0.01152026, "auxiliary_loss_mlp": 0.01031651, "balance_loss_clip": 1.04952478, "balance_loss_mlp": 1.02412891, "epoch": 0.5599711417062466, "flos": 23112542652960.0, "grad_norm": 1.979539135354679, "language_loss": 0.71435934, "learning_rate": 1.7096938227742584e-06, "loss": 0.73619604, "num_input_tokens_seen": 100442370, "step": 4657, "time_per_iteration": 2.5946900844573975 }, { "auxiliary_loss_clip": 0.01183661, "auxiliary_loss_mlp": 0.01029145, "balance_loss_clip": 1.05637813, "balance_loss_mlp": 1.02141452, "epoch": 0.5600913845968857, "flos": 22339348741920.0, "grad_norm": 1.9050906880765717, "language_loss": 0.83934516, "learning_rate": 1.70892312309274e-06, "loss": 0.8614732, "num_input_tokens_seen": 100460260, "step": 4658, "time_per_iteration": 2.503011465072632 }, { "auxiliary_loss_clip": 0.01150561, "auxiliary_loss_mlp": 0.0102693, "balance_loss_clip": 1.04560041, "balance_loss_mlp": 1.01915193, "epoch": 0.5602116274875248, "flos": 17633793981600.0, "grad_norm": 2.2721858075779298, "language_loss": 0.67946213, "learning_rate": 1.7081524675673523e-06, "loss": 0.70123702, "num_input_tokens_seen": 100475750, "step": 4659, "time_per_iteration": 2.5163135528564453 }, { "auxiliary_loss_clip": 0.01055728, "auxiliary_loss_mlp": 0.01002048, "balance_loss_clip": 1.01695681, "balance_loss_mlp": 1.00073111, "epoch": 0.5603318703781639, "flos": 70115957995200.0, "grad_norm": 0.7683910828916669, "language_loss": 0.59643751, "learning_rate": 1.7073818563150026e-06, "loss": 0.6170153, "num_input_tokens_seen": 100537830, "step": 4660, "time_per_iteration": 3.245110273361206 }, { "auxiliary_loss_clip": 0.01160222, "auxiliary_loss_mlp": 0.01028354, "balance_loss_clip": 1.04901624, "balance_loss_mlp": 1.02101052, "epoch": 0.560452113268803, "flos": 18545853223680.0, "grad_norm": 2.1523746137907995, "language_loss": 0.86718786, "learning_rate": 1.7066112894525935e-06, "loss": 0.88907361, "num_input_tokens_seen": 100555910, "step": 4661, "time_per_iteration": 2.493727684020996 }, { "auxiliary_loss_clip": 0.0114346, "auxiliary_loss_mlp": 0.01031936, "balance_loss_clip": 1.04732108, "balance_loss_mlp": 1.02391934, "epoch": 0.5605723561594421, "flos": 25264669154880.0, "grad_norm": 2.1300895367324286, "language_loss": 0.72782505, "learning_rate": 1.7058407670970177e-06, "loss": 0.74957901, "num_input_tokens_seen": 100577385, "step": 4662, "time_per_iteration": 2.610466480255127 }, { "auxiliary_loss_clip": 0.01169537, "auxiliary_loss_mlp": 0.01027506, "balance_loss_clip": 1.05246878, "balance_loss_mlp": 1.01916146, "epoch": 0.5606925990500812, "flos": 20594953132320.0, "grad_norm": 2.601666070757627, "language_loss": 0.6095953, "learning_rate": 1.7050702893651643e-06, "loss": 0.63156569, "num_input_tokens_seen": 100596965, "step": 4663, "time_per_iteration": 2.5126402378082275 }, { "auxiliary_loss_clip": 0.01170018, "auxiliary_loss_mlp": 0.01032286, "balance_loss_clip": 1.05533564, "balance_loss_mlp": 1.02449584, "epoch": 0.5608128419407202, "flos": 35006049461760.0, "grad_norm": 2.002252077306076, "language_loss": 0.75245023, "learning_rate": 1.7042998563739134e-06, "loss": 0.77447325, "num_input_tokens_seen": 100615315, "step": 4664, "time_per_iteration": 2.6222379207611084 }, { "auxiliary_loss_clip": 0.01161111, "auxiliary_loss_mlp": 0.01028945, "balance_loss_clip": 1.0502758, "balance_loss_mlp": 1.02076077, "epoch": 0.5609330848313594, "flos": 24639822652320.0, "grad_norm": 2.2645038976986753, "language_loss": 0.71412092, "learning_rate": 1.703529468240139e-06, "loss": 0.73602146, "num_input_tokens_seen": 100634185, "step": 4665, "time_per_iteration": 2.5664138793945312 }, { "auxiliary_loss_clip": 0.01145439, "auxiliary_loss_mlp": 0.01032455, "balance_loss_clip": 1.05002546, "balance_loss_mlp": 1.02536213, "epoch": 0.5610533277219985, "flos": 18762897000000.0, "grad_norm": 2.2271279781407767, "language_loss": 0.73383123, "learning_rate": 1.7027591250807088e-06, "loss": 0.75561023, "num_input_tokens_seen": 100651360, "step": 4666, "time_per_iteration": 3.298814535140991 }, { "auxiliary_loss_clip": 0.0118187, "auxiliary_loss_mlp": 0.01028747, "balance_loss_clip": 1.05377591, "balance_loss_mlp": 1.02120709, "epoch": 0.5611735706126375, "flos": 15012387694560.0, "grad_norm": 2.104068047354377, "language_loss": 0.84851754, "learning_rate": 1.7019888270124825e-06, "loss": 0.87062371, "num_input_tokens_seen": 100668525, "step": 4667, "time_per_iteration": 3.178725481033325 }, { "auxiliary_loss_clip": 0.01172465, "auxiliary_loss_mlp": 0.01029488, "balance_loss_clip": 1.05532622, "balance_loss_mlp": 1.0223031, "epoch": 0.5612938135032767, "flos": 16468169803200.0, "grad_norm": 1.7201147500381198, "language_loss": 0.82076669, "learning_rate": 1.7012185741523147e-06, "loss": 0.84278619, "num_input_tokens_seen": 100684850, "step": 4668, "time_per_iteration": 2.5168302059173584 }, { "auxiliary_loss_clip": 0.01184102, "auxiliary_loss_mlp": 0.01028765, "balance_loss_clip": 1.05684447, "balance_loss_mlp": 1.02044392, "epoch": 0.5614140563939157, "flos": 25666437831840.0, "grad_norm": 1.9875218044673983, "language_loss": 0.62807304, "learning_rate": 1.7004483666170514e-06, "loss": 0.65020168, "num_input_tokens_seen": 100705345, "step": 4669, "time_per_iteration": 2.5103633403778076 }, { "auxiliary_loss_clip": 0.0116576, "auxiliary_loss_mlp": 0.01030953, "balance_loss_clip": 1.05268466, "balance_loss_mlp": 1.02323401, "epoch": 0.5615342992845548, "flos": 24717570094080.0, "grad_norm": 1.9618139559121734, "language_loss": 0.80511475, "learning_rate": 1.699678204523533e-06, "loss": 0.82708192, "num_input_tokens_seen": 100725210, "step": 4670, "time_per_iteration": 2.5517473220825195 }, { "auxiliary_loss_clip": 0.01157294, "auxiliary_loss_mlp": 0.01033227, "balance_loss_clip": 1.05409074, "balance_loss_mlp": 1.02517486, "epoch": 0.5616545421751938, "flos": 22015938095040.0, "grad_norm": 2.1779421316343446, "language_loss": 0.68374574, "learning_rate": 1.6989080879885918e-06, "loss": 0.70565093, "num_input_tokens_seen": 100743070, "step": 4671, "time_per_iteration": 3.237985372543335 }, { "auxiliary_loss_clip": 0.01047532, "auxiliary_loss_mlp": 0.01002466, "balance_loss_clip": 1.01900995, "balance_loss_mlp": 1.00121391, "epoch": 0.561774785065833, "flos": 53760368696640.0, "grad_norm": 0.9124563954050742, "language_loss": 0.61059213, "learning_rate": 1.6981380171290544e-06, "loss": 0.63109213, "num_input_tokens_seen": 100804095, "step": 4672, "time_per_iteration": 3.1704888343811035 }, { "auxiliary_loss_clip": 0.01145277, "auxiliary_loss_mlp": 0.0102897, "balance_loss_clip": 1.04834533, "balance_loss_mlp": 1.02150726, "epoch": 0.5618950279564721, "flos": 19750009911840.0, "grad_norm": 1.701184297605387, "language_loss": 0.73957121, "learning_rate": 1.6973679920617396e-06, "loss": 0.76131368, "num_input_tokens_seen": 100821630, "step": 4673, "time_per_iteration": 3.2928404808044434 }, { "auxiliary_loss_clip": 0.01151595, "auxiliary_loss_mlp": 0.01029372, "balance_loss_clip": 1.05292225, "balance_loss_mlp": 1.02170968, "epoch": 0.5620152708471111, "flos": 16800595607040.0, "grad_norm": 2.0113585213306675, "language_loss": 0.85484159, "learning_rate": 1.6965980129034603e-06, "loss": 0.87665123, "num_input_tokens_seen": 100839015, "step": 4674, "time_per_iteration": 2.560805082321167 }, { "auxiliary_loss_clip": 0.01155457, "auxiliary_loss_mlp": 0.01028407, "balance_loss_clip": 1.05381012, "balance_loss_mlp": 1.02035403, "epoch": 0.5621355137377503, "flos": 26797803618720.0, "grad_norm": 1.5432180915439877, "language_loss": 0.7652669, "learning_rate": 1.6958280797710209e-06, "loss": 0.78710556, "num_input_tokens_seen": 100860940, "step": 4675, "time_per_iteration": 2.579139232635498 }, { "auxiliary_loss_clip": 0.01053587, "auxiliary_loss_mlp": 0.01000502, "balance_loss_clip": 1.01604915, "balance_loss_mlp": 0.99919623, "epoch": 0.5622557566283893, "flos": 61207057723680.0, "grad_norm": 0.7123139357515784, "language_loss": 0.54779142, "learning_rate": 1.6950581927812198e-06, "loss": 0.56833231, "num_input_tokens_seen": 100920510, "step": 4676, "time_per_iteration": 3.010270118713379 }, { "auxiliary_loss_clip": 0.0116904, "auxiliary_loss_mlp": 0.01028302, "balance_loss_clip": 1.05323195, "balance_loss_mlp": 1.02031446, "epoch": 0.5623759995190284, "flos": 26468538507360.0, "grad_norm": 1.8827392150714615, "language_loss": 0.78954858, "learning_rate": 1.6942883520508486e-06, "loss": 0.81152201, "num_input_tokens_seen": 100939245, "step": 4677, "time_per_iteration": 2.537836790084839 }, { "auxiliary_loss_clip": 0.01165543, "auxiliary_loss_mlp": 0.01023579, "balance_loss_clip": 1.05276442, "balance_loss_mlp": 1.01643252, "epoch": 0.5624962424096676, "flos": 19390904195040.0, "grad_norm": 2.024075847919177, "language_loss": 0.76959276, "learning_rate": 1.693518557696691e-06, "loss": 0.79148394, "num_input_tokens_seen": 100958385, "step": 4678, "time_per_iteration": 2.4870026111602783 }, { "auxiliary_loss_clip": 0.01162125, "auxiliary_loss_mlp": 0.01031719, "balance_loss_clip": 1.0485301, "balance_loss_mlp": 1.0241996, "epoch": 0.5626164853003066, "flos": 20667348947040.0, "grad_norm": 2.3829041555708486, "language_loss": 0.89094108, "learning_rate": 1.6927488098355252e-06, "loss": 0.91287947, "num_input_tokens_seen": 100976015, "step": 4679, "time_per_iteration": 2.490779161453247 }, { "auxiliary_loss_clip": 0.01036212, "auxiliary_loss_mlp": 0.01002225, "balance_loss_clip": 1.01483226, "balance_loss_mlp": 1.00072944, "epoch": 0.5627367281909457, "flos": 62766081927360.0, "grad_norm": 0.9059772811276514, "language_loss": 0.6315608, "learning_rate": 1.6919791085841201e-06, "loss": 0.65194523, "num_input_tokens_seen": 101033425, "step": 4680, "time_per_iteration": 3.1461100578308105 }, { "auxiliary_loss_clip": 0.01159229, "auxiliary_loss_mlp": 0.01026097, "balance_loss_clip": 1.04833174, "balance_loss_mlp": 1.01849103, "epoch": 0.5628569710815848, "flos": 12787147038720.0, "grad_norm": 2.763698491345357, "language_loss": 0.78932679, "learning_rate": 1.6912094540592396e-06, "loss": 0.81118, "num_input_tokens_seen": 101048945, "step": 4681, "time_per_iteration": 2.4689571857452393 }, { "auxiliary_loss_clip": 0.01166052, "auxiliary_loss_mlp": 0.01031316, "balance_loss_clip": 1.05240417, "balance_loss_mlp": 1.02347779, "epoch": 0.5629772139722239, "flos": 13762084101120.0, "grad_norm": 2.810003313784333, "language_loss": 0.80735993, "learning_rate": 1.6904398463776393e-06, "loss": 0.82933354, "num_input_tokens_seen": 101062745, "step": 4682, "time_per_iteration": 2.4515268802642822 }, { "auxiliary_loss_clip": 0.01166189, "auxiliary_loss_mlp": 0.01025045, "balance_loss_clip": 1.04983497, "balance_loss_mlp": 1.01775229, "epoch": 0.5630974568628629, "flos": 21467833359360.0, "grad_norm": 1.628085475496583, "language_loss": 0.72419661, "learning_rate": 1.6896702856560683e-06, "loss": 0.74610895, "num_input_tokens_seen": 101081840, "step": 4683, "time_per_iteration": 2.509634256362915 }, { "auxiliary_loss_clip": 0.01133949, "auxiliary_loss_mlp": 0.01029616, "balance_loss_clip": 1.04499674, "balance_loss_mlp": 1.02165866, "epoch": 0.5632176997535021, "flos": 14245910255040.0, "grad_norm": 2.8215549636109274, "language_loss": 0.69259131, "learning_rate": 1.6889007720112677e-06, "loss": 0.7142269, "num_input_tokens_seen": 101099585, "step": 4684, "time_per_iteration": 2.5673413276672363 }, { "auxiliary_loss_clip": 0.01168203, "auxiliary_loss_mlp": 0.0102251, "balance_loss_clip": 1.05184436, "balance_loss_mlp": 1.01494646, "epoch": 0.5633379426441412, "flos": 20812248327360.0, "grad_norm": 1.7370524079337173, "language_loss": 0.77116287, "learning_rate": 1.6881313055599734e-06, "loss": 0.79307008, "num_input_tokens_seen": 101119515, "step": 4685, "time_per_iteration": 2.5212628841400146 }, { "auxiliary_loss_clip": 0.01136558, "auxiliary_loss_mlp": 0.0102951, "balance_loss_clip": 1.04492784, "balance_loss_mlp": 1.02098703, "epoch": 0.5634581855347802, "flos": 22600887243360.0, "grad_norm": 2.763113750905142, "language_loss": 0.8239423, "learning_rate": 1.6873618864189117e-06, "loss": 0.84560299, "num_input_tokens_seen": 101135285, "step": 4686, "time_per_iteration": 2.5226314067840576 }, { "auxiliary_loss_clip": 0.01165373, "auxiliary_loss_mlp": 0.01028156, "balance_loss_clip": 1.05075026, "balance_loss_mlp": 1.02025867, "epoch": 0.5635784284254194, "flos": 21506976457440.0, "grad_norm": 2.1775198658834602, "language_loss": 0.77584076, "learning_rate": 1.686592514704803e-06, "loss": 0.7977761, "num_input_tokens_seen": 101152680, "step": 4687, "time_per_iteration": 2.5000269412994385 }, { "auxiliary_loss_clip": 0.01149142, "auxiliary_loss_mlp": 0.01027979, "balance_loss_clip": 1.05132139, "balance_loss_mlp": 1.02081776, "epoch": 0.5636986713160584, "flos": 19827470017920.0, "grad_norm": 2.381841949398283, "language_loss": 0.70825255, "learning_rate": 1.685823190534361e-06, "loss": 0.73002374, "num_input_tokens_seen": 101170920, "step": 4688, "time_per_iteration": 2.520798921585083 }, { "auxiliary_loss_clip": 0.01184062, "auxiliary_loss_mlp": 0.01030324, "balance_loss_clip": 1.05325842, "balance_loss_mlp": 1.02246249, "epoch": 0.5638189142066975, "flos": 19792457370240.0, "grad_norm": 1.6981313535763034, "language_loss": 0.83275521, "learning_rate": 1.6850539140242907e-06, "loss": 0.85489905, "num_input_tokens_seen": 101190180, "step": 4689, "time_per_iteration": 2.467721939086914 }, { "auxiliary_loss_clip": 0.01165321, "auxiliary_loss_mlp": 0.01027378, "balance_loss_clip": 1.0514636, "balance_loss_mlp": 1.0198915, "epoch": 0.5639391570973367, "flos": 22893774862560.0, "grad_norm": 1.8831837536052551, "language_loss": 0.81941533, "learning_rate": 1.684284685291292e-06, "loss": 0.84134233, "num_input_tokens_seen": 101211825, "step": 4690, "time_per_iteration": 2.545064687728882 }, { "auxiliary_loss_clip": 0.01180663, "auxiliary_loss_mlp": 0.01026301, "balance_loss_clip": 1.05331182, "balance_loss_mlp": 1.01826072, "epoch": 0.5640593999879757, "flos": 23727080988000.0, "grad_norm": 2.0572118694255503, "language_loss": 0.81417108, "learning_rate": 1.683515504452055e-06, "loss": 0.83624065, "num_input_tokens_seen": 101229200, "step": 4691, "time_per_iteration": 2.493093252182007 }, { "auxiliary_loss_clip": 0.01128701, "auxiliary_loss_mlp": 0.01031043, "balance_loss_clip": 1.04679394, "balance_loss_mlp": 1.02305865, "epoch": 0.5641796428786148, "flos": 22710127470720.0, "grad_norm": 1.4351921605843343, "language_loss": 0.65959591, "learning_rate": 1.6827463716232648e-06, "loss": 0.68119335, "num_input_tokens_seen": 101249860, "step": 4692, "time_per_iteration": 3.353193521499634 }, { "auxiliary_loss_clip": 0.01164493, "auxiliary_loss_mlp": 0.00762429, "balance_loss_clip": 1.05081272, "balance_loss_mlp": 1.00040698, "epoch": 0.5642998857692539, "flos": 19791990449760.0, "grad_norm": 1.8849852167939365, "language_loss": 0.75466311, "learning_rate": 1.6819772869215972e-06, "loss": 0.77393234, "num_input_tokens_seen": 101268940, "step": 4693, "time_per_iteration": 3.274623155593872 }, { "auxiliary_loss_clip": 0.01159433, "auxiliary_loss_mlp": 0.01022188, "balance_loss_clip": 1.05164921, "balance_loss_mlp": 1.0149101, "epoch": 0.564420128659893, "flos": 23185908225600.0, "grad_norm": 1.9316234234047518, "language_loss": 0.82108003, "learning_rate": 1.6812082504637228e-06, "loss": 0.84289622, "num_input_tokens_seen": 101290260, "step": 4694, "time_per_iteration": 2.5772933959960938 }, { "auxiliary_loss_clip": 0.01165337, "auxiliary_loss_mlp": 0.01024218, "balance_loss_clip": 1.05392158, "balance_loss_mlp": 1.01676702, "epoch": 0.564540371550532, "flos": 23258268123360.0, "grad_norm": 1.6610692447934816, "language_loss": 0.74191213, "learning_rate": 1.6804392623663025e-06, "loss": 0.76380759, "num_input_tokens_seen": 101311465, "step": 4695, "time_per_iteration": 2.521225690841675 }, { "auxiliary_loss_clip": 0.01158714, "auxiliary_loss_mlp": 0.01024834, "balance_loss_clip": 1.05033255, "balance_loss_mlp": 1.01714492, "epoch": 0.5646606144411712, "flos": 25010062626720.0, "grad_norm": 2.235252808262604, "language_loss": 0.78564435, "learning_rate": 1.6796703227459935e-06, "loss": 0.8074798, "num_input_tokens_seen": 101329420, "step": 4696, "time_per_iteration": 3.2553799152374268 }, { "auxiliary_loss_clip": 0.01112957, "auxiliary_loss_mlp": 0.01029771, "balance_loss_clip": 1.04319143, "balance_loss_mlp": 1.0227201, "epoch": 0.5647808573318103, "flos": 36539650846080.0, "grad_norm": 1.9980765093227912, "language_loss": 0.76063168, "learning_rate": 1.6789014317194407e-06, "loss": 0.78205895, "num_input_tokens_seen": 101350900, "step": 4697, "time_per_iteration": 2.709291458129883 }, { "auxiliary_loss_clip": 0.0115624, "auxiliary_loss_mlp": 0.01036455, "balance_loss_clip": 1.05251753, "balance_loss_mlp": 1.0286144, "epoch": 0.5649011002224493, "flos": 22528455511680.0, "grad_norm": 3.1156026926748144, "language_loss": 0.7313354, "learning_rate": 1.6781325894032853e-06, "loss": 0.7532624, "num_input_tokens_seen": 101369860, "step": 4698, "time_per_iteration": 2.5589704513549805 }, { "auxiliary_loss_clip": 0.0114627, "auxiliary_loss_mlp": 0.01030266, "balance_loss_clip": 1.05032372, "balance_loss_mlp": 1.02259457, "epoch": 0.5650213431130885, "flos": 18515150611200.0, "grad_norm": 1.9459853605797957, "language_loss": 0.91917878, "learning_rate": 1.6773637959141608e-06, "loss": 0.94094419, "num_input_tokens_seen": 101386835, "step": 4699, "time_per_iteration": 3.2173709869384766 }, { "auxiliary_loss_clip": 0.01142616, "auxiliary_loss_mlp": 0.01024835, "balance_loss_clip": 1.04843616, "balance_loss_mlp": 1.01756346, "epoch": 0.5651415860037275, "flos": 17526313685280.0, "grad_norm": 2.2902294947155393, "language_loss": 0.66324604, "learning_rate": 1.6765950513686915e-06, "loss": 0.68492055, "num_input_tokens_seen": 101404945, "step": 4700, "time_per_iteration": 2.5256640911102295 }, { "auxiliary_loss_clip": 0.01121338, "auxiliary_loss_mlp": 0.01030187, "balance_loss_clip": 1.04352427, "balance_loss_mlp": 1.02228355, "epoch": 0.5652618288943666, "flos": 25520029939200.0, "grad_norm": 1.7370692564644463, "language_loss": 0.76249826, "learning_rate": 1.675826355883496e-06, "loss": 0.78401351, "num_input_tokens_seen": 101424160, "step": 4701, "time_per_iteration": 2.6257102489471436 }, { "auxiliary_loss_clip": 0.01148174, "auxiliary_loss_mlp": 0.01029152, "balance_loss_clip": 1.0517565, "balance_loss_mlp": 1.02176678, "epoch": 0.5653820717850057, "flos": 19683109392000.0, "grad_norm": 1.9042598725493025, "language_loss": 0.79379976, "learning_rate": 1.6750577095751848e-06, "loss": 0.81557304, "num_input_tokens_seen": 101443270, "step": 4702, "time_per_iteration": 2.501971483230591 }, { "auxiliary_loss_clip": 0.0117898, "auxiliary_loss_mlp": 0.01029116, "balance_loss_clip": 1.05386531, "balance_loss_mlp": 1.0221746, "epoch": 0.5655023146756448, "flos": 26979726996480.0, "grad_norm": 2.0659953713347874, "language_loss": 0.73114967, "learning_rate": 1.6742891125603605e-06, "loss": 0.75323063, "num_input_tokens_seen": 101464175, "step": 4703, "time_per_iteration": 2.5127384662628174 }, { "auxiliary_loss_clip": 0.01168962, "auxiliary_loss_mlp": 0.01032705, "balance_loss_clip": 1.05379665, "balance_loss_mlp": 1.02441406, "epoch": 0.5656225575662839, "flos": 27669354918240.0, "grad_norm": 2.7952897306536784, "language_loss": 0.7238003, "learning_rate": 1.6735205649556185e-06, "loss": 0.74581695, "num_input_tokens_seen": 101484045, "step": 4704, "time_per_iteration": 2.5583245754241943 }, { "auxiliary_loss_clip": 0.0113943, "auxiliary_loss_mlp": 0.01023159, "balance_loss_clip": 1.04968524, "balance_loss_mlp": 1.01583076, "epoch": 0.5657428004569229, "flos": 24349736556000.0, "grad_norm": 1.5734876849267656, "language_loss": 0.84536618, "learning_rate": 1.6727520668775476e-06, "loss": 0.86699206, "num_input_tokens_seen": 101504330, "step": 4705, "time_per_iteration": 2.6063549518585205 }, { "auxiliary_loss_clip": 0.01182713, "auxiliary_loss_mlp": 0.01022663, "balance_loss_clip": 1.05236888, "balance_loss_mlp": 1.01455641, "epoch": 0.5658630433475621, "flos": 21944045117760.0, "grad_norm": 1.468266978531868, "language_loss": 0.75092524, "learning_rate": 1.6719836184427275e-06, "loss": 0.77297902, "num_input_tokens_seen": 101524635, "step": 4706, "time_per_iteration": 2.507624864578247 }, { "auxiliary_loss_clip": 0.01143299, "auxiliary_loss_mlp": 0.01025912, "balance_loss_clip": 1.04606533, "balance_loss_mlp": 1.01820493, "epoch": 0.5659832862382012, "flos": 30409016589600.0, "grad_norm": 1.677125892406569, "language_loss": 0.64248407, "learning_rate": 1.671215219767733e-06, "loss": 0.66417623, "num_input_tokens_seen": 101544095, "step": 4707, "time_per_iteration": 2.622321605682373 }, { "auxiliary_loss_clip": 0.01121433, "auxiliary_loss_mlp": 0.01034393, "balance_loss_clip": 1.04403591, "balance_loss_mlp": 1.02661443, "epoch": 0.5661035291288402, "flos": 13188197376480.0, "grad_norm": 1.8838451828853144, "language_loss": 0.75859725, "learning_rate": 1.670446870969127e-06, "loss": 0.78015554, "num_input_tokens_seen": 101561760, "step": 4708, "time_per_iteration": 2.579224109649658 }, { "auxiliary_loss_clip": 0.01153252, "auxiliary_loss_mlp": 0.01025711, "balance_loss_clip": 1.05255127, "balance_loss_mlp": 1.01870728, "epoch": 0.5662237720194794, "flos": 16143035142240.0, "grad_norm": 2.6045291664799657, "language_loss": 0.80035621, "learning_rate": 1.6696785721634685e-06, "loss": 0.82214588, "num_input_tokens_seen": 101576245, "step": 4709, "time_per_iteration": 2.507643699645996 }, { "auxiliary_loss_clip": 0.01168243, "auxiliary_loss_mlp": 0.01031485, "balance_loss_clip": 1.05326152, "balance_loss_mlp": 1.02350366, "epoch": 0.5663440149101184, "flos": 17676852028320.0, "grad_norm": 2.025724126972402, "language_loss": 0.73307681, "learning_rate": 1.6689103234673086e-06, "loss": 0.75507402, "num_input_tokens_seen": 101594565, "step": 4710, "time_per_iteration": 2.4661502838134766 }, { "auxiliary_loss_clip": 0.01149706, "auxiliary_loss_mlp": 0.01027451, "balance_loss_clip": 1.0499481, "balance_loss_mlp": 1.0202806, "epoch": 0.5664642578007575, "flos": 23368334440800.0, "grad_norm": 1.8547111682535187, "language_loss": 0.76608098, "learning_rate": 1.668142124997189e-06, "loss": 0.78785253, "num_input_tokens_seen": 101614225, "step": 4711, "time_per_iteration": 2.567796230316162 }, { "auxiliary_loss_clip": 0.01055791, "auxiliary_loss_mlp": 0.01003856, "balance_loss_clip": 1.02279615, "balance_loss_mlp": 1.00270021, "epoch": 0.5665845006913967, "flos": 65516482769760.0, "grad_norm": 0.727387746503673, "language_loss": 0.59819806, "learning_rate": 1.6673739768696453e-06, "loss": 0.6187945, "num_input_tokens_seen": 101680795, "step": 4712, "time_per_iteration": 3.1728506088256836 }, { "auxiliary_loss_clip": 0.01156696, "auxiliary_loss_mlp": 0.01028256, "balance_loss_clip": 1.04885936, "balance_loss_mlp": 1.02072167, "epoch": 0.5667047435820357, "flos": 26140889659200.0, "grad_norm": 1.6693001381880797, "language_loss": 0.77705014, "learning_rate": 1.6666058792012052e-06, "loss": 0.79889965, "num_input_tokens_seen": 101701680, "step": 4713, "time_per_iteration": 2.613492250442505 }, { "auxiliary_loss_clip": 0.01065902, "auxiliary_loss_mlp": 0.01001602, "balance_loss_clip": 1.0145694, "balance_loss_mlp": 1.00033844, "epoch": 0.5668249864726748, "flos": 71866962325440.0, "grad_norm": 0.8817258594252062, "language_loss": 0.68857372, "learning_rate": 1.6658378321083878e-06, "loss": 0.70924878, "num_input_tokens_seen": 101766010, "step": 4714, "time_per_iteration": 3.155271291732788 }, { "auxiliary_loss_clip": 0.01112371, "auxiliary_loss_mlp": 0.01027055, "balance_loss_clip": 1.0483582, "balance_loss_mlp": 1.01941371, "epoch": 0.5669452293633139, "flos": 22195670538240.0, "grad_norm": 2.1908221949999684, "language_loss": 0.82450557, "learning_rate": 1.6650698357077055e-06, "loss": 0.84589982, "num_input_tokens_seen": 101783055, "step": 4715, "time_per_iteration": 2.61789608001709 }, { "auxiliary_loss_clip": 0.01153294, "auxiliary_loss_mlp": 0.01026282, "balance_loss_clip": 1.04765642, "balance_loss_mlp": 1.01889062, "epoch": 0.567065472253953, "flos": 18223196832960.0, "grad_norm": 2.7736397881955472, "language_loss": 0.8151592, "learning_rate": 1.6643018901156632e-06, "loss": 0.83695495, "num_input_tokens_seen": 101802150, "step": 4716, "time_per_iteration": 2.4971413612365723 }, { "auxiliary_loss_clip": 0.01153143, "auxiliary_loss_mlp": 0.0102685, "balance_loss_clip": 1.0478369, "balance_loss_mlp": 1.01943517, "epoch": 0.567185715144592, "flos": 20371552054080.0, "grad_norm": 2.4738336176011386, "language_loss": 0.79577696, "learning_rate": 1.6635339954487566e-06, "loss": 0.817577, "num_input_tokens_seen": 101818025, "step": 4717, "time_per_iteration": 2.5301594734191895 }, { "auxiliary_loss_clip": 0.01155593, "auxiliary_loss_mlp": 0.01027046, "balance_loss_clip": 1.05017424, "balance_loss_mlp": 1.01942897, "epoch": 0.5673059580352312, "flos": 23221351876800.0, "grad_norm": 1.6405376529524662, "language_loss": 0.8256917, "learning_rate": 1.6627661518234765e-06, "loss": 0.84751809, "num_input_tokens_seen": 101837280, "step": 4718, "time_per_iteration": 3.3521227836608887 }, { "auxiliary_loss_clip": 0.01127072, "auxiliary_loss_mlp": 0.01026261, "balance_loss_clip": 1.04898167, "balance_loss_mlp": 1.01818442, "epoch": 0.5674262009258703, "flos": 21719602447680.0, "grad_norm": 1.5289492158497648, "language_loss": 0.85386842, "learning_rate": 1.661998359356302e-06, "loss": 0.87540179, "num_input_tokens_seen": 101856310, "step": 4719, "time_per_iteration": 3.357093572616577 }, { "auxiliary_loss_clip": 0.01073293, "auxiliary_loss_mlp": 0.01002397, "balance_loss_clip": 1.01405561, "balance_loss_mlp": 1.00115132, "epoch": 0.5675464438165093, "flos": 67470379594080.0, "grad_norm": 0.753290482709989, "language_loss": 0.55780196, "learning_rate": 1.6612306181637077e-06, "loss": 0.57855886, "num_input_tokens_seen": 101915635, "step": 4720, "time_per_iteration": 3.0760793685913086 }, { "auxiliary_loss_clip": 0.0113621, "auxiliary_loss_mlp": 0.01023292, "balance_loss_clip": 1.04817581, "balance_loss_mlp": 1.01587152, "epoch": 0.5676666867071485, "flos": 18879176951520.0, "grad_norm": 2.073893410044646, "language_loss": 0.65516025, "learning_rate": 1.6604629283621598e-06, "loss": 0.67675531, "num_input_tokens_seen": 101933565, "step": 4721, "time_per_iteration": 2.528409957885742 }, { "auxiliary_loss_clip": 0.01183675, "auxiliary_loss_mlp": 0.01029031, "balance_loss_clip": 1.05380511, "balance_loss_mlp": 1.02102041, "epoch": 0.5677869295977875, "flos": 33546783408000.0, "grad_norm": 2.7356678908621133, "language_loss": 0.74439156, "learning_rate": 1.6596952900681152e-06, "loss": 0.76651865, "num_input_tokens_seen": 101954325, "step": 4722, "time_per_iteration": 3.333489418029785 }, { "auxiliary_loss_clip": 0.0111662, "auxiliary_loss_mlp": 0.01028892, "balance_loss_clip": 1.0487175, "balance_loss_mlp": 1.02047586, "epoch": 0.5679071724884266, "flos": 28037260290240.0, "grad_norm": 2.197615207956367, "language_loss": 0.81953657, "learning_rate": 1.658927703398025e-06, "loss": 0.84099174, "num_input_tokens_seen": 101974390, "step": 4723, "time_per_iteration": 2.6557440757751465 }, { "auxiliary_loss_clip": 0.0112175, "auxiliary_loss_mlp": 0.01025426, "balance_loss_clip": 1.04175794, "balance_loss_mlp": 1.01796985, "epoch": 0.5680274153790658, "flos": 23550114150720.0, "grad_norm": 2.2917105371716837, "language_loss": 0.77762806, "learning_rate": 1.6581601684683309e-06, "loss": 0.7990998, "num_input_tokens_seen": 101994815, "step": 4724, "time_per_iteration": 2.60770583152771 }, { "auxiliary_loss_clip": 0.0116763, "auxiliary_loss_mlp": 0.01028324, "balance_loss_clip": 1.05308604, "balance_loss_mlp": 1.02085567, "epoch": 0.5681476582697048, "flos": 22455161772960.0, "grad_norm": 3.006890178623854, "language_loss": 0.68044043, "learning_rate": 1.6573926853954674e-06, "loss": 0.70239997, "num_input_tokens_seen": 102012400, "step": 4725, "time_per_iteration": 3.2041616439819336 }, { "auxiliary_loss_clip": 0.01145278, "auxiliary_loss_mlp": 0.01028772, "balance_loss_clip": 1.04673171, "balance_loss_mlp": 1.02183425, "epoch": 0.5682679011603439, "flos": 19536917001120.0, "grad_norm": 1.815881243672858, "language_loss": 0.82785124, "learning_rate": 1.6566252542958608e-06, "loss": 0.84959179, "num_input_tokens_seen": 102031900, "step": 4726, "time_per_iteration": 2.5223095417022705 }, { "auxiliary_loss_clip": 0.01128901, "auxiliary_loss_mlp": 0.01026201, "balance_loss_clip": 1.04771209, "balance_loss_mlp": 1.01900399, "epoch": 0.568388144050983, "flos": 28765492555680.0, "grad_norm": 2.095305927853241, "language_loss": 0.78870207, "learning_rate": 1.6558578752859305e-06, "loss": 0.81025308, "num_input_tokens_seen": 102050860, "step": 4727, "time_per_iteration": 2.634359121322632 }, { "auxiliary_loss_clip": 0.01134423, "auxiliary_loss_mlp": 0.01032399, "balance_loss_clip": 1.04623604, "balance_loss_mlp": 1.02457023, "epoch": 0.5685083869416221, "flos": 21209455550400.0, "grad_norm": 1.9694111054279804, "language_loss": 0.78839767, "learning_rate": 1.6550905484820865e-06, "loss": 0.81006587, "num_input_tokens_seen": 102069320, "step": 4728, "time_per_iteration": 2.5741918087005615 }, { "auxiliary_loss_clip": 0.01181454, "auxiliary_loss_mlp": 0.01036714, "balance_loss_clip": 1.05254078, "balance_loss_mlp": 1.02826178, "epoch": 0.5686286298322611, "flos": 24827025823200.0, "grad_norm": 2.208789401333348, "language_loss": 0.784343, "learning_rate": 1.6543232740007328e-06, "loss": 0.80652475, "num_input_tokens_seen": 102086435, "step": 4729, "time_per_iteration": 2.5085294246673584 }, { "auxiliary_loss_clip": 0.0116974, "auxiliary_loss_mlp": 0.01030439, "balance_loss_clip": 1.0531323, "balance_loss_mlp": 1.02284539, "epoch": 0.5687488727229003, "flos": 26615125984800.0, "grad_norm": 2.4111667621789645, "language_loss": 0.66869104, "learning_rate": 1.653556051958263e-06, "loss": 0.69069278, "num_input_tokens_seen": 102106115, "step": 4730, "time_per_iteration": 2.5360846519470215 }, { "auxiliary_loss_clip": 0.01091218, "auxiliary_loss_mlp": 0.01029651, "balance_loss_clip": 1.0418551, "balance_loss_mlp": 1.02189064, "epoch": 0.5688691156135394, "flos": 20808728465280.0, "grad_norm": 1.6622290942496096, "language_loss": 0.73568082, "learning_rate": 1.6527888824710642e-06, "loss": 0.75688958, "num_input_tokens_seen": 102125715, "step": 4731, "time_per_iteration": 2.6487083435058594 }, { "auxiliary_loss_clip": 0.01129219, "auxiliary_loss_mlp": 0.01031013, "balance_loss_clip": 1.04495156, "balance_loss_mlp": 1.02272797, "epoch": 0.5689893585041784, "flos": 25880967420960.0, "grad_norm": 2.1138052054982404, "language_loss": 0.76446986, "learning_rate": 1.6520217656555166e-06, "loss": 0.78607213, "num_input_tokens_seen": 102145005, "step": 4732, "time_per_iteration": 2.5933401584625244 }, { "auxiliary_loss_clip": 0.01138013, "auxiliary_loss_mlp": 0.01025454, "balance_loss_clip": 1.04597855, "balance_loss_mlp": 1.01830125, "epoch": 0.5691096013948175, "flos": 23477466917280.0, "grad_norm": 1.4676898074506204, "language_loss": 0.71243447, "learning_rate": 1.65125470162799e-06, "loss": 0.73406911, "num_input_tokens_seen": 102165360, "step": 4733, "time_per_iteration": 2.5695841312408447 }, { "auxiliary_loss_clip": 0.01141024, "auxiliary_loss_mlp": 0.01032109, "balance_loss_clip": 1.04717386, "balance_loss_mlp": 1.02422047, "epoch": 0.5692298442854566, "flos": 18075603680640.0, "grad_norm": 2.1470968898929566, "language_loss": 0.69821984, "learning_rate": 1.6504876905048485e-06, "loss": 0.71995115, "num_input_tokens_seen": 102182320, "step": 4734, "time_per_iteration": 2.53476881980896 }, { "auxiliary_loss_clip": 0.0117745, "auxiliary_loss_mlp": 0.01029513, "balance_loss_clip": 1.05305517, "balance_loss_mlp": 1.02210093, "epoch": 0.5693500871760957, "flos": 23039320748160.0, "grad_norm": 2.03922865354761, "language_loss": 0.72085571, "learning_rate": 1.6497207324024464e-06, "loss": 0.74292529, "num_input_tokens_seen": 102201220, "step": 4735, "time_per_iteration": 2.480253219604492 }, { "auxiliary_loss_clip": 0.01159756, "auxiliary_loss_mlp": 0.01032304, "balance_loss_clip": 1.0521183, "balance_loss_mlp": 1.02453136, "epoch": 0.5694703300667348, "flos": 18989674272480.0, "grad_norm": 1.8274454058954537, "language_loss": 0.82292593, "learning_rate": 1.6489538274371305e-06, "loss": 0.84484649, "num_input_tokens_seen": 102219825, "step": 4736, "time_per_iteration": 2.558070182800293 }, { "auxiliary_loss_clip": 0.01160564, "auxiliary_loss_mlp": 0.01030766, "balance_loss_clip": 1.0519917, "balance_loss_mlp": 1.02339292, "epoch": 0.5695905729573739, "flos": 21908709217440.0, "grad_norm": 1.919627016737931, "language_loss": 0.83026409, "learning_rate": 1.6481869757252396e-06, "loss": 0.85217732, "num_input_tokens_seen": 102238160, "step": 4737, "time_per_iteration": 2.5008671283721924 }, { "auxiliary_loss_clip": 0.01165462, "auxiliary_loss_mlp": 0.01029866, "balance_loss_clip": 1.05165386, "balance_loss_mlp": 1.02296078, "epoch": 0.569710815848013, "flos": 28476663552960.0, "grad_norm": 1.3781012205104977, "language_loss": 0.71757066, "learning_rate": 1.647420177383105e-06, "loss": 0.73952389, "num_input_tokens_seen": 102261030, "step": 4738, "time_per_iteration": 2.557896852493286 }, { "auxiliary_loss_clip": 0.01162073, "auxiliary_loss_mlp": 0.01028645, "balance_loss_clip": 1.05379701, "balance_loss_mlp": 1.02136433, "epoch": 0.569831058738652, "flos": 28366166232000.0, "grad_norm": 1.7749567743415442, "language_loss": 0.72671944, "learning_rate": 1.646653432527049e-06, "loss": 0.74862659, "num_input_tokens_seen": 102281670, "step": 4739, "time_per_iteration": 2.536358118057251 }, { "auxiliary_loss_clip": 0.01140921, "auxiliary_loss_mlp": 0.01026751, "balance_loss_clip": 1.04989827, "balance_loss_mlp": 1.01996469, "epoch": 0.5699513016292912, "flos": 25849977472800.0, "grad_norm": 1.5566570012107157, "language_loss": 0.74707448, "learning_rate": 1.645886741273387e-06, "loss": 0.76875126, "num_input_tokens_seen": 102303485, "step": 4740, "time_per_iteration": 2.6702284812927246 }, { "auxiliary_loss_clip": 0.01133726, "auxiliary_loss_mlp": 0.01027553, "balance_loss_clip": 1.05167127, "balance_loss_mlp": 1.01937556, "epoch": 0.5700715445199303, "flos": 18037861344000.0, "grad_norm": 2.1741829185234725, "language_loss": 0.73775482, "learning_rate": 1.645120103738424e-06, "loss": 0.75936759, "num_input_tokens_seen": 102320995, "step": 4741, "time_per_iteration": 2.534181833267212 }, { "auxiliary_loss_clip": 0.01153087, "auxiliary_loss_mlp": 0.00761786, "balance_loss_clip": 1.04830325, "balance_loss_mlp": 1.00043011, "epoch": 0.5701917874105693, "flos": 11473355036640.0, "grad_norm": 2.0547003163598014, "language_loss": 0.84337914, "learning_rate": 1.6443535200384591e-06, "loss": 0.86252785, "num_input_tokens_seen": 102339170, "step": 4742, "time_per_iteration": 2.4757750034332275 }, { "auxiliary_loss_clip": 0.01178684, "auxiliary_loss_mlp": 0.01031744, "balance_loss_clip": 1.05354667, "balance_loss_mlp": 1.02410865, "epoch": 0.5703120303012085, "flos": 21761762570400.0, "grad_norm": 1.6862865570122314, "language_loss": 0.7055406, "learning_rate": 1.6435869902897827e-06, "loss": 0.72764486, "num_input_tokens_seen": 102357750, "step": 4743, "time_per_iteration": 2.464538335800171 }, { "auxiliary_loss_clip": 0.01039074, "auxiliary_loss_mlp": 0.01003623, "balance_loss_clip": 1.01175284, "balance_loss_mlp": 1.00233507, "epoch": 0.5704322731918475, "flos": 56746268244480.0, "grad_norm": 0.7906621223655387, "language_loss": 0.62096274, "learning_rate": 1.6428205146086764e-06, "loss": 0.64138973, "num_input_tokens_seen": 102419730, "step": 4744, "time_per_iteration": 3.939958333969116 }, { "auxiliary_loss_clip": 0.0115558, "auxiliary_loss_mlp": 0.0102982, "balance_loss_clip": 1.049155, "balance_loss_mlp": 1.0225482, "epoch": 0.5705525160824866, "flos": 20741145523200.0, "grad_norm": 1.4613475409214745, "language_loss": 0.70478296, "learning_rate": 1.6420540931114142e-06, "loss": 0.72663689, "num_input_tokens_seen": 102440320, "step": 4745, "time_per_iteration": 3.322477102279663 }, { "auxiliary_loss_clip": 0.01154214, "auxiliary_loss_mlp": 0.01022475, "balance_loss_clip": 1.05053031, "balance_loss_mlp": 1.01464319, "epoch": 0.5706727589731257, "flos": 18771265651680.0, "grad_norm": 1.5919551063669386, "language_loss": 0.78647542, "learning_rate": 1.6412877259142616e-06, "loss": 0.80824226, "num_input_tokens_seen": 102460240, "step": 4746, "time_per_iteration": 2.531956672668457 }, { "auxiliary_loss_clip": 0.0114733, "auxiliary_loss_mlp": 0.01027479, "balance_loss_clip": 1.0498445, "balance_loss_mlp": 1.02021301, "epoch": 0.5707930018637648, "flos": 27634737357120.0, "grad_norm": 5.600078088617803, "language_loss": 0.73629928, "learning_rate": 1.6405214131334757e-06, "loss": 0.75804734, "num_input_tokens_seen": 102478765, "step": 4747, "time_per_iteration": 2.558173656463623 }, { "auxiliary_loss_clip": 0.01120692, "auxiliary_loss_mlp": 0.01030808, "balance_loss_clip": 1.04974604, "balance_loss_mlp": 1.02330947, "epoch": 0.5709132447544039, "flos": 27597677442720.0, "grad_norm": 1.8526543097278028, "language_loss": 0.79475081, "learning_rate": 1.6397551548853052e-06, "loss": 0.81626582, "num_input_tokens_seen": 102496930, "step": 4748, "time_per_iteration": 2.634350299835205 }, { "auxiliary_loss_clip": 0.01154617, "auxiliary_loss_mlp": 0.01029916, "balance_loss_clip": 1.05234861, "balance_loss_mlp": 1.02215505, "epoch": 0.571033487645043, "flos": 21686098312320.0, "grad_norm": 1.6388301460413643, "language_loss": 0.706936, "learning_rate": 1.6389889512859917e-06, "loss": 0.72878134, "num_input_tokens_seen": 102516590, "step": 4749, "time_per_iteration": 3.2918636798858643 }, { "auxiliary_loss_clip": 0.01050405, "auxiliary_loss_mlp": 0.01000484, "balance_loss_clip": 1.0128876, "balance_loss_mlp": 0.99928015, "epoch": 0.5711537305356821, "flos": 70181457753600.0, "grad_norm": 0.8145879731070319, "language_loss": 0.60399973, "learning_rate": 1.638222802451767e-06, "loss": 0.62450862, "num_input_tokens_seen": 102578070, "step": 4750, "time_per_iteration": 3.1250457763671875 }, { "auxiliary_loss_clip": 0.01159456, "auxiliary_loss_mlp": 0.01030269, "balance_loss_clip": 1.05118608, "balance_loss_mlp": 1.02292585, "epoch": 0.5712739734263211, "flos": 24717498260160.0, "grad_norm": 2.0407900918775628, "language_loss": 0.75220972, "learning_rate": 1.6374567084988561e-06, "loss": 0.77410698, "num_input_tokens_seen": 102599255, "step": 4751, "time_per_iteration": 3.247545003890991 }, { "auxiliary_loss_clip": 0.01154797, "auxiliary_loss_mlp": 0.01030087, "balance_loss_clip": 1.0529511, "balance_loss_mlp": 1.02211201, "epoch": 0.5713942163169603, "flos": 26578173821280.0, "grad_norm": 1.881856471374428, "language_loss": 0.76380146, "learning_rate": 1.6366906695434738e-06, "loss": 0.78565025, "num_input_tokens_seen": 102621775, "step": 4752, "time_per_iteration": 2.5617830753326416 }, { "auxiliary_loss_clip": 0.01164974, "auxiliary_loss_mlp": 0.01024098, "balance_loss_clip": 1.05172563, "balance_loss_mlp": 1.01647413, "epoch": 0.5715144592075994, "flos": 21142447279680.0, "grad_norm": 1.926335921756652, "language_loss": 0.86078101, "learning_rate": 1.6359246857018275e-06, "loss": 0.88267171, "num_input_tokens_seen": 102639305, "step": 4753, "time_per_iteration": 2.4782657623291016 }, { "auxiliary_loss_clip": 0.01115915, "auxiliary_loss_mlp": 0.01032979, "balance_loss_clip": 1.04273701, "balance_loss_mlp": 1.02518868, "epoch": 0.5716347020982384, "flos": 23330268851520.0, "grad_norm": 1.724132990885601, "language_loss": 0.78062105, "learning_rate": 1.6351587570901178e-06, "loss": 0.80210996, "num_input_tokens_seen": 102659430, "step": 4754, "time_per_iteration": 2.5877761840820312 }, { "auxiliary_loss_clip": 0.01137779, "auxiliary_loss_mlp": 0.0102548, "balance_loss_clip": 1.04910159, "balance_loss_mlp": 1.01805902, "epoch": 0.5717549449888776, "flos": 17009558067360.0, "grad_norm": 2.3801467248807207, "language_loss": 0.75631106, "learning_rate": 1.634392883824534e-06, "loss": 0.77794367, "num_input_tokens_seen": 102671430, "step": 4755, "time_per_iteration": 2.4885265827178955 }, { "auxiliary_loss_clip": 0.01124811, "auxiliary_loss_mlp": 0.01031852, "balance_loss_clip": 1.04730058, "balance_loss_mlp": 1.02475333, "epoch": 0.5718751878795166, "flos": 35518135874880.0, "grad_norm": 3.1633238038650684, "language_loss": 0.68015242, "learning_rate": 1.6336270660212595e-06, "loss": 0.70171905, "num_input_tokens_seen": 102693025, "step": 4756, "time_per_iteration": 2.6794304847717285 }, { "auxiliary_loss_clip": 0.0114991, "auxiliary_loss_mlp": 0.01025982, "balance_loss_clip": 1.05286312, "balance_loss_mlp": 1.01859736, "epoch": 0.5719954307701557, "flos": 38613993989280.0, "grad_norm": 2.047751379667113, "language_loss": 0.663252, "learning_rate": 1.6328613037964676e-06, "loss": 0.68501091, "num_input_tokens_seen": 102716090, "step": 4757, "time_per_iteration": 2.657871723175049 }, { "auxiliary_loss_clip": 0.01164533, "auxiliary_loss_mlp": 0.01027776, "balance_loss_clip": 1.05014968, "balance_loss_mlp": 1.02011728, "epoch": 0.5721156736607949, "flos": 20631115122720.0, "grad_norm": 1.7740597237677334, "language_loss": 0.68021959, "learning_rate": 1.6320955972663241e-06, "loss": 0.70214272, "num_input_tokens_seen": 102735685, "step": 4758, "time_per_iteration": 2.4853978157043457 }, { "auxiliary_loss_clip": 0.01162019, "auxiliary_loss_mlp": 0.01028712, "balance_loss_clip": 1.04890561, "balance_loss_mlp": 1.02141047, "epoch": 0.5722359165514339, "flos": 37415081177280.0, "grad_norm": 1.840726697139043, "language_loss": 0.65604115, "learning_rate": 1.6313299465469857e-06, "loss": 0.67794842, "num_input_tokens_seen": 102758415, "step": 4759, "time_per_iteration": 2.617710828781128 }, { "auxiliary_loss_clip": 0.0115709, "auxiliary_loss_mlp": 0.01029554, "balance_loss_clip": 1.04816449, "balance_loss_mlp": 1.02122092, "epoch": 0.572356159442073, "flos": 21972880048320.0, "grad_norm": 2.2397358789946975, "language_loss": 0.79474652, "learning_rate": 1.6305643517546014e-06, "loss": 0.81661296, "num_input_tokens_seen": 102773795, "step": 4760, "time_per_iteration": 2.458984375 }, { "auxiliary_loss_clip": 0.01177828, "auxiliary_loss_mlp": 0.01026155, "balance_loss_clip": 1.05313718, "balance_loss_mlp": 1.01912153, "epoch": 0.5724764023327121, "flos": 19135543410720.0, "grad_norm": 1.791149973247976, "language_loss": 0.84530997, "learning_rate": 1.629798813005311e-06, "loss": 0.86734986, "num_input_tokens_seen": 102793515, "step": 4761, "time_per_iteration": 2.456334352493286 }, { "auxiliary_loss_clip": 0.0112472, "auxiliary_loss_mlp": 0.01026307, "balance_loss_clip": 1.04963863, "balance_loss_mlp": 1.01914227, "epoch": 0.5725966452233512, "flos": 22819762784640.0, "grad_norm": 1.964588954519478, "language_loss": 0.71141338, "learning_rate": 1.6290333304152473e-06, "loss": 0.73292363, "num_input_tokens_seen": 102813390, "step": 4762, "time_per_iteration": 2.589035987854004 }, { "auxiliary_loss_clip": 0.01146488, "auxiliary_loss_mlp": 0.01032278, "balance_loss_clip": 1.05074561, "balance_loss_mlp": 1.02429068, "epoch": 0.5727168881139902, "flos": 41496615525120.0, "grad_norm": 1.8126567644034564, "language_loss": 0.56513739, "learning_rate": 1.6282679041005314e-06, "loss": 0.58692503, "num_input_tokens_seen": 102838980, "step": 4763, "time_per_iteration": 2.6926462650299072 }, { "auxiliary_loss_clip": 0.0114195, "auxiliary_loss_mlp": 0.01023608, "balance_loss_clip": 1.04646993, "balance_loss_mlp": 1.01678658, "epoch": 0.5728371310046293, "flos": 14647678932000.0, "grad_norm": 2.0004260393526248, "language_loss": 0.871979, "learning_rate": 1.6275025341772789e-06, "loss": 0.89363456, "num_input_tokens_seen": 102855285, "step": 4764, "time_per_iteration": 2.5287818908691406 }, { "auxiliary_loss_clip": 0.01151223, "auxiliary_loss_mlp": 0.01034785, "balance_loss_clip": 1.04772067, "balance_loss_mlp": 1.02685165, "epoch": 0.5729573738952685, "flos": 21506617287840.0, "grad_norm": 2.2955543703042887, "language_loss": 0.81920993, "learning_rate": 1.626737220761596e-06, "loss": 0.84106994, "num_input_tokens_seen": 102872750, "step": 4765, "time_per_iteration": 2.516873359680176 }, { "auxiliary_loss_clip": 0.01163691, "auxiliary_loss_mlp": 0.01037185, "balance_loss_clip": 1.05388415, "balance_loss_mlp": 1.02963042, "epoch": 0.5730776167859075, "flos": 23621683875360.0, "grad_norm": 4.147413967842906, "language_loss": 0.78736764, "learning_rate": 1.62597196396958e-06, "loss": 0.80937642, "num_input_tokens_seen": 102890920, "step": 4766, "time_per_iteration": 2.520954132080078 }, { "auxiliary_loss_clip": 0.01163201, "auxiliary_loss_mlp": 0.01032325, "balance_loss_clip": 1.05048323, "balance_loss_mlp": 1.02480841, "epoch": 0.5731978596765466, "flos": 25739228733120.0, "grad_norm": 1.6312789777041223, "language_loss": 0.8553884, "learning_rate": 1.6252067639173197e-06, "loss": 0.87734365, "num_input_tokens_seen": 102912830, "step": 4767, "time_per_iteration": 2.518085479736328 }, { "auxiliary_loss_clip": 0.01167493, "auxiliary_loss_mlp": 0.010313, "balance_loss_clip": 1.05275869, "balance_loss_mlp": 1.02367699, "epoch": 0.5733181025671857, "flos": 26359513781760.0, "grad_norm": 1.833041468586415, "language_loss": 0.69251817, "learning_rate": 1.6244416207208956e-06, "loss": 0.71450615, "num_input_tokens_seen": 102933765, "step": 4768, "time_per_iteration": 2.5479800701141357 }, { "auxiliary_loss_clip": 0.01136955, "auxiliary_loss_mlp": 0.01027975, "balance_loss_clip": 1.04839623, "balance_loss_mlp": 1.02037859, "epoch": 0.5734383454578248, "flos": 29423879110560.0, "grad_norm": 1.6091559806741573, "language_loss": 0.73564416, "learning_rate": 1.6236765344963787e-06, "loss": 0.75729346, "num_input_tokens_seen": 102955025, "step": 4769, "time_per_iteration": 2.6058189868927 }, { "auxiliary_loss_clip": 0.01149502, "auxiliary_loss_mlp": 0.01025085, "balance_loss_clip": 1.04938841, "balance_loss_mlp": 1.01647174, "epoch": 0.5735585883484638, "flos": 34969959305280.0, "grad_norm": 2.117109836204105, "language_loss": 0.69073874, "learning_rate": 1.6229115053598322e-06, "loss": 0.7124846, "num_input_tokens_seen": 102976780, "step": 4770, "time_per_iteration": 3.436656951904297 }, { "auxiliary_loss_clip": 0.0116884, "auxiliary_loss_mlp": 0.01036319, "balance_loss_clip": 1.05474901, "balance_loss_mlp": 1.02823043, "epoch": 0.573678831239103, "flos": 18770762814240.0, "grad_norm": 2.339617292855786, "language_loss": 0.71717596, "learning_rate": 1.6221465334273108e-06, "loss": 0.73922753, "num_input_tokens_seen": 102995990, "step": 4771, "time_per_iteration": 3.388969898223877 }, { "auxiliary_loss_clip": 0.01141286, "auxiliary_loss_mlp": 0.01030865, "balance_loss_clip": 1.04874015, "balance_loss_mlp": 1.02315199, "epoch": 0.5737990741297421, "flos": 25702887157920.0, "grad_norm": 2.0530184083302188, "language_loss": 0.61343628, "learning_rate": 1.6213816188148593e-06, "loss": 0.63515776, "num_input_tokens_seen": 103014695, "step": 4772, "time_per_iteration": 2.6140995025634766 }, { "auxiliary_loss_clip": 0.01141885, "auxiliary_loss_mlp": 0.0103551, "balance_loss_clip": 1.05057383, "balance_loss_mlp": 1.02767777, "epoch": 0.5739193170203811, "flos": 27269238421440.0, "grad_norm": 1.6421996751803771, "language_loss": 0.77115768, "learning_rate": 1.6206167616385162e-06, "loss": 0.79293162, "num_input_tokens_seen": 103035760, "step": 4773, "time_per_iteration": 2.572930097579956 }, { "auxiliary_loss_clip": 0.0115949, "auxiliary_loss_mlp": 0.01033125, "balance_loss_clip": 1.05274117, "balance_loss_mlp": 1.0250963, "epoch": 0.5740395599110203, "flos": 12239724725280.0, "grad_norm": 1.9000565043973419, "language_loss": 0.7340517, "learning_rate": 1.6198519620143078e-06, "loss": 0.75597787, "num_input_tokens_seen": 103052915, "step": 4774, "time_per_iteration": 3.276904344558716 }, { "auxiliary_loss_clip": 0.01142703, "auxiliary_loss_mlp": 0.01024814, "balance_loss_clip": 1.05054426, "balance_loss_mlp": 1.01767373, "epoch": 0.5741598028016593, "flos": 25921403529600.0, "grad_norm": 1.936844125003406, "language_loss": 0.78167546, "learning_rate": 1.6190872200582546e-06, "loss": 0.80335057, "num_input_tokens_seen": 103074655, "step": 4775, "time_per_iteration": 2.61263108253479 }, { "auxiliary_loss_clip": 0.01144108, "auxiliary_loss_mlp": 0.00762627, "balance_loss_clip": 1.04646611, "balance_loss_mlp": 1.00028133, "epoch": 0.5742800456922984, "flos": 19244137132800.0, "grad_norm": 2.061126930993719, "language_loss": 0.78147233, "learning_rate": 1.6183225358863676e-06, "loss": 0.80053961, "num_input_tokens_seen": 103091550, "step": 4776, "time_per_iteration": 2.502619981765747 }, { "auxiliary_loss_clip": 0.01142628, "auxiliary_loss_mlp": 0.0102578, "balance_loss_clip": 1.04788685, "balance_loss_mlp": 1.01767325, "epoch": 0.5744002885829376, "flos": 30920492414400.0, "grad_norm": 2.1772761994072183, "language_loss": 0.71945763, "learning_rate": 1.617557909614648e-06, "loss": 0.74114168, "num_input_tokens_seen": 103110985, "step": 4777, "time_per_iteration": 3.311955451965332 }, { "auxiliary_loss_clip": 0.01135617, "auxiliary_loss_mlp": 0.01035187, "balance_loss_clip": 1.04784536, "balance_loss_mlp": 1.02742696, "epoch": 0.5745205314735766, "flos": 23840020662240.0, "grad_norm": 1.8672193066907519, "language_loss": 0.85613024, "learning_rate": 1.6167933413590899e-06, "loss": 0.87783831, "num_input_tokens_seen": 103129890, "step": 4778, "time_per_iteration": 2.5565872192382812 }, { "auxiliary_loss_clip": 0.01163034, "auxiliary_loss_mlp": 0.01028928, "balance_loss_clip": 1.04993629, "balance_loss_mlp": 1.02130723, "epoch": 0.5746407743642157, "flos": 12311905038240.0, "grad_norm": 2.2126382480742417, "language_loss": 0.91097641, "learning_rate": 1.6160288312356773e-06, "loss": 0.93289602, "num_input_tokens_seen": 103147020, "step": 4779, "time_per_iteration": 2.465362310409546 }, { "auxiliary_loss_clip": 0.01168355, "auxiliary_loss_mlp": 0.0102932, "balance_loss_clip": 1.05072749, "balance_loss_mlp": 1.02192295, "epoch": 0.5747610172548548, "flos": 24133662537600.0, "grad_norm": 1.5983514318596233, "language_loss": 0.81641126, "learning_rate": 1.6152643793603857e-06, "loss": 0.83838809, "num_input_tokens_seen": 103167370, "step": 4780, "time_per_iteration": 2.5115692615509033 }, { "auxiliary_loss_clip": 0.0118048, "auxiliary_loss_mlp": 0.01031025, "balance_loss_clip": 1.05406237, "balance_loss_mlp": 1.02279341, "epoch": 0.5748812601454939, "flos": 25408455109440.0, "grad_norm": 1.7645538992022274, "language_loss": 0.8789221, "learning_rate": 1.6144999858491815e-06, "loss": 0.9010371, "num_input_tokens_seen": 103186000, "step": 4781, "time_per_iteration": 2.523819923400879 }, { "auxiliary_loss_clip": 0.01156371, "auxiliary_loss_mlp": 0.01029213, "balance_loss_clip": 1.04955053, "balance_loss_mlp": 1.02096343, "epoch": 0.575001503036133, "flos": 30624946940160.0, "grad_norm": 2.0418863852913303, "language_loss": 0.85736883, "learning_rate": 1.6137356508180232e-06, "loss": 0.87922466, "num_input_tokens_seen": 103207710, "step": 4782, "time_per_iteration": 2.5950794219970703 }, { "auxiliary_loss_clip": 0.01179698, "auxiliary_loss_mlp": 0.00762784, "balance_loss_clip": 1.05229402, "balance_loss_mlp": 1.00033796, "epoch": 0.5751217459267721, "flos": 21726570337920.0, "grad_norm": 1.6542276999324041, "language_loss": 0.81088907, "learning_rate": 1.6129713743828593e-06, "loss": 0.83031392, "num_input_tokens_seen": 103226720, "step": 4783, "time_per_iteration": 2.4860541820526123 }, { "auxiliary_loss_clip": 0.01149511, "auxiliary_loss_mlp": 0.01028348, "balance_loss_clip": 1.04671001, "balance_loss_mlp": 1.02118099, "epoch": 0.5752419888174112, "flos": 21651624419040.0, "grad_norm": 1.6389163230414883, "language_loss": 0.7533282, "learning_rate": 1.6122071566596306e-06, "loss": 0.77510679, "num_input_tokens_seen": 103246995, "step": 4784, "time_per_iteration": 2.532273530960083 }, { "auxiliary_loss_clip": 0.01166946, "auxiliary_loss_mlp": 0.01025076, "balance_loss_clip": 1.05126858, "balance_loss_mlp": 1.01725602, "epoch": 0.5753622317080502, "flos": 17775999589920.0, "grad_norm": 2.968948276861731, "language_loss": 0.83399212, "learning_rate": 1.6114429977642674e-06, "loss": 0.85591239, "num_input_tokens_seen": 103261500, "step": 4785, "time_per_iteration": 2.4691009521484375 }, { "auxiliary_loss_clip": 0.01164901, "auxiliary_loss_mlp": 0.01028614, "balance_loss_clip": 1.05281019, "balance_loss_mlp": 1.0205971, "epoch": 0.5754824745986894, "flos": 19789619930400.0, "grad_norm": 2.9599151651381477, "language_loss": 0.73915094, "learning_rate": 1.6106788978126926e-06, "loss": 0.76108611, "num_input_tokens_seen": 103280475, "step": 4786, "time_per_iteration": 2.515125274658203 }, { "auxiliary_loss_clip": 0.01118065, "auxiliary_loss_mlp": 0.010265, "balance_loss_clip": 1.0433774, "balance_loss_mlp": 1.01904392, "epoch": 0.5756027174893285, "flos": 30985668920160.0, "grad_norm": 2.123683248249591, "language_loss": 0.78581816, "learning_rate": 1.6099148569208196e-06, "loss": 0.80726379, "num_input_tokens_seen": 103297695, "step": 4787, "time_per_iteration": 2.6519713401794434 }, { "auxiliary_loss_clip": 0.01151739, "auxiliary_loss_mlp": 0.01027476, "balance_loss_clip": 1.05202889, "balance_loss_mlp": 1.01980805, "epoch": 0.5757229603799675, "flos": 28546868433120.0, "grad_norm": 1.7334011352787424, "language_loss": 0.63235235, "learning_rate": 1.6091508752045523e-06, "loss": 0.65414453, "num_input_tokens_seen": 103318575, "step": 4788, "time_per_iteration": 2.582115650177002 }, { "auxiliary_loss_clip": 0.01126803, "auxiliary_loss_mlp": 0.01027154, "balance_loss_clip": 1.04361033, "balance_loss_mlp": 1.02001977, "epoch": 0.5758432032706067, "flos": 22999028307360.0, "grad_norm": 1.6270350572784056, "language_loss": 0.86463469, "learning_rate": 1.608386952779787e-06, "loss": 0.88617432, "num_input_tokens_seen": 103337945, "step": 4789, "time_per_iteration": 2.562433958053589 }, { "auxiliary_loss_clip": 0.01154327, "auxiliary_loss_mlp": 0.01030476, "balance_loss_clip": 1.04841793, "balance_loss_mlp": 1.02375817, "epoch": 0.5759634461612457, "flos": 25739731570560.0, "grad_norm": 1.6721509147622753, "language_loss": 0.74525928, "learning_rate": 1.6076230897624098e-06, "loss": 0.76710731, "num_input_tokens_seen": 103360150, "step": 4790, "time_per_iteration": 2.57922625541687 }, { "auxiliary_loss_clip": 0.01165564, "auxiliary_loss_mlp": 0.0103011, "balance_loss_clip": 1.04798377, "balance_loss_mlp": 1.02228963, "epoch": 0.5760836890518848, "flos": 30591766057440.0, "grad_norm": 1.9046824924682189, "language_loss": 0.77272087, "learning_rate": 1.6068592862682974e-06, "loss": 0.79467756, "num_input_tokens_seen": 103378305, "step": 4791, "time_per_iteration": 2.554779052734375 }, { "auxiliary_loss_clip": 0.01154281, "auxiliary_loss_mlp": 0.01034202, "balance_loss_clip": 1.05100632, "balance_loss_mlp": 1.02635169, "epoch": 0.576203931942524, "flos": 36538968423840.0, "grad_norm": 2.066708224596109, "language_loss": 0.7363385, "learning_rate": 1.6060955424133187e-06, "loss": 0.75822341, "num_input_tokens_seen": 103399230, "step": 4792, "time_per_iteration": 2.646186590194702 }, { "auxiliary_loss_clip": 0.01164233, "auxiliary_loss_mlp": 0.01025015, "balance_loss_clip": 1.05112028, "balance_loss_mlp": 1.01734948, "epoch": 0.576324174833163, "flos": 25516941080640.0, "grad_norm": 1.8754327604711842, "language_loss": 0.89470863, "learning_rate": 1.6053318583133332e-06, "loss": 0.91660106, "num_input_tokens_seen": 103420100, "step": 4793, "time_per_iteration": 2.557560920715332 }, { "auxiliary_loss_clip": 0.011648, "auxiliary_loss_mlp": 0.0102751, "balance_loss_clip": 1.05093133, "balance_loss_mlp": 1.01939178, "epoch": 0.5764444177238021, "flos": 25119266937120.0, "grad_norm": 1.9982538529502833, "language_loss": 0.75419688, "learning_rate": 1.6045682340841907e-06, "loss": 0.77611995, "num_input_tokens_seen": 103439025, "step": 4794, "time_per_iteration": 2.5270309448242188 }, { "auxiliary_loss_clip": 0.01038906, "auxiliary_loss_mlp": 0.00752524, "balance_loss_clip": 1.01073182, "balance_loss_mlp": 0.99982512, "epoch": 0.5765646606144411, "flos": 62212697398560.0, "grad_norm": 0.7538932303458636, "language_loss": 0.58062434, "learning_rate": 1.6038046698417336e-06, "loss": 0.59853876, "num_input_tokens_seen": 103499920, "step": 4795, "time_per_iteration": 3.1340157985687256 }, { "auxiliary_loss_clip": 0.01162883, "auxiliary_loss_mlp": 0.01029955, "balance_loss_clip": 1.04839027, "balance_loss_mlp": 1.02229023, "epoch": 0.5766849035050803, "flos": 25118764099680.0, "grad_norm": 2.2415419047195777, "language_loss": 0.68980688, "learning_rate": 1.6030411657017919e-06, "loss": 0.71173525, "num_input_tokens_seen": 103519575, "step": 4796, "time_per_iteration": 3.316415786743164 }, { "auxiliary_loss_clip": 0.01156897, "auxiliary_loss_mlp": 0.01032181, "balance_loss_clip": 1.04874492, "balance_loss_mlp": 1.02472496, "epoch": 0.5768051463957193, "flos": 15991096037760.0, "grad_norm": 1.6540720743579105, "language_loss": 0.84116441, "learning_rate": 1.6022777217801903e-06, "loss": 0.86305523, "num_input_tokens_seen": 103536530, "step": 4797, "time_per_iteration": 3.222970962524414 }, { "auxiliary_loss_clip": 0.01139571, "auxiliary_loss_mlp": 0.01031505, "balance_loss_clip": 1.05173039, "balance_loss_mlp": 1.02361965, "epoch": 0.5769253892863584, "flos": 22163638998240.0, "grad_norm": 1.9802060248625255, "language_loss": 0.7356835, "learning_rate": 1.601514338192742e-06, "loss": 0.75739431, "num_input_tokens_seen": 103556460, "step": 4798, "time_per_iteration": 2.5794248580932617 }, { "auxiliary_loss_clip": 0.01176571, "auxiliary_loss_mlp": 0.01024194, "balance_loss_clip": 1.05280662, "balance_loss_mlp": 1.01741958, "epoch": 0.5770456321769976, "flos": 22856391695520.0, "grad_norm": 2.2930114109874955, "language_loss": 0.71271861, "learning_rate": 1.6007510150552514e-06, "loss": 0.73472625, "num_input_tokens_seen": 103574520, "step": 4799, "time_per_iteration": 2.457435369491577 }, { "auxiliary_loss_clip": 0.01170174, "auxiliary_loss_mlp": 0.0103464, "balance_loss_clip": 1.04958987, "balance_loss_mlp": 1.02602673, "epoch": 0.5771658750676366, "flos": 46353678386400.0, "grad_norm": 1.4742655464185102, "language_loss": 0.61835515, "learning_rate": 1.599987752483515e-06, "loss": 0.64040327, "num_input_tokens_seen": 103598965, "step": 4800, "time_per_iteration": 3.3896470069885254 }, { "auxiliary_loss_clip": 0.01130051, "auxiliary_loss_mlp": 0.01028254, "balance_loss_clip": 1.0454936, "balance_loss_mlp": 1.02090204, "epoch": 0.5772861179582757, "flos": 22159975468320.0, "grad_norm": 1.6813242121951566, "language_loss": 0.67669755, "learning_rate": 1.5992245505933184e-06, "loss": 0.69828057, "num_input_tokens_seen": 103618665, "step": 4801, "time_per_iteration": 2.5774049758911133 }, { "auxiliary_loss_clip": 0.01182356, "auxiliary_loss_mlp": 0.01026947, "balance_loss_clip": 1.0541054, "balance_loss_mlp": 1.0192045, "epoch": 0.5774063608489148, "flos": 31248931435680.0, "grad_norm": 2.3457739150717085, "language_loss": 0.71509308, "learning_rate": 1.5984614095004388e-06, "loss": 0.73718619, "num_input_tokens_seen": 103639800, "step": 4802, "time_per_iteration": 2.5284924507141113 }, { "auxiliary_loss_clip": 0.01161473, "auxiliary_loss_mlp": 0.0103157, "balance_loss_clip": 1.04982686, "balance_loss_mlp": 1.02366006, "epoch": 0.5775266037395539, "flos": 22527126584160.0, "grad_norm": 2.4518111135039242, "language_loss": 0.80606294, "learning_rate": 1.5976983293206438e-06, "loss": 0.82799333, "num_input_tokens_seen": 103655605, "step": 4803, "time_per_iteration": 3.227806568145752 }, { "auxiliary_loss_clip": 0.01146558, "auxiliary_loss_mlp": 0.01027793, "balance_loss_clip": 1.04679942, "balance_loss_mlp": 1.02005661, "epoch": 0.577646846630193, "flos": 21068794371360.0, "grad_norm": 2.1419270781714252, "language_loss": 0.71034801, "learning_rate": 1.5969353101696928e-06, "loss": 0.73209155, "num_input_tokens_seen": 103674045, "step": 4804, "time_per_iteration": 2.5080554485321045 }, { "auxiliary_loss_clip": 0.0116462, "auxiliary_loss_mlp": 0.01030298, "balance_loss_clip": 1.05049694, "balance_loss_mlp": 1.02279949, "epoch": 0.5777670895208321, "flos": 29714288459520.0, "grad_norm": 3.8186546115430473, "language_loss": 0.79717898, "learning_rate": 1.5961723521633341e-06, "loss": 0.81912816, "num_input_tokens_seen": 103695285, "step": 4805, "time_per_iteration": 2.5683176517486572 }, { "auxiliary_loss_clip": 0.01144318, "auxiliary_loss_mlp": 0.01029133, "balance_loss_clip": 1.04839242, "balance_loss_mlp": 1.02166474, "epoch": 0.5778873324114712, "flos": 19500431758080.0, "grad_norm": 2.319897104428174, "language_loss": 0.91096872, "learning_rate": 1.5954094554173097e-06, "loss": 0.93270314, "num_input_tokens_seen": 103713275, "step": 4806, "time_per_iteration": 2.5062763690948486 }, { "auxiliary_loss_clip": 0.01156007, "auxiliary_loss_mlp": 0.01022978, "balance_loss_clip": 1.05161214, "balance_loss_mlp": 1.015962, "epoch": 0.5780075753021102, "flos": 14136849612480.0, "grad_norm": 2.0497319422762765, "language_loss": 0.79175234, "learning_rate": 1.5946466200473482e-06, "loss": 0.81354225, "num_input_tokens_seen": 103731185, "step": 4807, "time_per_iteration": 2.5127549171447754 }, { "auxiliary_loss_clip": 0.01155155, "auxiliary_loss_mlp": 0.01024389, "balance_loss_clip": 1.04911375, "balance_loss_mlp": 1.01724827, "epoch": 0.5781278181927494, "flos": 15262181350080.0, "grad_norm": 1.7675797683616628, "language_loss": 0.83296204, "learning_rate": 1.5938838461691723e-06, "loss": 0.85475743, "num_input_tokens_seen": 103748095, "step": 4808, "time_per_iteration": 2.489649534225464 }, { "auxiliary_loss_clip": 0.01181165, "auxiliary_loss_mlp": 0.0102671, "balance_loss_clip": 1.05355382, "balance_loss_mlp": 1.01884806, "epoch": 0.5782480610833884, "flos": 16726834947840.0, "grad_norm": 3.388480332063422, "language_loss": 0.83189905, "learning_rate": 1.593121133898494e-06, "loss": 0.8539778, "num_input_tokens_seen": 103765300, "step": 4809, "time_per_iteration": 2.448169231414795 }, { "auxiliary_loss_clip": 0.01172228, "auxiliary_loss_mlp": 0.01031229, "balance_loss_clip": 1.05204666, "balance_loss_mlp": 1.02370644, "epoch": 0.5783683039740275, "flos": 25482143934720.0, "grad_norm": 9.04556193107422, "language_loss": 0.79329073, "learning_rate": 1.592358483351016e-06, "loss": 0.81532532, "num_input_tokens_seen": 103785475, "step": 4810, "time_per_iteration": 2.520634651184082 }, { "auxiliary_loss_clip": 0.01162608, "auxiliary_loss_mlp": 0.0102521, "balance_loss_clip": 1.05105019, "balance_loss_mlp": 1.0179615, "epoch": 0.5784885468646667, "flos": 18405838549920.0, "grad_norm": 1.954581147460458, "language_loss": 0.72235096, "learning_rate": 1.5915958946424326e-06, "loss": 0.74422914, "num_input_tokens_seen": 103804160, "step": 4811, "time_per_iteration": 2.482891798019409 }, { "auxiliary_loss_clip": 0.01139279, "auxiliary_loss_mlp": 0.00762972, "balance_loss_clip": 1.04796779, "balance_loss_mlp": 1.00033712, "epoch": 0.5786087897553057, "flos": 46100724038400.0, "grad_norm": 1.5583174115386575, "language_loss": 0.74135345, "learning_rate": 1.5908333678884271e-06, "loss": 0.76037598, "num_input_tokens_seen": 103830580, "step": 4812, "time_per_iteration": 2.7811176776885986 }, { "auxiliary_loss_clip": 0.01164677, "auxiliary_loss_mlp": 0.01029593, "balance_loss_clip": 1.05140996, "balance_loss_mlp": 1.02127278, "epoch": 0.5787290326459448, "flos": 12385953033120.0, "grad_norm": 1.7964743991909684, "language_loss": 0.73667401, "learning_rate": 1.5900709032046743e-06, "loss": 0.75861675, "num_input_tokens_seen": 103848655, "step": 4813, "time_per_iteration": 2.4596168994903564 }, { "auxiliary_loss_clip": 0.01147502, "auxiliary_loss_mlp": 0.01027853, "balance_loss_clip": 1.05066919, "balance_loss_mlp": 1.02068543, "epoch": 0.5788492755365839, "flos": 23290335580320.0, "grad_norm": 2.3710440427330077, "language_loss": 0.78272843, "learning_rate": 1.5893085007068391e-06, "loss": 0.80448198, "num_input_tokens_seen": 103866215, "step": 4814, "time_per_iteration": 2.5049846172332764 }, { "auxiliary_loss_clip": 0.01138544, "auxiliary_loss_mlp": 0.01031792, "balance_loss_clip": 1.04713213, "balance_loss_mlp": 1.02402532, "epoch": 0.578969518427223, "flos": 24061051221120.0, "grad_norm": 1.8153832034115256, "language_loss": 0.70510125, "learning_rate": 1.5885461605105786e-06, "loss": 0.72680461, "num_input_tokens_seen": 103887815, "step": 4815, "time_per_iteration": 2.5422070026397705 }, { "auxiliary_loss_clip": 0.01152896, "auxiliary_loss_mlp": 0.01030107, "balance_loss_clip": 1.04955256, "balance_loss_mlp": 1.02213764, "epoch": 0.579089761317862, "flos": 21871828887840.0, "grad_norm": 2.9036228920705356, "language_loss": 0.76528692, "learning_rate": 1.5877838827315375e-06, "loss": 0.787117, "num_input_tokens_seen": 103906360, "step": 4816, "time_per_iteration": 2.5038721561431885 }, { "auxiliary_loss_clip": 0.01176912, "auxiliary_loss_mlp": 0.01030146, "balance_loss_clip": 1.05147815, "balance_loss_mlp": 1.02221894, "epoch": 0.5792100042085012, "flos": 22929685434240.0, "grad_norm": 1.7401165698547467, "language_loss": 0.6996007, "learning_rate": 1.587021667485355e-06, "loss": 0.72167134, "num_input_tokens_seen": 103925730, "step": 4817, "time_per_iteration": 2.456021547317505 }, { "auxiliary_loss_clip": 0.01153038, "auxiliary_loss_mlp": 0.01026674, "balance_loss_clip": 1.04768646, "balance_loss_mlp": 1.01946509, "epoch": 0.5793302470991403, "flos": 21470060210880.0, "grad_norm": 1.8187454955614677, "language_loss": 0.78729343, "learning_rate": 1.5862595148876559e-06, "loss": 0.8090905, "num_input_tokens_seen": 103945835, "step": 4818, "time_per_iteration": 2.515930414199829 }, { "auxiliary_loss_clip": 0.0112528, "auxiliary_loss_mlp": 0.01025819, "balance_loss_clip": 1.04644012, "balance_loss_mlp": 1.01788878, "epoch": 0.5794504899897793, "flos": 12711015860160.0, "grad_norm": 2.771683031175648, "language_loss": 0.75753391, "learning_rate": 1.58549742505406e-06, "loss": 0.77904499, "num_input_tokens_seen": 103960580, "step": 4819, "time_per_iteration": 2.5408146381378174 }, { "auxiliary_loss_clip": 0.01180734, "auxiliary_loss_mlp": 0.01030351, "balance_loss_clip": 1.05344784, "balance_loss_mlp": 1.02240014, "epoch": 0.5795707328804185, "flos": 14867057310720.0, "grad_norm": 2.3212841502337263, "language_loss": 0.756037, "learning_rate": 1.5847353981001747e-06, "loss": 0.77814782, "num_input_tokens_seen": 103977760, "step": 4820, "time_per_iteration": 2.427328586578369 }, { "auxiliary_loss_clip": 0.01144455, "auxiliary_loss_mlp": 0.01026089, "balance_loss_clip": 1.0461024, "balance_loss_mlp": 1.01862025, "epoch": 0.5796909757710575, "flos": 36430051449120.0, "grad_norm": 1.6117007232578415, "language_loss": 0.69801235, "learning_rate": 1.5839734341415993e-06, "loss": 0.71971774, "num_input_tokens_seen": 103999960, "step": 4821, "time_per_iteration": 2.6537814140319824 }, { "auxiliary_loss_clip": 0.01162851, "auxiliary_loss_mlp": 0.01024586, "balance_loss_clip": 1.05531478, "balance_loss_mlp": 1.01750457, "epoch": 0.5798112186616966, "flos": 23039895419520.0, "grad_norm": 1.7250063793943244, "language_loss": 0.76400381, "learning_rate": 1.5832115332939238e-06, "loss": 0.78587818, "num_input_tokens_seen": 104018400, "step": 4822, "time_per_iteration": 3.266045093536377 }, { "auxiliary_loss_clip": 0.01166474, "auxiliary_loss_mlp": 0.01030854, "balance_loss_clip": 1.05107415, "balance_loss_mlp": 1.02310514, "epoch": 0.5799314615523358, "flos": 16652607368160.0, "grad_norm": 1.6522328090738938, "language_loss": 0.74640441, "learning_rate": 1.5824496956727272e-06, "loss": 0.76837772, "num_input_tokens_seen": 104035605, "step": 4823, "time_per_iteration": 3.2201266288757324 }, { "auxiliary_loss_clip": 0.01152139, "auxiliary_loss_mlp": 0.01021884, "balance_loss_clip": 1.04945159, "balance_loss_mlp": 1.01476693, "epoch": 0.5800517044429748, "flos": 20485676988000.0, "grad_norm": 1.8890249669364603, "language_loss": 0.73112231, "learning_rate": 1.5816879213935797e-06, "loss": 0.75286251, "num_input_tokens_seen": 104054415, "step": 4824, "time_per_iteration": 2.558314085006714 }, { "auxiliary_loss_clip": 0.01162172, "auxiliary_loss_mlp": 0.0102821, "balance_loss_clip": 1.052742, "balance_loss_mlp": 1.02100706, "epoch": 0.5801719473336139, "flos": 31538263275840.0, "grad_norm": 1.5346581071267664, "language_loss": 0.79658484, "learning_rate": 1.5809262105720416e-06, "loss": 0.81848872, "num_input_tokens_seen": 104075455, "step": 4825, "time_per_iteration": 2.63403582572937 }, { "auxiliary_loss_clip": 0.01177507, "auxiliary_loss_mlp": 0.01027899, "balance_loss_clip": 1.05222487, "balance_loss_mlp": 1.02073455, "epoch": 0.580292190224253, "flos": 20375969840160.0, "grad_norm": 1.8530502124143453, "language_loss": 0.79924154, "learning_rate": 1.5801645633236644e-06, "loss": 0.82129556, "num_input_tokens_seen": 104096440, "step": 4826, "time_per_iteration": 3.198915958404541 }, { "auxiliary_loss_clip": 0.01142739, "auxiliary_loss_mlp": 0.01030241, "balance_loss_clip": 1.04642582, "balance_loss_mlp": 1.02264762, "epoch": 0.5804124331148921, "flos": 26615377403520.0, "grad_norm": 1.7728564308832004, "language_loss": 0.76944256, "learning_rate": 1.579402979763989e-06, "loss": 0.79117239, "num_input_tokens_seen": 104116775, "step": 4827, "time_per_iteration": 2.5622875690460205 }, { "auxiliary_loss_clip": 0.01117251, "auxiliary_loss_mlp": 0.01028305, "balance_loss_clip": 1.04542291, "balance_loss_mlp": 1.02074099, "epoch": 0.5805326760055312, "flos": 13478498974560.0, "grad_norm": 2.3531666897048384, "language_loss": 0.81342953, "learning_rate": 1.578641460008548e-06, "loss": 0.83488512, "num_input_tokens_seen": 104134510, "step": 4828, "time_per_iteration": 2.566131114959717 }, { "auxiliary_loss_clip": 0.01161664, "auxiliary_loss_mlp": 0.01023215, "balance_loss_clip": 1.05107093, "balance_loss_mlp": 1.0162977, "epoch": 0.5806529188961702, "flos": 12091377316800.0, "grad_norm": 2.028803470326585, "language_loss": 0.67861021, "learning_rate": 1.5778800041728613e-06, "loss": 0.700459, "num_input_tokens_seen": 104150800, "step": 4829, "time_per_iteration": 3.197800874710083 }, { "auxiliary_loss_clip": 0.01158224, "auxiliary_loss_mlp": 0.01025989, "balance_loss_clip": 1.0495863, "balance_loss_mlp": 1.01886368, "epoch": 0.5807731617868094, "flos": 26214111564000.0, "grad_norm": 1.546248447627082, "language_loss": 0.66217268, "learning_rate": 1.577118612372443e-06, "loss": 0.6840148, "num_input_tokens_seen": 104172640, "step": 4830, "time_per_iteration": 2.5436336994171143 }, { "auxiliary_loss_clip": 0.0114475, "auxiliary_loss_mlp": 0.00763472, "balance_loss_clip": 1.04663241, "balance_loss_mlp": 1.00035679, "epoch": 0.5808934046774484, "flos": 37962144321120.0, "grad_norm": 1.7466176490233645, "language_loss": 0.70451891, "learning_rate": 1.5763572847227943e-06, "loss": 0.7236011, "num_input_tokens_seen": 104193525, "step": 4831, "time_per_iteration": 2.6660776138305664 }, { "auxiliary_loss_clip": 0.01163545, "auxiliary_loss_mlp": 0.0102329, "balance_loss_clip": 1.04961479, "balance_loss_mlp": 1.01600957, "epoch": 0.5810136475680875, "flos": 20485856572800.0, "grad_norm": 1.9094314462883992, "language_loss": 0.8149122, "learning_rate": 1.5755960213394091e-06, "loss": 0.83678049, "num_input_tokens_seen": 104210625, "step": 4832, "time_per_iteration": 2.4804344177246094 }, { "auxiliary_loss_clip": 0.01138212, "auxiliary_loss_mlp": 0.01022249, "balance_loss_clip": 1.04884291, "balance_loss_mlp": 1.01465225, "epoch": 0.5811338904587267, "flos": 17530156800000.0, "grad_norm": 1.796294322055097, "language_loss": 0.78414166, "learning_rate": 1.5748348223377703e-06, "loss": 0.80574632, "num_input_tokens_seen": 104228180, "step": 4833, "time_per_iteration": 2.54868483543396 }, { "auxiliary_loss_clip": 0.01150229, "auxiliary_loss_mlp": 0.01030293, "balance_loss_clip": 1.05127072, "balance_loss_mlp": 1.02298534, "epoch": 0.5812541333493657, "flos": 19458020216640.0, "grad_norm": 1.4859896548111982, "language_loss": 0.7786932, "learning_rate": 1.5740736878333507e-06, "loss": 0.80049849, "num_input_tokens_seen": 104246020, "step": 4834, "time_per_iteration": 2.523548126220703 }, { "auxiliary_loss_clip": 0.01153806, "auxiliary_loss_mlp": 0.01024585, "balance_loss_clip": 1.04928219, "balance_loss_mlp": 1.01663375, "epoch": 0.5813743762400048, "flos": 20594953132320.0, "grad_norm": 2.2962249426457366, "language_loss": 0.78179121, "learning_rate": 1.5733126179416143e-06, "loss": 0.80357516, "num_input_tokens_seen": 104260505, "step": 4835, "time_per_iteration": 2.503859281539917 }, { "auxiliary_loss_clip": 0.0116342, "auxiliary_loss_mlp": 0.01027029, "balance_loss_clip": 1.05072784, "balance_loss_mlp": 1.01925027, "epoch": 0.5814946191306439, "flos": 33178231530720.0, "grad_norm": 2.0137188611476375, "language_loss": 0.72808635, "learning_rate": 1.5725516127780137e-06, "loss": 0.74999082, "num_input_tokens_seen": 104282640, "step": 4836, "time_per_iteration": 2.598302125930786 }, { "auxiliary_loss_clip": 0.01169507, "auxiliary_loss_mlp": 0.01027573, "balance_loss_clip": 1.05010128, "balance_loss_mlp": 1.01992536, "epoch": 0.581614862021283, "flos": 16143286560960.0, "grad_norm": 2.579114993222021, "language_loss": 0.88359451, "learning_rate": 1.5717906724579943e-06, "loss": 0.90556526, "num_input_tokens_seen": 104299700, "step": 4837, "time_per_iteration": 2.467763900756836 }, { "auxiliary_loss_clip": 0.01145172, "auxiliary_loss_mlp": 0.01025064, "balance_loss_clip": 1.04862177, "balance_loss_mlp": 1.0179112, "epoch": 0.581735104911922, "flos": 33802647029760.0, "grad_norm": 2.095875295686702, "language_loss": 0.68351972, "learning_rate": 1.571029797096989e-06, "loss": 0.70522201, "num_input_tokens_seen": 104320805, "step": 4838, "time_per_iteration": 2.656186103820801 }, { "auxiliary_loss_clip": 0.01174172, "auxiliary_loss_mlp": 0.01031204, "balance_loss_clip": 1.04967046, "balance_loss_mlp": 1.02406967, "epoch": 0.5818553478025612, "flos": 23331166775520.0, "grad_norm": 2.1653248644960206, "language_loss": 0.78676641, "learning_rate": 1.570268986810423e-06, "loss": 0.80882013, "num_input_tokens_seen": 104340700, "step": 4839, "time_per_iteration": 2.471614360809326 }, { "auxiliary_loss_clip": 0.01150637, "auxiliary_loss_mlp": 0.01028494, "balance_loss_clip": 1.04952657, "balance_loss_mlp": 1.02137709, "epoch": 0.5819755906932003, "flos": 20996147137920.0, "grad_norm": 1.9135979680010695, "language_loss": 0.75389034, "learning_rate": 1.5695082417137096e-06, "loss": 0.77568161, "num_input_tokens_seen": 104358575, "step": 4840, "time_per_iteration": 2.499537467956543 }, { "auxiliary_loss_clip": 0.01147552, "auxiliary_loss_mlp": 0.01024831, "balance_loss_clip": 1.0476191, "balance_loss_mlp": 1.01812541, "epoch": 0.5820958335838393, "flos": 21431671368960.0, "grad_norm": 1.9357068570682139, "language_loss": 0.75386542, "learning_rate": 1.5687475619222539e-06, "loss": 0.77558929, "num_input_tokens_seen": 104378530, "step": 4841, "time_per_iteration": 2.569702625274658 }, { "auxiliary_loss_clip": 0.01146488, "auxiliary_loss_mlp": 0.01028705, "balance_loss_clip": 1.04754186, "balance_loss_mlp": 1.02107, "epoch": 0.5822160764744785, "flos": 17967476879040.0, "grad_norm": 2.126527762979449, "language_loss": 0.73536259, "learning_rate": 1.5679869475514496e-06, "loss": 0.75711453, "num_input_tokens_seen": 104395465, "step": 4842, "time_per_iteration": 2.4982659816741943 }, { "auxiliary_loss_clip": 0.01162767, "auxiliary_loss_mlp": 0.01025934, "balance_loss_clip": 1.05008328, "balance_loss_mlp": 1.01809347, "epoch": 0.5823363193651175, "flos": 23033861370240.0, "grad_norm": 2.600423390497603, "language_loss": 0.81278133, "learning_rate": 1.567226398716682e-06, "loss": 0.8346684, "num_input_tokens_seen": 104415380, "step": 4843, "time_per_iteration": 2.5437049865722656 }, { "auxiliary_loss_clip": 0.01157403, "auxiliary_loss_mlp": 0.01024984, "balance_loss_clip": 1.04977679, "balance_loss_mlp": 1.01764059, "epoch": 0.5824565622557566, "flos": 32891844881280.0, "grad_norm": 1.8394153395918256, "language_loss": 0.6164065, "learning_rate": 1.566465915533326e-06, "loss": 0.63823038, "num_input_tokens_seen": 104437410, "step": 4844, "time_per_iteration": 2.617631435394287 }, { "auxiliary_loss_clip": 0.01160228, "auxiliary_loss_mlp": 0.01024936, "balance_loss_clip": 1.04928052, "balance_loss_mlp": 1.0173099, "epoch": 0.5825768051463958, "flos": 22229677511040.0, "grad_norm": 2.199232854626378, "language_loss": 0.88294458, "learning_rate": 1.5657054981167458e-06, "loss": 0.90479618, "num_input_tokens_seen": 104456305, "step": 4845, "time_per_iteration": 2.5199854373931885 }, { "auxiliary_loss_clip": 0.01159942, "auxiliary_loss_mlp": 0.0103202, "balance_loss_clip": 1.05028725, "balance_loss_mlp": 1.02489114, "epoch": 0.5826970480370348, "flos": 28001565220320.0, "grad_norm": 1.7058033692484789, "language_loss": 0.6786139, "learning_rate": 1.5649451465822965e-06, "loss": 0.70053351, "num_input_tokens_seen": 104477695, "step": 4846, "time_per_iteration": 2.535196542739868 }, { "auxiliary_loss_clip": 0.01121196, "auxiliary_loss_mlp": 0.01029273, "balance_loss_clip": 1.0480926, "balance_loss_mlp": 1.02184665, "epoch": 0.5828172909276739, "flos": 17858057066880.0, "grad_norm": 1.5937386725026028, "language_loss": 0.83614516, "learning_rate": 1.5641848610453218e-06, "loss": 0.85764992, "num_input_tokens_seen": 104496355, "step": 4847, "time_per_iteration": 2.5582923889160156 }, { "auxiliary_loss_clip": 0.01159491, "auxiliary_loss_mlp": 0.0102335, "balance_loss_clip": 1.05081785, "balance_loss_mlp": 1.01576805, "epoch": 0.582937533818313, "flos": 19865248271520.0, "grad_norm": 2.1367089739811154, "language_loss": 0.85959792, "learning_rate": 1.563424641621158e-06, "loss": 0.88142627, "num_input_tokens_seen": 104515535, "step": 4848, "time_per_iteration": 3.212151288986206 }, { "auxiliary_loss_clip": 0.01156156, "auxiliary_loss_mlp": 0.01023852, "balance_loss_clip": 1.05237627, "balance_loss_mlp": 1.01595497, "epoch": 0.5830577767089521, "flos": 26870738187840.0, "grad_norm": 2.1220538345812696, "language_loss": 0.70040154, "learning_rate": 1.5626644884251282e-06, "loss": 0.72220159, "num_input_tokens_seen": 104535055, "step": 4849, "time_per_iteration": 3.304354190826416 }, { "auxiliary_loss_clip": 0.01174053, "auxiliary_loss_mlp": 0.01024068, "balance_loss_clip": 1.04988408, "balance_loss_mlp": 1.01726687, "epoch": 0.5831780195995911, "flos": 25298209207200.0, "grad_norm": 1.6576492984790392, "language_loss": 0.88006485, "learning_rate": 1.5619044015725488e-06, "loss": 0.90204608, "num_input_tokens_seen": 104554745, "step": 4850, "time_per_iteration": 2.4756851196289062 }, { "auxiliary_loss_clip": 0.01186568, "auxiliary_loss_mlp": 0.01033139, "balance_loss_clip": 1.05739021, "balance_loss_mlp": 1.02500856, "epoch": 0.5832982624902303, "flos": 14756990993280.0, "grad_norm": 2.1544143189810474, "language_loss": 0.86968839, "learning_rate": 1.5611443811787224e-06, "loss": 0.8918854, "num_input_tokens_seen": 104568870, "step": 4851, "time_per_iteration": 2.424057960510254 }, { "auxiliary_loss_clip": 0.01160421, "auxiliary_loss_mlp": 0.01030348, "balance_loss_clip": 1.0508672, "balance_loss_mlp": 1.02376127, "epoch": 0.5834185053808694, "flos": 20444558457120.0, "grad_norm": 2.2776766882832726, "language_loss": 0.69665098, "learning_rate": 1.560384427358945e-06, "loss": 0.71855867, "num_input_tokens_seen": 104588415, "step": 4852, "time_per_iteration": 3.227605104446411 }, { "auxiliary_loss_clip": 0.01140557, "auxiliary_loss_mlp": 0.01031464, "balance_loss_clip": 1.04270947, "balance_loss_mlp": 1.02419233, "epoch": 0.5835387482715084, "flos": 27200398385760.0, "grad_norm": 1.5292920662940959, "language_loss": 0.72718465, "learning_rate": 1.5596245402284998e-06, "loss": 0.74890488, "num_input_tokens_seen": 104611940, "step": 4853, "time_per_iteration": 2.5801162719726562 }, { "auxiliary_loss_clip": 0.01166441, "auxiliary_loss_mlp": 0.0103026, "balance_loss_clip": 1.05384004, "balance_loss_mlp": 1.02254748, "epoch": 0.5836589911621476, "flos": 16654618717920.0, "grad_norm": 1.904215444437234, "language_loss": 0.81830102, "learning_rate": 1.5588647199026619e-06, "loss": 0.84026802, "num_input_tokens_seen": 104629675, "step": 4854, "time_per_iteration": 2.4648597240448 }, { "auxiliary_loss_clip": 0.01179594, "auxiliary_loss_mlp": 0.01024634, "balance_loss_clip": 1.05296516, "balance_loss_mlp": 1.01687908, "epoch": 0.5837792340527866, "flos": 20446821225600.0, "grad_norm": 2.145918353786838, "language_loss": 0.87292564, "learning_rate": 1.5581049664966956e-06, "loss": 0.89496797, "num_input_tokens_seen": 104647435, "step": 4855, "time_per_iteration": 3.1140661239624023 }, { "auxiliary_loss_clip": 0.01021403, "auxiliary_loss_mlp": 0.01003837, "balance_loss_clip": 1.01038361, "balance_loss_mlp": 1.00267506, "epoch": 0.5838994769434257, "flos": 65995489662720.0, "grad_norm": 0.9861135061878473, "language_loss": 0.65083241, "learning_rate": 1.5573452801258545e-06, "loss": 0.67108476, "num_input_tokens_seen": 104694605, "step": 4856, "time_per_iteration": 3.1018810272216797 }, { "auxiliary_loss_clip": 0.01173324, "auxiliary_loss_mlp": 0.01028856, "balance_loss_clip": 1.05519843, "balance_loss_mlp": 1.02145278, "epoch": 0.5840197198340649, "flos": 21470527131360.0, "grad_norm": 2.2488248431195683, "language_loss": 0.63671517, "learning_rate": 1.5565856609053824e-06, "loss": 0.658737, "num_input_tokens_seen": 104713400, "step": 4857, "time_per_iteration": 2.679495096206665 }, { "auxiliary_loss_clip": 0.01179174, "auxiliary_loss_mlp": 0.01031837, "balance_loss_clip": 1.05341434, "balance_loss_mlp": 1.02434468, "epoch": 0.5841399627247039, "flos": 19135148324160.0, "grad_norm": 1.7537266601078603, "language_loss": 0.79822648, "learning_rate": 1.5558261089505127e-06, "loss": 0.82033658, "num_input_tokens_seen": 104732130, "step": 4858, "time_per_iteration": 2.4452319145202637 }, { "auxiliary_loss_clip": 0.01164236, "auxiliary_loss_mlp": 0.01030628, "balance_loss_clip": 1.05108547, "balance_loss_mlp": 1.02302217, "epoch": 0.584260205615343, "flos": 26425695962400.0, "grad_norm": 1.9798133727763525, "language_loss": 0.79688787, "learning_rate": 1.5550666243764697e-06, "loss": 0.81883645, "num_input_tokens_seen": 104750290, "step": 4859, "time_per_iteration": 2.531619071960449 }, { "auxiliary_loss_clip": 0.01163293, "auxiliary_loss_mlp": 0.0102747, "balance_loss_clip": 1.05057716, "balance_loss_mlp": 1.02043056, "epoch": 0.584380448505982, "flos": 13881812080800.0, "grad_norm": 2.171822579970041, "language_loss": 0.77659249, "learning_rate": 1.554307207298465e-06, "loss": 0.79850012, "num_input_tokens_seen": 104768550, "step": 4860, "time_per_iteration": 2.4967780113220215 }, { "auxiliary_loss_clip": 0.01179606, "auxiliary_loss_mlp": 0.01029447, "balance_loss_clip": 1.0518651, "balance_loss_mlp": 1.02164745, "epoch": 0.5845006913966212, "flos": 21543713119200.0, "grad_norm": 1.8312584669621883, "language_loss": 0.78658199, "learning_rate": 1.553547857831704e-06, "loss": 0.80867255, "num_input_tokens_seen": 104785060, "step": 4861, "time_per_iteration": 2.4445626735687256 }, { "auxiliary_loss_clip": 0.01072127, "auxiliary_loss_mlp": 0.01004617, "balance_loss_clip": 1.01294672, "balance_loss_mlp": 1.00343132, "epoch": 0.5846209342872603, "flos": 58375461411360.0, "grad_norm": 0.8792923247799113, "language_loss": 0.64121222, "learning_rate": 1.5527885760913771e-06, "loss": 0.66197968, "num_input_tokens_seen": 104834950, "step": 4862, "time_per_iteration": 2.891733169555664 }, { "auxiliary_loss_clip": 0.01149228, "auxiliary_loss_mlp": 0.01026465, "balance_loss_clip": 1.05104303, "balance_loss_mlp": 1.01982224, "epoch": 0.5847411771778993, "flos": 18588049263360.0, "grad_norm": 1.5776268001620797, "language_loss": 0.76475477, "learning_rate": 1.552029362192668e-06, "loss": 0.78651166, "num_input_tokens_seen": 104854210, "step": 4863, "time_per_iteration": 2.525087356567383 }, { "auxiliary_loss_clip": 0.01129482, "auxiliary_loss_mlp": 0.01023958, "balance_loss_clip": 1.04608262, "balance_loss_mlp": 1.01673973, "epoch": 0.5848614200685385, "flos": 24240783664320.0, "grad_norm": 1.7742169596793487, "language_loss": 0.7229073, "learning_rate": 1.5512702162507478e-06, "loss": 0.74444175, "num_input_tokens_seen": 104874525, "step": 4864, "time_per_iteration": 2.5740468502044678 }, { "auxiliary_loss_clip": 0.010496, "auxiliary_loss_mlp": 0.01000885, "balance_loss_clip": 1.01117122, "balance_loss_mlp": 0.99971038, "epoch": 0.5849816629591775, "flos": 71660256245280.0, "grad_norm": 1.5570367859712917, "language_loss": 0.55763483, "learning_rate": 1.5505111383807792e-06, "loss": 0.57813972, "num_input_tokens_seen": 104937195, "step": 4865, "time_per_iteration": 3.171213150024414 }, { "auxiliary_loss_clip": 0.01122329, "auxiliary_loss_mlp": 0.01023772, "balance_loss_clip": 1.04355073, "balance_loss_mlp": 1.01659524, "epoch": 0.5851019058498166, "flos": 23802098740800.0, "grad_norm": 1.8682285420788078, "language_loss": 0.80643475, "learning_rate": 1.5497521286979138e-06, "loss": 0.8278957, "num_input_tokens_seen": 104957435, "step": 4866, "time_per_iteration": 2.5819919109344482 }, { "auxiliary_loss_clip": 0.01139285, "auxiliary_loss_mlp": 0.0102854, "balance_loss_clip": 1.04806674, "balance_loss_mlp": 1.02088702, "epoch": 0.5852221487404557, "flos": 24388520484480.0, "grad_norm": 1.857057237472617, "language_loss": 0.73918617, "learning_rate": 1.5489931873172927e-06, "loss": 0.7608645, "num_input_tokens_seen": 104978755, "step": 4867, "time_per_iteration": 2.5891621112823486 }, { "auxiliary_loss_clip": 0.01089071, "auxiliary_loss_mlp": 0.01029213, "balance_loss_clip": 1.03966856, "balance_loss_mlp": 1.02142239, "epoch": 0.5853423916310948, "flos": 27271429356000.0, "grad_norm": 1.8551876551945947, "language_loss": 0.79043871, "learning_rate": 1.5482343143540467e-06, "loss": 0.81162155, "num_input_tokens_seen": 105000020, "step": 4868, "time_per_iteration": 2.6467278003692627 }, { "auxiliary_loss_clip": 0.01132582, "auxiliary_loss_mlp": 0.00761489, "balance_loss_clip": 1.04574442, "balance_loss_mlp": 1.00025594, "epoch": 0.5854626345217339, "flos": 11983789269600.0, "grad_norm": 1.9947758799545219, "language_loss": 0.8330701, "learning_rate": 1.547475509923295e-06, "loss": 0.85201085, "num_input_tokens_seen": 105017060, "step": 4869, "time_per_iteration": 2.526979446411133 }, { "auxiliary_loss_clip": 0.01027652, "auxiliary_loss_mlp": 0.01005867, "balance_loss_clip": 1.01059854, "balance_loss_mlp": 1.00460899, "epoch": 0.585582877412373, "flos": 64342346271840.0, "grad_norm": 0.7432735775159861, "language_loss": 0.56074691, "learning_rate": 1.5467167741401495e-06, "loss": 0.58108211, "num_input_tokens_seen": 105078540, "step": 4870, "time_per_iteration": 3.173628330230713 }, { "auxiliary_loss_clip": 0.01146862, "auxiliary_loss_mlp": 0.01031219, "balance_loss_clip": 1.04545188, "balance_loss_mlp": 1.02364278, "epoch": 0.5857031203030121, "flos": 17011928586720.0, "grad_norm": 2.2266098008349338, "language_loss": 0.70998549, "learning_rate": 1.5459581071197083e-06, "loss": 0.73176628, "num_input_tokens_seen": 105094200, "step": 4871, "time_per_iteration": 2.5054800510406494 }, { "auxiliary_loss_clip": 0.01164827, "auxiliary_loss_mlp": 0.01025921, "balance_loss_clip": 1.05203545, "balance_loss_mlp": 1.01826799, "epoch": 0.5858233631936511, "flos": 20885685733920.0, "grad_norm": 2.397965280156941, "language_loss": 0.8306706, "learning_rate": 1.5451995089770624e-06, "loss": 0.8525781, "num_input_tokens_seen": 105113985, "step": 4872, "time_per_iteration": 2.474717855453491 }, { "auxiliary_loss_clip": 0.01174317, "auxiliary_loss_mlp": 0.0102104, "balance_loss_clip": 1.05121911, "balance_loss_mlp": 1.01427507, "epoch": 0.5859436060842903, "flos": 23191906357920.0, "grad_norm": 1.9866938371780158, "language_loss": 0.71783119, "learning_rate": 1.5444409798272885e-06, "loss": 0.73978472, "num_input_tokens_seen": 105138075, "step": 4873, "time_per_iteration": 2.5614211559295654 }, { "auxiliary_loss_clip": 0.01136292, "auxiliary_loss_mlp": 0.01025439, "balance_loss_clip": 1.0474503, "balance_loss_mlp": 1.01817036, "epoch": 0.5860638489749294, "flos": 22492652690880.0, "grad_norm": 1.7264627151106018, "language_loss": 0.80405891, "learning_rate": 1.543682519785456e-06, "loss": 0.8256762, "num_input_tokens_seen": 105156555, "step": 4874, "time_per_iteration": 3.3256161212921143 }, { "auxiliary_loss_clip": 0.01147499, "auxiliary_loss_mlp": 0.0103083, "balance_loss_clip": 1.04777026, "balance_loss_mlp": 1.02360559, "epoch": 0.5861840918655684, "flos": 17566246956480.0, "grad_norm": 2.5519633182853454, "language_loss": 0.80137026, "learning_rate": 1.5429241289666219e-06, "loss": 0.8231535, "num_input_tokens_seen": 105174055, "step": 4875, "time_per_iteration": 3.260133743286133 }, { "auxiliary_loss_clip": 0.01140122, "auxiliary_loss_mlp": 0.01026567, "balance_loss_clip": 1.04596829, "balance_loss_mlp": 1.0196377, "epoch": 0.5863043347562076, "flos": 25556156012640.0, "grad_norm": 2.2280149243684866, "language_loss": 0.69999051, "learning_rate": 1.5421658074858342e-06, "loss": 0.7216574, "num_input_tokens_seen": 105192160, "step": 4876, "time_per_iteration": 2.5756754875183105 }, { "auxiliary_loss_clip": 0.01144852, "auxiliary_loss_mlp": 0.01031959, "balance_loss_clip": 1.04862547, "balance_loss_mlp": 1.02464509, "epoch": 0.5864245776468466, "flos": 20667528531840.0, "grad_norm": 2.747921192403719, "language_loss": 0.66625059, "learning_rate": 1.5414075554581298e-06, "loss": 0.68801868, "num_input_tokens_seen": 105210205, "step": 4877, "time_per_iteration": 2.5014352798461914 }, { "auxiliary_loss_clip": 0.01179829, "auxiliary_loss_mlp": 0.01024507, "balance_loss_clip": 1.05313754, "balance_loss_mlp": 1.01749146, "epoch": 0.5865448205374857, "flos": 28913911798080.0, "grad_norm": 2.205011062027092, "language_loss": 0.78382987, "learning_rate": 1.5406493729985348e-06, "loss": 0.80587322, "num_input_tokens_seen": 105229400, "step": 4878, "time_per_iteration": 3.280076026916504 }, { "auxiliary_loss_clip": 0.01124462, "auxiliary_loss_mlp": 0.0076182, "balance_loss_clip": 1.04913712, "balance_loss_mlp": 1.00032365, "epoch": 0.5866650634281249, "flos": 25842578579040.0, "grad_norm": 2.083581575365003, "language_loss": 0.72307324, "learning_rate": 1.5398912602220644e-06, "loss": 0.74193609, "num_input_tokens_seen": 105248675, "step": 4879, "time_per_iteration": 2.598153591156006 }, { "auxiliary_loss_clip": 0.01130064, "auxiliary_loss_mlp": 0.01019932, "balance_loss_clip": 1.04672372, "balance_loss_mlp": 1.01240134, "epoch": 0.5867853063187639, "flos": 17052328778400.0, "grad_norm": 2.0746142782160732, "language_loss": 0.7843129, "learning_rate": 1.539133217243724e-06, "loss": 0.80581284, "num_input_tokens_seen": 105265695, "step": 4880, "time_per_iteration": 2.5572755336761475 }, { "auxiliary_loss_clip": 0.01140005, "auxiliary_loss_mlp": 0.01030549, "balance_loss_clip": 1.04601598, "balance_loss_mlp": 1.0225563, "epoch": 0.586905549209403, "flos": 24645030611520.0, "grad_norm": 1.9800553752173475, "language_loss": 0.75964016, "learning_rate": 1.5383752441785081e-06, "loss": 0.78134573, "num_input_tokens_seen": 105284920, "step": 4881, "time_per_iteration": 3.3301680088043213 }, { "auxiliary_loss_clip": 0.01167361, "auxiliary_loss_mlp": 0.0103433, "balance_loss_clip": 1.05092204, "balance_loss_mlp": 1.02684402, "epoch": 0.5870257921000421, "flos": 14720541667200.0, "grad_norm": 2.183939472269457, "language_loss": 0.85380894, "learning_rate": 1.5376173411414003e-06, "loss": 0.87582582, "num_input_tokens_seen": 105302960, "step": 4882, "time_per_iteration": 2.4682133197784424 }, { "auxiliary_loss_clip": 0.01149234, "auxiliary_loss_mlp": 0.01025102, "balance_loss_clip": 1.04596019, "balance_loss_mlp": 1.01756191, "epoch": 0.5871460349906812, "flos": 23915002498080.0, "grad_norm": 1.8387089471448252, "language_loss": 0.79262114, "learning_rate": 1.5368595082473753e-06, "loss": 0.81436449, "num_input_tokens_seen": 105321260, "step": 4883, "time_per_iteration": 2.549302577972412 }, { "auxiliary_loss_clip": 0.01164231, "auxiliary_loss_mlp": 0.0102393, "balance_loss_clip": 1.04932332, "balance_loss_mlp": 1.01714957, "epoch": 0.5872662778813202, "flos": 22164177752640.0, "grad_norm": 1.7381353181334662, "language_loss": 0.78301001, "learning_rate": 1.5361017456113935e-06, "loss": 0.80489159, "num_input_tokens_seen": 105341610, "step": 4884, "time_per_iteration": 2.4896793365478516 }, { "auxiliary_loss_clip": 0.01164046, "auxiliary_loss_mlp": 0.01026568, "balance_loss_clip": 1.04838204, "balance_loss_mlp": 1.01847935, "epoch": 0.5873865207719594, "flos": 18441928706400.0, "grad_norm": 2.5282817574690704, "language_loss": 0.85344523, "learning_rate": 1.5353440533484085e-06, "loss": 0.87535131, "num_input_tokens_seen": 105360465, "step": 4885, "time_per_iteration": 2.4721362590789795 }, { "auxiliary_loss_clip": 0.01148975, "auxiliary_loss_mlp": 0.0102606, "balance_loss_clip": 1.04772043, "balance_loss_mlp": 1.01772773, "epoch": 0.5875067636625985, "flos": 54015328006080.0, "grad_norm": 1.6744540638719236, "language_loss": 0.6557501, "learning_rate": 1.534586431573361e-06, "loss": 0.67750049, "num_input_tokens_seen": 105385405, "step": 4886, "time_per_iteration": 2.8042869567871094 }, { "auxiliary_loss_clip": 0.01105458, "auxiliary_loss_mlp": 0.0102668, "balance_loss_clip": 1.04066253, "balance_loss_mlp": 1.01842451, "epoch": 0.5876270065532375, "flos": 27995710755840.0, "grad_norm": 2.0276121014754764, "language_loss": 0.78979683, "learning_rate": 1.5338288804011817e-06, "loss": 0.81111813, "num_input_tokens_seen": 105404905, "step": 4887, "time_per_iteration": 2.676326274871826 }, { "auxiliary_loss_clip": 0.01144332, "auxiliary_loss_mlp": 0.01026067, "balance_loss_clip": 1.0457443, "balance_loss_mlp": 1.01825261, "epoch": 0.5877472494438767, "flos": 21361466488800.0, "grad_norm": 1.9812868592555832, "language_loss": 0.70823586, "learning_rate": 1.533071399946791e-06, "loss": 0.72993982, "num_input_tokens_seen": 105423650, "step": 4888, "time_per_iteration": 2.5179121494293213 }, { "auxiliary_loss_clip": 0.01148271, "auxiliary_loss_mlp": 0.01032952, "balance_loss_clip": 1.04794216, "balance_loss_mlp": 1.02551651, "epoch": 0.5878674923345157, "flos": 22383017376960.0, "grad_norm": 1.8977124760969521, "language_loss": 0.56960607, "learning_rate": 1.5323139903250977e-06, "loss": 0.59141827, "num_input_tokens_seen": 105444255, "step": 4889, "time_per_iteration": 2.5313079357147217 }, { "auxiliary_loss_clip": 0.01154022, "auxiliary_loss_mlp": 0.01029156, "balance_loss_clip": 1.05208611, "balance_loss_mlp": 1.02109146, "epoch": 0.5879877352251548, "flos": 21868668195360.0, "grad_norm": 1.6006293080337932, "language_loss": 0.76914525, "learning_rate": 1.5315566516510002e-06, "loss": 0.79097706, "num_input_tokens_seen": 105462425, "step": 4890, "time_per_iteration": 2.504659414291382 }, { "auxiliary_loss_clip": 0.01177644, "auxiliary_loss_mlp": 0.01033802, "balance_loss_clip": 1.05219293, "balance_loss_mlp": 1.0262084, "epoch": 0.5881079781157939, "flos": 17493815224800.0, "grad_norm": 1.7471940438249156, "language_loss": 0.67304122, "learning_rate": 1.5307993840393857e-06, "loss": 0.69515568, "num_input_tokens_seen": 105480505, "step": 4891, "time_per_iteration": 2.438293218612671 }, { "auxiliary_loss_clip": 0.01175899, "auxiliary_loss_mlp": 0.01025149, "balance_loss_clip": 1.05013204, "balance_loss_mlp": 1.01860476, "epoch": 0.588228221006433, "flos": 22601857001280.0, "grad_norm": 1.8537699802932253, "language_loss": 0.79997945, "learning_rate": 1.530042187605132e-06, "loss": 0.82199001, "num_input_tokens_seen": 105499760, "step": 4892, "time_per_iteration": 2.5624186992645264 }, { "auxiliary_loss_clip": 0.01165314, "auxiliary_loss_mlp": 0.00761534, "balance_loss_clip": 1.05211604, "balance_loss_mlp": 1.00033581, "epoch": 0.5883484638970721, "flos": 26176943898720.0, "grad_norm": 1.4628151623581684, "language_loss": 0.84170526, "learning_rate": 1.5292850624631044e-06, "loss": 0.86097372, "num_input_tokens_seen": 105521955, "step": 4893, "time_per_iteration": 2.542515754699707 }, { "auxiliary_loss_clip": 0.01160831, "auxiliary_loss_mlp": 0.01028976, "balance_loss_clip": 1.05129552, "balance_loss_mlp": 1.02171063, "epoch": 0.5884687067877111, "flos": 30443741901600.0, "grad_norm": 1.9046309102667403, "language_loss": 0.80158043, "learning_rate": 1.5285280087281593e-06, "loss": 0.82347858, "num_input_tokens_seen": 105542685, "step": 4894, "time_per_iteration": 2.5343985557556152 }, { "auxiliary_loss_clip": 0.01052437, "auxiliary_loss_mlp": 0.010021, "balance_loss_clip": 1.01140547, "balance_loss_mlp": 1.00080633, "epoch": 0.5885889496783503, "flos": 70507562172480.0, "grad_norm": 0.6431791399460807, "language_loss": 0.56611562, "learning_rate": 1.5277710265151398e-06, "loss": 0.58666098, "num_input_tokens_seen": 105612165, "step": 4895, "time_per_iteration": 3.2724106311798096 }, { "auxiliary_loss_clip": 0.01163448, "auxiliary_loss_mlp": 0.01033805, "balance_loss_clip": 1.0494653, "balance_loss_mlp": 1.02557981, "epoch": 0.5887091925689893, "flos": 19098770832000.0, "grad_norm": 2.9163719479758163, "language_loss": 0.77600396, "learning_rate": 1.5270141159388803e-06, "loss": 0.79797643, "num_input_tokens_seen": 105629185, "step": 4896, "time_per_iteration": 2.467050313949585 }, { "auxiliary_loss_clip": 0.01179045, "auxiliary_loss_mlp": 0.01024194, "balance_loss_clip": 1.05167305, "balance_loss_mlp": 1.01684451, "epoch": 0.5888294354596284, "flos": 23294286445920.0, "grad_norm": 3.7214854618586717, "language_loss": 0.80547857, "learning_rate": 1.526257277114203e-06, "loss": 0.82751107, "num_input_tokens_seen": 105650260, "step": 4897, "time_per_iteration": 2.5188703536987305 }, { "auxiliary_loss_clip": 0.01146972, "auxiliary_loss_mlp": 0.01026808, "balance_loss_clip": 1.05046272, "balance_loss_mlp": 1.01936328, "epoch": 0.5889496783502676, "flos": 21981536035680.0, "grad_norm": 1.828995280624483, "language_loss": 0.79416883, "learning_rate": 1.5255005101559201e-06, "loss": 0.81590652, "num_input_tokens_seen": 105667870, "step": 4898, "time_per_iteration": 2.554734468460083 }, { "auxiliary_loss_clip": 0.0116246, "auxiliary_loss_mlp": 0.01024031, "balance_loss_clip": 1.05060887, "balance_loss_mlp": 1.01666689, "epoch": 0.5890699212409066, "flos": 21685236305280.0, "grad_norm": 2.187982268906777, "language_loss": 0.7664479, "learning_rate": 1.524743815178833e-06, "loss": 0.78831279, "num_input_tokens_seen": 105685830, "step": 4899, "time_per_iteration": 2.5254743099212646 }, { "auxiliary_loss_clip": 0.01147853, "auxiliary_loss_mlp": 0.01025364, "balance_loss_clip": 1.04464555, "balance_loss_mlp": 1.01836956, "epoch": 0.5891901641315457, "flos": 19464557103360.0, "grad_norm": 1.8097574207629787, "language_loss": 0.8108269, "learning_rate": 1.5239871922977315e-06, "loss": 0.83255911, "num_input_tokens_seen": 105705745, "step": 4900, "time_per_iteration": 3.3562116622924805 }, { "auxiliary_loss_clip": 0.01147075, "auxiliary_loss_mlp": 0.01025293, "balance_loss_clip": 1.04627264, "balance_loss_mlp": 1.01838195, "epoch": 0.5893104070221848, "flos": 19609887487200.0, "grad_norm": 1.9678975871571012, "language_loss": 0.89685494, "learning_rate": 1.523230641627394e-06, "loss": 0.91857862, "num_input_tokens_seen": 105724730, "step": 4901, "time_per_iteration": 3.318577527999878 }, { "auxiliary_loss_clip": 0.01120174, "auxiliary_loss_mlp": 0.01028707, "balance_loss_clip": 1.04090858, "balance_loss_mlp": 1.02101839, "epoch": 0.5894306499128239, "flos": 29060068272000.0, "grad_norm": 1.9769052833781868, "language_loss": 0.72738242, "learning_rate": 1.5224741632825888e-06, "loss": 0.74887121, "num_input_tokens_seen": 105744920, "step": 4902, "time_per_iteration": 2.6731598377227783 }, { "auxiliary_loss_clip": 0.01181836, "auxiliary_loss_mlp": 0.01027064, "balance_loss_clip": 1.0541296, "balance_loss_mlp": 1.0187906, "epoch": 0.589550892803463, "flos": 42298895785440.0, "grad_norm": 1.729791300098252, "language_loss": 0.69443893, "learning_rate": 1.521717757378074e-06, "loss": 0.716528, "num_input_tokens_seen": 105765465, "step": 4903, "time_per_iteration": 2.655749559402466 }, { "auxiliary_loss_clip": 0.01169341, "auxiliary_loss_mlp": 0.01028772, "balance_loss_clip": 1.05161846, "balance_loss_mlp": 1.0203793, "epoch": 0.5896711356941021, "flos": 14137065114240.0, "grad_norm": 2.6885556624766296, "language_loss": 0.69465697, "learning_rate": 1.5209614240285943e-06, "loss": 0.71663809, "num_input_tokens_seen": 105783120, "step": 4904, "time_per_iteration": 3.229659080505371 }, { "auxiliary_loss_clip": 0.01175346, "auxiliary_loss_mlp": 0.00762052, "balance_loss_clip": 1.04931331, "balance_loss_mlp": 1.00026143, "epoch": 0.5897913785847412, "flos": 17201358609120.0, "grad_norm": 2.0581334980866486, "language_loss": 0.84386587, "learning_rate": 1.520205163348887e-06, "loss": 0.86323988, "num_input_tokens_seen": 105801055, "step": 4905, "time_per_iteration": 2.466888904571533 }, { "auxiliary_loss_clip": 0.01044528, "auxiliary_loss_mlp": 0.01001346, "balance_loss_clip": 1.01177609, "balance_loss_mlp": 0.99996293, "epoch": 0.5899116214753802, "flos": 48794173358880.0, "grad_norm": 0.7795193280676392, "language_loss": 0.5697127, "learning_rate": 1.519448975453674e-06, "loss": 0.59017146, "num_input_tokens_seen": 105856155, "step": 4906, "time_per_iteration": 3.041015148162842 }, { "auxiliary_loss_clip": 0.01166264, "auxiliary_loss_mlp": 0.00762339, "balance_loss_clip": 1.05352688, "balance_loss_mlp": 1.00035477, "epoch": 0.5900318643660194, "flos": 21103663351200.0, "grad_norm": 1.9558440071331789, "language_loss": 0.76453233, "learning_rate": 1.5186928604576696e-06, "loss": 0.78381836, "num_input_tokens_seen": 105873350, "step": 4907, "time_per_iteration": 3.251967191696167 }, { "auxiliary_loss_clip": 0.01150677, "auxiliary_loss_mlp": 0.01026612, "balance_loss_clip": 1.04798329, "balance_loss_mlp": 1.01930404, "epoch": 0.5901521072566585, "flos": 21178393768320.0, "grad_norm": 2.0195276732262952, "language_loss": 0.7708379, "learning_rate": 1.5179368184755752e-06, "loss": 0.79261076, "num_input_tokens_seen": 105891435, "step": 4908, "time_per_iteration": 2.5862503051757812 }, { "auxiliary_loss_clip": 0.01149943, "auxiliary_loss_mlp": 0.01024057, "balance_loss_clip": 1.04987252, "balance_loss_mlp": 1.01680279, "epoch": 0.5902723501472975, "flos": 20225970251520.0, "grad_norm": 1.46688278446821, "language_loss": 0.82449985, "learning_rate": 1.5171808496220821e-06, "loss": 0.84623981, "num_input_tokens_seen": 105910190, "step": 4909, "time_per_iteration": 2.5511081218719482 }, { "auxiliary_loss_clip": 0.0115364, "auxiliary_loss_mlp": 0.01032628, "balance_loss_clip": 1.04685271, "balance_loss_mlp": 1.02558851, "epoch": 0.5903925930379367, "flos": 22964410746240.0, "grad_norm": 1.8320709546204597, "language_loss": 0.81533873, "learning_rate": 1.5164249540118708e-06, "loss": 0.83720136, "num_input_tokens_seen": 105929315, "step": 4910, "time_per_iteration": 2.5131492614746094 }, { "auxiliary_loss_clip": 0.01112638, "auxiliary_loss_mlp": 0.01025916, "balance_loss_clip": 1.04492807, "balance_loss_mlp": 1.01872158, "epoch": 0.5905128359285757, "flos": 23367723852480.0, "grad_norm": 1.6046856373849196, "language_loss": 0.83131778, "learning_rate": 1.5156691317596093e-06, "loss": 0.85270333, "num_input_tokens_seen": 105950740, "step": 4911, "time_per_iteration": 2.656482458114624 }, { "auxiliary_loss_clip": 0.01166151, "auxiliary_loss_mlp": 0.00761922, "balance_loss_clip": 1.05026162, "balance_loss_mlp": 1.00024605, "epoch": 0.5906330788192148, "flos": 28032339666720.0, "grad_norm": 2.038671374507978, "language_loss": 0.66574198, "learning_rate": 1.5149133829799556e-06, "loss": 0.68502271, "num_input_tokens_seen": 105968735, "step": 4912, "time_per_iteration": 2.5309925079345703 }, { "auxiliary_loss_clip": 0.01154809, "auxiliary_loss_mlp": 0.01027111, "balance_loss_clip": 1.04849935, "balance_loss_mlp": 1.01997316, "epoch": 0.590753321709854, "flos": 18477947028960.0, "grad_norm": 1.741443195577454, "language_loss": 0.80583525, "learning_rate": 1.5141577077875556e-06, "loss": 0.82765448, "num_input_tokens_seen": 105986060, "step": 4913, "time_per_iteration": 2.513505458831787 }, { "auxiliary_loss_clip": 0.01164733, "auxiliary_loss_mlp": 0.01030091, "balance_loss_clip": 1.04961705, "balance_loss_mlp": 1.02282214, "epoch": 0.590873564600493, "flos": 16873709760960.0, "grad_norm": 1.993393928969475, "language_loss": 0.72551054, "learning_rate": 1.5134021062970451e-06, "loss": 0.74745882, "num_input_tokens_seen": 106004440, "step": 4914, "time_per_iteration": 2.4926671981811523 }, { "auxiliary_loss_clip": 0.01128006, "auxiliary_loss_mlp": 0.01026871, "balance_loss_clip": 1.04865456, "balance_loss_mlp": 1.01925385, "epoch": 0.5909938074911321, "flos": 13516169477280.0, "grad_norm": 1.8444853038860654, "language_loss": 0.80729228, "learning_rate": 1.5126465786230483e-06, "loss": 0.82884109, "num_input_tokens_seen": 106021215, "step": 4915, "time_per_iteration": 2.5605287551879883 }, { "auxiliary_loss_clip": 0.01177372, "auxiliary_loss_mlp": 0.01027617, "balance_loss_clip": 1.05220914, "balance_loss_mlp": 1.02064061, "epoch": 0.5911140503817712, "flos": 26024070953280.0, "grad_norm": 1.694248328488331, "language_loss": 0.82104564, "learning_rate": 1.5118911248801787e-06, "loss": 0.84309554, "num_input_tokens_seen": 106039225, "step": 4916, "time_per_iteration": 2.496915340423584 }, { "auxiliary_loss_clip": 0.01159757, "auxiliary_loss_mlp": 0.01025752, "balance_loss_clip": 1.04972744, "balance_loss_mlp": 1.01901031, "epoch": 0.5912342932724103, "flos": 23258735043840.0, "grad_norm": 2.092348925346796, "language_loss": 0.80269569, "learning_rate": 1.5111357451830364e-06, "loss": 0.82455075, "num_input_tokens_seen": 106057920, "step": 4917, "time_per_iteration": 2.5146658420562744 }, { "auxiliary_loss_clip": 0.01164266, "auxiliary_loss_mlp": 0.01026553, "balance_loss_clip": 1.04912722, "balance_loss_mlp": 1.01963925, "epoch": 0.5913545361630493, "flos": 19573043074560.0, "grad_norm": 2.00623621541242, "language_loss": 0.71169221, "learning_rate": 1.5103804396462131e-06, "loss": 0.73360038, "num_input_tokens_seen": 106077855, "step": 4918, "time_per_iteration": 2.477827548980713 }, { "auxiliary_loss_clip": 0.01168846, "auxiliary_loss_mlp": 0.01031275, "balance_loss_clip": 1.04937553, "balance_loss_mlp": 1.02343369, "epoch": 0.5914747790536885, "flos": 26213536892640.0, "grad_norm": 2.012068855053061, "language_loss": 0.80006164, "learning_rate": 1.5096252083842877e-06, "loss": 0.82206285, "num_input_tokens_seen": 106097065, "step": 4919, "time_per_iteration": 2.5477182865142822 }, { "auxiliary_loss_clip": 0.01160479, "auxiliary_loss_mlp": 0.01026332, "balance_loss_clip": 1.04611719, "balance_loss_mlp": 1.01873875, "epoch": 0.5915950219443276, "flos": 27417549912960.0, "grad_norm": 1.7383464376506468, "language_loss": 0.8522861, "learning_rate": 1.5088700515118285e-06, "loss": 0.87415421, "num_input_tokens_seen": 106116385, "step": 4920, "time_per_iteration": 2.5351033210754395 }, { "auxiliary_loss_clip": 0.01131955, "auxiliary_loss_mlp": 0.0102821, "balance_loss_clip": 1.04765582, "balance_loss_mlp": 1.0211556, "epoch": 0.5917152648349666, "flos": 21907883127360.0, "grad_norm": 1.5652361565362078, "language_loss": 0.6640197, "learning_rate": 1.508114969143392e-06, "loss": 0.68562138, "num_input_tokens_seen": 106136370, "step": 4921, "time_per_iteration": 2.564302682876587 }, { "auxiliary_loss_clip": 0.01150822, "auxiliary_loss_mlp": 0.0102962, "balance_loss_clip": 1.04674959, "balance_loss_mlp": 1.02256322, "epoch": 0.5918355077256057, "flos": 28109189184480.0, "grad_norm": 1.5883853163910182, "language_loss": 0.77571118, "learning_rate": 1.5073599613935238e-06, "loss": 0.79751551, "num_input_tokens_seen": 106158490, "step": 4922, "time_per_iteration": 2.5826756954193115 }, { "auxiliary_loss_clip": 0.01148544, "auxiliary_loss_mlp": 0.0102926, "balance_loss_clip": 1.04695725, "balance_loss_mlp": 1.02201176, "epoch": 0.5919557506162448, "flos": 28183811850720.0, "grad_norm": 1.8662105306938606, "language_loss": 0.57420254, "learning_rate": 1.5066050283767574e-06, "loss": 0.59598064, "num_input_tokens_seen": 106179170, "step": 4923, "time_per_iteration": 2.5824756622314453 }, { "auxiliary_loss_clip": 0.01141309, "auxiliary_loss_mlp": 0.01029982, "balance_loss_clip": 1.04793692, "balance_loss_mlp": 1.02313089, "epoch": 0.5920759935068839, "flos": 12094358424480.0, "grad_norm": 1.8957038616265505, "language_loss": 0.82653677, "learning_rate": 1.505850170207616e-06, "loss": 0.84824967, "num_input_tokens_seen": 106196035, "step": 4924, "time_per_iteration": 2.484563112258911 }, { "auxiliary_loss_clip": 0.01148473, "auxiliary_loss_mlp": 0.01031977, "balance_loss_clip": 1.04675305, "balance_loss_mlp": 1.02508116, "epoch": 0.592196236397523, "flos": 29424777034560.0, "grad_norm": 1.9394575710342932, "language_loss": 0.77674943, "learning_rate": 1.505095387000611e-06, "loss": 0.79855394, "num_input_tokens_seen": 106218335, "step": 4925, "time_per_iteration": 2.61273455619812 }, { "auxiliary_loss_clip": 0.01139292, "auxiliary_loss_mlp": 0.0102506, "balance_loss_clip": 1.04656816, "balance_loss_mlp": 1.01768124, "epoch": 0.5923164792881621, "flos": 24384713286720.0, "grad_norm": 1.8642966141024848, "language_loss": 0.74460065, "learning_rate": 1.504340678870242e-06, "loss": 0.76624417, "num_input_tokens_seen": 106236550, "step": 4926, "time_per_iteration": 3.3406505584716797 }, { "auxiliary_loss_clip": 0.01162451, "auxiliary_loss_mlp": 0.01027764, "balance_loss_clip": 1.0495559, "balance_loss_mlp": 1.01972365, "epoch": 0.5924367221788012, "flos": 24024242725440.0, "grad_norm": 1.9980163305248662, "language_loss": 0.89511132, "learning_rate": 1.5035860459309989e-06, "loss": 0.91701353, "num_input_tokens_seen": 106254265, "step": 4927, "time_per_iteration": 2.493417978286743 }, { "auxiliary_loss_clip": 0.01143294, "auxiliary_loss_mlp": 0.01033182, "balance_loss_clip": 1.04543018, "balance_loss_mlp": 1.02603006, "epoch": 0.5925569650694402, "flos": 26870594520000.0, "grad_norm": 1.770714378318074, "language_loss": 0.63474977, "learning_rate": 1.5028314882973568e-06, "loss": 0.65651453, "num_input_tokens_seen": 106274670, "step": 4928, "time_per_iteration": 3.35701322555542 }, { "auxiliary_loss_clip": 0.01149653, "auxiliary_loss_mlp": 0.01028107, "balance_loss_clip": 1.047984, "balance_loss_mlp": 1.02068079, "epoch": 0.5926772079600794, "flos": 22302791664960.0, "grad_norm": 2.1236034043634278, "language_loss": 0.84402752, "learning_rate": 1.502077006083783e-06, "loss": 0.86580509, "num_input_tokens_seen": 106293330, "step": 4929, "time_per_iteration": 2.50453519821167 }, { "auxiliary_loss_clip": 0.01162354, "auxiliary_loss_mlp": 0.0076193, "balance_loss_clip": 1.04904449, "balance_loss_mlp": 1.00027812, "epoch": 0.5927974508507184, "flos": 19865248271520.0, "grad_norm": 1.9198945517725197, "language_loss": 0.76560509, "learning_rate": 1.5013225994047315e-06, "loss": 0.78484786, "num_input_tokens_seen": 106310960, "step": 4930, "time_per_iteration": 3.228936195373535 }, { "auxiliary_loss_clip": 0.01165702, "auxiliary_loss_mlp": 0.00762005, "balance_loss_clip": 1.05134296, "balance_loss_mlp": 1.00028467, "epoch": 0.5929176937413575, "flos": 15776746033440.0, "grad_norm": 1.581650811991155, "language_loss": 0.80626929, "learning_rate": 1.5005682683746452e-06, "loss": 0.82554638, "num_input_tokens_seen": 106329475, "step": 4931, "time_per_iteration": 2.479245662689209 }, { "auxiliary_loss_clip": 0.01165699, "auxiliary_loss_mlp": 0.01026798, "balance_loss_clip": 1.05306935, "balance_loss_mlp": 1.01965392, "epoch": 0.5930379366319967, "flos": 17601475105920.0, "grad_norm": 1.9696294367828888, "language_loss": 0.72631532, "learning_rate": 1.4998140131079553e-06, "loss": 0.74824023, "num_input_tokens_seen": 106345565, "step": 4932, "time_per_iteration": 2.47196888923645 }, { "auxiliary_loss_clip": 0.01105856, "auxiliary_loss_mlp": 0.0076226, "balance_loss_clip": 1.04511344, "balance_loss_mlp": 1.00023937, "epoch": 0.5931581795226357, "flos": 17704286197440.0, "grad_norm": 1.7211027115916548, "language_loss": 0.73115087, "learning_rate": 1.4990598337190821e-06, "loss": 0.74983203, "num_input_tokens_seen": 106361920, "step": 4933, "time_per_iteration": 3.3295528888702393 }, { "auxiliary_loss_clip": 0.01178064, "auxiliary_loss_mlp": 0.00762039, "balance_loss_clip": 1.05184686, "balance_loss_mlp": 1.0002532, "epoch": 0.5932784224132748, "flos": 24280106347200.0, "grad_norm": 1.6443409584374362, "language_loss": 0.67729294, "learning_rate": 1.4983057303224338e-06, "loss": 0.69669402, "num_input_tokens_seen": 106381735, "step": 4934, "time_per_iteration": 2.4748458862304688 }, { "auxiliary_loss_clip": 0.01119775, "auxiliary_loss_mlp": 0.01029103, "balance_loss_clip": 1.04453635, "balance_loss_mlp": 1.02227581, "epoch": 0.5933986653039139, "flos": 22926704326560.0, "grad_norm": 1.567793987003978, "language_loss": 0.87440002, "learning_rate": 1.4975517030324072e-06, "loss": 0.89588886, "num_input_tokens_seen": 106399745, "step": 4935, "time_per_iteration": 2.583130359649658 }, { "auxiliary_loss_clip": 0.01073435, "auxiliary_loss_mlp": 0.0075246, "balance_loss_clip": 1.01330662, "balance_loss_mlp": 0.99975532, "epoch": 0.593518908194553, "flos": 71121741337920.0, "grad_norm": 0.7794829002089525, "language_loss": 0.61821842, "learning_rate": 1.4967977519633882e-06, "loss": 0.63647741, "num_input_tokens_seen": 106457205, "step": 4936, "time_per_iteration": 3.1777827739715576 }, { "auxiliary_loss_clip": 0.0113264, "auxiliary_loss_mlp": 0.01031194, "balance_loss_clip": 1.04609597, "balance_loss_mlp": 1.02393365, "epoch": 0.593639151085192, "flos": 20448652990560.0, "grad_norm": 2.045183416085654, "language_loss": 0.78443497, "learning_rate": 1.4960438772297494e-06, "loss": 0.80607331, "num_input_tokens_seen": 106474250, "step": 4937, "time_per_iteration": 2.540648937225342 }, { "auxiliary_loss_clip": 0.01151912, "auxiliary_loss_mlp": 0.01033879, "balance_loss_clip": 1.04709506, "balance_loss_mlp": 1.0263629, "epoch": 0.5937593939758312, "flos": 30883432500000.0, "grad_norm": 3.528865816788163, "language_loss": 0.73611736, "learning_rate": 1.495290078945855e-06, "loss": 0.75797534, "num_input_tokens_seen": 106494015, "step": 4938, "time_per_iteration": 2.588373899459839 }, { "auxiliary_loss_clip": 0.01175321, "auxiliary_loss_mlp": 0.01030581, "balance_loss_clip": 1.05043805, "balance_loss_mlp": 1.02320504, "epoch": 0.5938796368664703, "flos": 36898074140640.0, "grad_norm": 2.3904507038987814, "language_loss": 0.74399126, "learning_rate": 1.4945363572260529e-06, "loss": 0.76605022, "num_input_tokens_seen": 106515010, "step": 4939, "time_per_iteration": 2.5920333862304688 }, { "auxiliary_loss_clip": 0.01164933, "auxiliary_loss_mlp": 0.0102375, "balance_loss_clip": 1.05053854, "balance_loss_mlp": 1.01670456, "epoch": 0.5939998797571093, "flos": 23842929936000.0, "grad_norm": 2.163264624688953, "language_loss": 0.68055332, "learning_rate": 1.4937827121846845e-06, "loss": 0.70244014, "num_input_tokens_seen": 106535265, "step": 4940, "time_per_iteration": 2.534188747406006 }, { "auxiliary_loss_clip": 0.01131448, "auxiliary_loss_mlp": 0.01029106, "balance_loss_clip": 1.049263, "balance_loss_mlp": 1.02212977, "epoch": 0.5941201226477485, "flos": 25191411333120.0, "grad_norm": 5.111731259774065, "language_loss": 0.73478204, "learning_rate": 1.4930291439360755e-06, "loss": 0.75638759, "num_input_tokens_seen": 106557830, "step": 4941, "time_per_iteration": 2.6014816761016846 }, { "auxiliary_loss_clip": 0.01165282, "auxiliary_loss_mlp": 0.0103073, "balance_loss_clip": 1.05123878, "balance_loss_mlp": 1.02309132, "epoch": 0.5942403655383875, "flos": 22418999782560.0, "grad_norm": 1.7380498617001159, "language_loss": 0.78811991, "learning_rate": 1.4922756525945427e-06, "loss": 0.81008005, "num_input_tokens_seen": 106577140, "step": 4942, "time_per_iteration": 2.4885642528533936 }, { "auxiliary_loss_clip": 0.01062072, "auxiliary_loss_mlp": 0.01002056, "balance_loss_clip": 1.0137229, "balance_loss_mlp": 1.0007391, "epoch": 0.5943606084290266, "flos": 67629322505760.0, "grad_norm": 0.7770747882347183, "language_loss": 0.59601122, "learning_rate": 1.4915222382743894e-06, "loss": 0.61665249, "num_input_tokens_seen": 106635975, "step": 4943, "time_per_iteration": 3.142963409423828 }, { "auxiliary_loss_clip": 0.01166958, "auxiliary_loss_mlp": 0.01025789, "balance_loss_clip": 1.05269277, "balance_loss_mlp": 1.01818955, "epoch": 0.5944808513196658, "flos": 18223160916000.0, "grad_norm": 2.2972209138011617, "language_loss": 0.7216475, "learning_rate": 1.4907689010899085e-06, "loss": 0.74357498, "num_input_tokens_seen": 106653555, "step": 4944, "time_per_iteration": 2.453847885131836 }, { "auxiliary_loss_clip": 0.01149713, "auxiliary_loss_mlp": 0.01027115, "balance_loss_clip": 1.04848099, "balance_loss_mlp": 1.01998019, "epoch": 0.5946010942103048, "flos": 24790827915840.0, "grad_norm": 1.8252250976482127, "language_loss": 0.62019777, "learning_rate": 1.4900156411553804e-06, "loss": 0.64196604, "num_input_tokens_seen": 106673385, "step": 4945, "time_per_iteration": 2.5606887340545654 }, { "auxiliary_loss_clip": 0.01151745, "auxiliary_loss_mlp": 0.01030615, "balance_loss_clip": 1.04904294, "balance_loss_mlp": 1.02339995, "epoch": 0.5947213371009439, "flos": 15231622405440.0, "grad_norm": 2.017004130125578, "language_loss": 0.85453463, "learning_rate": 1.4892624585850739e-06, "loss": 0.87635827, "num_input_tokens_seen": 106691740, "step": 4946, "time_per_iteration": 2.489022970199585 }, { "auxiliary_loss_clip": 0.01182605, "auxiliary_loss_mlp": 0.01030962, "balance_loss_clip": 1.05352557, "balance_loss_mlp": 1.02331483, "epoch": 0.594841579991583, "flos": 25848073873920.0, "grad_norm": 1.896229823844114, "language_loss": 0.79272902, "learning_rate": 1.4885093534932465e-06, "loss": 0.8148647, "num_input_tokens_seen": 106709705, "step": 4947, "time_per_iteration": 2.5139822959899902 }, { "auxiliary_loss_clip": 0.01150239, "auxiliary_loss_mlp": 0.01030052, "balance_loss_clip": 1.05062199, "balance_loss_mlp": 1.02210069, "epoch": 0.5949618228822221, "flos": 23981112844800.0, "grad_norm": 1.9543706517303674, "language_loss": 0.71052468, "learning_rate": 1.4877563259941433e-06, "loss": 0.73232758, "num_input_tokens_seen": 106727560, "step": 4948, "time_per_iteration": 2.516765832901001 }, { "auxiliary_loss_clip": 0.01172839, "auxiliary_loss_mlp": 0.01029634, "balance_loss_clip": 1.05319691, "balance_loss_mlp": 1.02221322, "epoch": 0.5950820657728612, "flos": 40547496368640.0, "grad_norm": 1.8563654479995857, "language_loss": 0.67673445, "learning_rate": 1.4870033762019988e-06, "loss": 0.6987592, "num_input_tokens_seen": 106747725, "step": 4949, "time_per_iteration": 2.650087833404541 }, { "auxiliary_loss_clip": 0.01147008, "auxiliary_loss_mlp": 0.010343, "balance_loss_clip": 1.04741573, "balance_loss_mlp": 1.02702224, "epoch": 0.5952023086635003, "flos": 23184471547200.0, "grad_norm": 1.565049181152439, "language_loss": 0.73572862, "learning_rate": 1.4862505042310334e-06, "loss": 0.75754166, "num_input_tokens_seen": 106767010, "step": 4950, "time_per_iteration": 2.508655309677124 }, { "auxiliary_loss_clip": 0.01141091, "auxiliary_loss_mlp": 0.0102286, "balance_loss_clip": 1.04698968, "balance_loss_mlp": 1.01549554, "epoch": 0.5953225515541394, "flos": 33653293946400.0, "grad_norm": 1.9642988331450353, "language_loss": 0.69432974, "learning_rate": 1.4854977101954587e-06, "loss": 0.7159692, "num_input_tokens_seen": 106789230, "step": 4951, "time_per_iteration": 2.6162750720977783 }, { "auxiliary_loss_clip": 0.0116564, "auxiliary_loss_mlp": 0.01031691, "balance_loss_clip": 1.04782569, "balance_loss_mlp": 1.02377534, "epoch": 0.5954427944447784, "flos": 24459623288640.0, "grad_norm": 1.8855895605063813, "language_loss": 0.86484796, "learning_rate": 1.4847449942094716e-06, "loss": 0.88682127, "num_input_tokens_seen": 106808110, "step": 4952, "time_per_iteration": 3.267587900161743 }, { "auxiliary_loss_clip": 0.01147353, "auxiliary_loss_mlp": 0.01029021, "balance_loss_clip": 1.04997635, "balance_loss_mlp": 1.02162731, "epoch": 0.5955630373354175, "flos": 18551851356000.0, "grad_norm": 1.8198465021855594, "language_loss": 0.86127543, "learning_rate": 1.4839923563872598e-06, "loss": 0.88303918, "num_input_tokens_seen": 106826650, "step": 4953, "time_per_iteration": 2.5170626640319824 }, { "auxiliary_loss_clip": 0.011346, "auxiliary_loss_mlp": 0.01034931, "balance_loss_clip": 1.04767895, "balance_loss_mlp": 1.02691472, "epoch": 0.5956832802260567, "flos": 19791703114080.0, "grad_norm": 2.1317375205354283, "language_loss": 0.75707352, "learning_rate": 1.483239796842997e-06, "loss": 0.77876878, "num_input_tokens_seen": 106844680, "step": 4954, "time_per_iteration": 3.325248956680298 }, { "auxiliary_loss_clip": 0.01132895, "auxiliary_loss_mlp": 0.0102955, "balance_loss_clip": 1.04962873, "balance_loss_mlp": 1.02226591, "epoch": 0.5958035231166957, "flos": 19750871918880.0, "grad_norm": 1.733346710388223, "language_loss": 0.8386476, "learning_rate": 1.4824873156908462e-06, "loss": 0.86027205, "num_input_tokens_seen": 106862605, "step": 4955, "time_per_iteration": 2.541189193725586 }, { "auxiliary_loss_clip": 0.01166283, "auxiliary_loss_mlp": 0.00762628, "balance_loss_clip": 1.05265093, "balance_loss_mlp": 1.00022876, "epoch": 0.5959237660073348, "flos": 21652809678720.0, "grad_norm": 1.4921403133259208, "language_loss": 0.75570774, "learning_rate": 1.4817349130449584e-06, "loss": 0.77499688, "num_input_tokens_seen": 106882325, "step": 4956, "time_per_iteration": 3.2400052547454834 }, { "auxiliary_loss_clip": 0.01161687, "auxiliary_loss_mlp": 0.01026918, "balance_loss_clip": 1.05078006, "balance_loss_mlp": 1.01975381, "epoch": 0.5960440088979739, "flos": 21171210376320.0, "grad_norm": 1.7858770212483115, "language_loss": 0.82870024, "learning_rate": 1.4809825890194717e-06, "loss": 0.8505863, "num_input_tokens_seen": 106900995, "step": 4957, "time_per_iteration": 2.508543014526367 }, { "auxiliary_loss_clip": 0.01143885, "auxiliary_loss_mlp": 0.01024473, "balance_loss_clip": 1.04647505, "balance_loss_mlp": 1.01778793, "epoch": 0.596164251788613, "flos": 14757529747680.0, "grad_norm": 1.799061786111808, "language_loss": 0.77091324, "learning_rate": 1.4802303437285139e-06, "loss": 0.79259682, "num_input_tokens_seen": 106918265, "step": 4958, "time_per_iteration": 2.497875690460205 }, { "auxiliary_loss_clip": 0.01148024, "auxiliary_loss_mlp": 0.01023279, "balance_loss_clip": 1.04672921, "balance_loss_mlp": 1.01580453, "epoch": 0.596284494679252, "flos": 20485928406720.0, "grad_norm": 2.1414386813985766, "language_loss": 0.80859923, "learning_rate": 1.4794781772861994e-06, "loss": 0.83031225, "num_input_tokens_seen": 106934760, "step": 4959, "time_per_iteration": 3.2915947437286377 }, { "auxiliary_loss_clip": 0.01149004, "auxiliary_loss_mlp": 0.00762248, "balance_loss_clip": 1.04871249, "balance_loss_mlp": 1.00021911, "epoch": 0.5964047375698912, "flos": 31212266607840.0, "grad_norm": 1.993539252957994, "language_loss": 0.66936356, "learning_rate": 1.4787260898066324e-06, "loss": 0.68847603, "num_input_tokens_seen": 106954760, "step": 4960, "time_per_iteration": 2.6068601608276367 }, { "auxiliary_loss_clip": 0.01174893, "auxiliary_loss_mlp": 0.01033062, "balance_loss_clip": 1.05102587, "balance_loss_mlp": 1.025599, "epoch": 0.5965249804605303, "flos": 27483624342720.0, "grad_norm": 1.9477771616279518, "language_loss": 0.85937965, "learning_rate": 1.4779740814039023e-06, "loss": 0.88145924, "num_input_tokens_seen": 106974845, "step": 4961, "time_per_iteration": 2.4956700801849365 }, { "auxiliary_loss_clip": 0.01177515, "auxiliary_loss_mlp": 0.0102742, "balance_loss_clip": 1.05138814, "balance_loss_mlp": 1.01969481, "epoch": 0.5966452233511693, "flos": 30773940853920.0, "grad_norm": 1.8805282930550749, "language_loss": 0.68467343, "learning_rate": 1.4772221521920894e-06, "loss": 0.70672274, "num_input_tokens_seen": 106994870, "step": 4962, "time_per_iteration": 2.522557020187378 }, { "auxiliary_loss_clip": 0.01147871, "auxiliary_loss_mlp": 0.0102629, "balance_loss_clip": 1.0490365, "balance_loss_mlp": 1.01892924, "epoch": 0.5967654662418085, "flos": 25481174176800.0, "grad_norm": 2.5869643275031935, "language_loss": 0.7408573, "learning_rate": 1.4764703022852598e-06, "loss": 0.76259887, "num_input_tokens_seen": 107015390, "step": 4963, "time_per_iteration": 2.5645081996917725 }, { "auxiliary_loss_clip": 0.01093305, "auxiliary_loss_mlp": 0.01024073, "balance_loss_clip": 1.04131162, "balance_loss_mlp": 1.0166043, "epoch": 0.5968857091324475, "flos": 19099130001600.0, "grad_norm": 1.918615937969897, "language_loss": 0.76874828, "learning_rate": 1.4757185317974696e-06, "loss": 0.78992212, "num_input_tokens_seen": 107033775, "step": 4964, "time_per_iteration": 2.5690715312957764 }, { "auxiliary_loss_clip": 0.01164261, "auxiliary_loss_mlp": 0.01024051, "balance_loss_clip": 1.05061364, "balance_loss_mlp": 1.01661873, "epoch": 0.5970059520230866, "flos": 23692714845600.0, "grad_norm": 2.8665600317722837, "language_loss": 0.71154475, "learning_rate": 1.474966840842761e-06, "loss": 0.73342788, "num_input_tokens_seen": 107053355, "step": 4965, "time_per_iteration": 2.5058627128601074 }, { "auxiliary_loss_clip": 0.01166425, "auxiliary_loss_mlp": 0.0102849, "balance_loss_clip": 1.05023122, "balance_loss_mlp": 1.02127838, "epoch": 0.5971261949137258, "flos": 23185549056000.0, "grad_norm": 1.750391220660223, "language_loss": 0.87139845, "learning_rate": 1.4742152295351655e-06, "loss": 0.89334756, "num_input_tokens_seen": 107072510, "step": 4966, "time_per_iteration": 2.4805707931518555 }, { "auxiliary_loss_clip": 0.01166374, "auxiliary_loss_mlp": 0.00762345, "balance_loss_clip": 1.05223513, "balance_loss_mlp": 1.00017142, "epoch": 0.5972464378043648, "flos": 20557713633120.0, "grad_norm": 5.086959623532529, "language_loss": 0.64084172, "learning_rate": 1.4734636979887016e-06, "loss": 0.66012889, "num_input_tokens_seen": 107089970, "step": 4967, "time_per_iteration": 2.470893144607544 }, { "auxiliary_loss_clip": 0.01140071, "auxiliary_loss_mlp": 0.01023903, "balance_loss_clip": 1.04724765, "balance_loss_mlp": 1.01644599, "epoch": 0.5973666806950039, "flos": 29387034697920.0, "grad_norm": 2.4189424226559813, "language_loss": 0.90370709, "learning_rate": 1.4727122463173755e-06, "loss": 0.92534679, "num_input_tokens_seen": 107108500, "step": 4968, "time_per_iteration": 2.5967538356781006 }, { "auxiliary_loss_clip": 0.01150289, "auxiliary_loss_mlp": 0.01030727, "balance_loss_clip": 1.05055976, "balance_loss_mlp": 1.02346158, "epoch": 0.597486923585643, "flos": 22273525730880.0, "grad_norm": 2.1594589114897618, "language_loss": 0.64338964, "learning_rate": 1.471960874635183e-06, "loss": 0.66519976, "num_input_tokens_seen": 107128060, "step": 4969, "time_per_iteration": 2.5384116172790527 }, { "auxiliary_loss_clip": 0.01144136, "auxiliary_loss_mlp": 0.01031026, "balance_loss_clip": 1.04736042, "balance_loss_mlp": 1.02347445, "epoch": 0.5976071664762821, "flos": 13772464102560.0, "grad_norm": 4.1196807986807, "language_loss": 0.71123451, "learning_rate": 1.4712095830561055e-06, "loss": 0.73298615, "num_input_tokens_seen": 107146550, "step": 4970, "time_per_iteration": 2.4939610958099365 }, { "auxiliary_loss_clip": 0.01147671, "auxiliary_loss_mlp": 0.01021176, "balance_loss_clip": 1.04618049, "balance_loss_mlp": 1.01448202, "epoch": 0.5977274093669211, "flos": 19098627164160.0, "grad_norm": 1.8092171052713528, "language_loss": 0.81265914, "learning_rate": 1.4704583716941147e-06, "loss": 0.83434761, "num_input_tokens_seen": 107165415, "step": 4971, "time_per_iteration": 2.5169625282287598 }, { "auxiliary_loss_clip": 0.01155432, "auxiliary_loss_mlp": 0.01026119, "balance_loss_clip": 1.05000067, "balance_loss_mlp": 1.01848662, "epoch": 0.5978476522575603, "flos": 20376005757120.0, "grad_norm": 1.8723080849053093, "language_loss": 0.72299469, "learning_rate": 1.4697072406631672e-06, "loss": 0.74481016, "num_input_tokens_seen": 107185320, "step": 4972, "time_per_iteration": 2.4907684326171875 }, { "auxiliary_loss_clip": 0.01124625, "auxiliary_loss_mlp": 0.01028274, "balance_loss_clip": 1.04746234, "balance_loss_mlp": 1.02078152, "epoch": 0.5979678951481994, "flos": 29023152025440.0, "grad_norm": 1.6773648588831562, "language_loss": 0.73162842, "learning_rate": 1.4689561900772097e-06, "loss": 0.75315738, "num_input_tokens_seen": 107205380, "step": 4973, "time_per_iteration": 2.6430070400238037 }, { "auxiliary_loss_clip": 0.01146106, "auxiliary_loss_mlp": 0.01024378, "balance_loss_clip": 1.04690552, "balance_loss_mlp": 1.01762223, "epoch": 0.5980881380388384, "flos": 17967692380800.0, "grad_norm": 2.3566408316350556, "language_loss": 0.72304749, "learning_rate": 1.4682052200501758e-06, "loss": 0.74475235, "num_input_tokens_seen": 107222585, "step": 4974, "time_per_iteration": 2.4800970554351807 }, { "auxiliary_loss_clip": 0.01175071, "auxiliary_loss_mlp": 0.01029488, "balance_loss_clip": 1.05027616, "balance_loss_mlp": 1.02273798, "epoch": 0.5982083809294776, "flos": 22962830400000.0, "grad_norm": 1.8662591826227484, "language_loss": 0.80130947, "learning_rate": 1.4674543306959876e-06, "loss": 0.82335508, "num_input_tokens_seen": 107242055, "step": 4975, "time_per_iteration": 2.4722118377685547 }, { "auxiliary_loss_clip": 0.01154519, "auxiliary_loss_mlp": 0.0102902, "balance_loss_clip": 1.04968977, "balance_loss_mlp": 1.02112269, "epoch": 0.5983286238201166, "flos": 20991944853600.0, "grad_norm": 2.239624590035263, "language_loss": 0.84017229, "learning_rate": 1.4667035221285535e-06, "loss": 0.86200768, "num_input_tokens_seen": 107259695, "step": 4976, "time_per_iteration": 2.508525848388672 }, { "auxiliary_loss_clip": 0.01161668, "auxiliary_loss_mlp": 0.01026637, "balance_loss_clip": 1.0514698, "balance_loss_mlp": 1.0194217, "epoch": 0.5984488667107557, "flos": 28183452681120.0, "grad_norm": 1.7828006471745135, "language_loss": 0.74037588, "learning_rate": 1.4659527944617715e-06, "loss": 0.76225889, "num_input_tokens_seen": 107279640, "step": 4977, "time_per_iteration": 2.5331525802612305 }, { "auxiliary_loss_clip": 0.01102195, "auxiliary_loss_mlp": 0.01028178, "balance_loss_clip": 1.04341769, "balance_loss_mlp": 1.02112675, "epoch": 0.5985691096013949, "flos": 16471797416160.0, "grad_norm": 1.7950546937309106, "language_loss": 0.76166618, "learning_rate": 1.465202147809526e-06, "loss": 0.78296989, "num_input_tokens_seen": 107298135, "step": 4978, "time_per_iteration": 3.312619686126709 }, { "auxiliary_loss_clip": 0.0117858, "auxiliary_loss_mlp": 0.01029296, "balance_loss_clip": 1.05310678, "balance_loss_mlp": 1.02225709, "epoch": 0.5986893524920339, "flos": 26719050502080.0, "grad_norm": 1.8698449426835682, "language_loss": 0.76341969, "learning_rate": 1.4644515822856888e-06, "loss": 0.7854985, "num_input_tokens_seen": 107316570, "step": 4979, "time_per_iteration": 2.5167396068573 }, { "auxiliary_loss_clip": 0.01042526, "auxiliary_loss_mlp": 0.01001975, "balance_loss_clip": 1.0130403, "balance_loss_mlp": 1.00054443, "epoch": 0.598809595382673, "flos": 61608072144480.0, "grad_norm": 0.7546468880408337, "language_loss": 0.56544799, "learning_rate": 1.4637010980041215e-06, "loss": 0.58589303, "num_input_tokens_seen": 107378680, "step": 4980, "time_per_iteration": 3.9115004539489746 }, { "auxiliary_loss_clip": 0.01179347, "auxiliary_loss_mlp": 0.01027353, "balance_loss_clip": 1.0528549, "balance_loss_mlp": 1.01980078, "epoch": 0.5989298382733121, "flos": 11801722224000.0, "grad_norm": 2.199875600762352, "language_loss": 0.89505726, "learning_rate": 1.4629506950786707e-06, "loss": 0.91712427, "num_input_tokens_seen": 107394860, "step": 4981, "time_per_iteration": 2.4508845806121826 }, { "auxiliary_loss_clip": 0.01073042, "auxiliary_loss_mlp": 0.01001443, "balance_loss_clip": 1.0128144, "balance_loss_mlp": 1.00011349, "epoch": 0.5990500811639512, "flos": 60025809667680.0, "grad_norm": 0.8081441893957247, "language_loss": 0.56085777, "learning_rate": 1.4622003736231733e-06, "loss": 0.58160257, "num_input_tokens_seen": 107453850, "step": 4982, "time_per_iteration": 3.802550792694092 }, { "auxiliary_loss_clip": 0.01162622, "auxiliary_loss_mlp": 0.01022336, "balance_loss_clip": 1.05123067, "balance_loss_mlp": 1.01498389, "epoch": 0.5991703240545903, "flos": 18222729912480.0, "grad_norm": 1.9609603393028197, "language_loss": 0.81079936, "learning_rate": 1.461450133751451e-06, "loss": 0.83264899, "num_input_tokens_seen": 107471920, "step": 4983, "time_per_iteration": 2.463637590408325 }, { "auxiliary_loss_clip": 0.0116509, "auxiliary_loss_mlp": 0.01025197, "balance_loss_clip": 1.05252385, "balance_loss_mlp": 1.01831555, "epoch": 0.5992905669452293, "flos": 27709898777760.0, "grad_norm": 1.7505385568210643, "language_loss": 0.75869465, "learning_rate": 1.4606999755773153e-06, "loss": 0.78059751, "num_input_tokens_seen": 107493125, "step": 4984, "time_per_iteration": 2.545457363128662 }, { "auxiliary_loss_clip": 0.01178182, "auxiliary_loss_mlp": 0.01027089, "balance_loss_clip": 1.05295134, "balance_loss_mlp": 1.01983488, "epoch": 0.5994108098358685, "flos": 20449012160160.0, "grad_norm": 1.5766525565046754, "language_loss": 0.82310259, "learning_rate": 1.4599498992145643e-06, "loss": 0.84515536, "num_input_tokens_seen": 107513150, "step": 4985, "time_per_iteration": 3.1662192344665527 }, { "auxiliary_loss_clip": 0.01158243, "auxiliary_loss_mlp": 0.00761887, "balance_loss_clip": 1.05190945, "balance_loss_mlp": 1.00021422, "epoch": 0.5995310527265075, "flos": 22269969951840.0, "grad_norm": 1.873864071968721, "language_loss": 0.7053743, "learning_rate": 1.4591999047769846e-06, "loss": 0.72457558, "num_input_tokens_seen": 107532005, "step": 4986, "time_per_iteration": 2.5418097972869873 }, { "auxiliary_loss_clip": 0.01106013, "auxiliary_loss_mlp": 0.010292, "balance_loss_clip": 1.04361343, "balance_loss_mlp": 1.02197933, "epoch": 0.5996512956171466, "flos": 18916955205120.0, "grad_norm": 1.6581766586789486, "language_loss": 0.75164056, "learning_rate": 1.4584499923783486e-06, "loss": 0.77299273, "num_input_tokens_seen": 107550585, "step": 4987, "time_per_iteration": 2.610240936279297 }, { "auxiliary_loss_clip": 0.01149465, "auxiliary_loss_mlp": 0.0102111, "balance_loss_clip": 1.04892755, "balance_loss_mlp": 1.01437759, "epoch": 0.5997715385077858, "flos": 15370918740000.0, "grad_norm": 2.0049078475795903, "language_loss": 0.76576507, "learning_rate": 1.457700162132419e-06, "loss": 0.78747082, "num_input_tokens_seen": 107567575, "step": 4988, "time_per_iteration": 2.5061795711517334 }, { "auxiliary_loss_clip": 0.01118973, "auxiliary_loss_mlp": 0.01023671, "balance_loss_clip": 1.04788136, "balance_loss_mlp": 1.01648021, "epoch": 0.5998917813984248, "flos": 25264848739680.0, "grad_norm": 2.372434381292361, "language_loss": 0.72424018, "learning_rate": 1.4569504141529433e-06, "loss": 0.74566662, "num_input_tokens_seen": 107585410, "step": 4989, "time_per_iteration": 2.5873591899871826 }, { "auxiliary_loss_clip": 0.01160524, "auxiliary_loss_mlp": 0.01028007, "balance_loss_clip": 1.05020046, "balance_loss_mlp": 1.02036881, "epoch": 0.6000120242890639, "flos": 22054506521760.0, "grad_norm": 2.918089085752445, "language_loss": 0.72025692, "learning_rate": 1.456200748553658e-06, "loss": 0.7421422, "num_input_tokens_seen": 107603405, "step": 4990, "time_per_iteration": 2.5050179958343506 }, { "auxiliary_loss_clip": 0.01181975, "auxiliary_loss_mlp": 0.01034491, "balance_loss_clip": 1.05507612, "balance_loss_mlp": 1.02714491, "epoch": 0.600132267179703, "flos": 29863426041120.0, "grad_norm": 1.5095042020233753, "language_loss": 0.7855981, "learning_rate": 1.455451165448287e-06, "loss": 0.80776274, "num_input_tokens_seen": 107626060, "step": 4991, "time_per_iteration": 2.5232977867126465 }, { "auxiliary_loss_clip": 0.01150742, "auxiliary_loss_mlp": 0.01032212, "balance_loss_clip": 1.05211735, "balance_loss_mlp": 1.02459502, "epoch": 0.6002525100703421, "flos": 25045362610080.0, "grad_norm": 2.4910376430693644, "language_loss": 0.73307514, "learning_rate": 1.4547016649505407e-06, "loss": 0.75490469, "num_input_tokens_seen": 107644070, "step": 4992, "time_per_iteration": 2.5434482097625732 }, { "auxiliary_loss_clip": 0.01134665, "auxiliary_loss_mlp": 0.01029097, "balance_loss_clip": 1.04688048, "balance_loss_mlp": 1.02102661, "epoch": 0.6003727529609811, "flos": 20849595577440.0, "grad_norm": 1.9714594555791023, "language_loss": 0.8493976, "learning_rate": 1.4539522471741193e-06, "loss": 0.87103522, "num_input_tokens_seen": 107661495, "step": 4993, "time_per_iteration": 2.640915870666504 }, { "auxiliary_loss_clip": 0.01169115, "auxiliary_loss_mlp": 0.01034625, "balance_loss_clip": 1.05151784, "balance_loss_mlp": 1.02595806, "epoch": 0.6004929958516203, "flos": 15594607153920.0, "grad_norm": 2.018483051930779, "language_loss": 0.71028119, "learning_rate": 1.4532029122327067e-06, "loss": 0.73231864, "num_input_tokens_seen": 107678280, "step": 4994, "time_per_iteration": 2.4773612022399902 }, { "auxiliary_loss_clip": 0.01131077, "auxiliary_loss_mlp": 0.01026266, "balance_loss_clip": 1.0508976, "balance_loss_mlp": 1.01895308, "epoch": 0.6006132387422594, "flos": 21763271082720.0, "grad_norm": 2.5218182884754303, "language_loss": 0.75590253, "learning_rate": 1.4524536602399783e-06, "loss": 0.77747601, "num_input_tokens_seen": 107697370, "step": 4995, "time_per_iteration": 2.5996923446655273 }, { "auxiliary_loss_clip": 0.01148151, "auxiliary_loss_mlp": 0.01028686, "balance_loss_clip": 1.05309343, "balance_loss_mlp": 1.0213902, "epoch": 0.6007334816328984, "flos": 22858546713120.0, "grad_norm": 1.4627713115104122, "language_loss": 0.77296788, "learning_rate": 1.4517044913095938e-06, "loss": 0.79473621, "num_input_tokens_seen": 107717790, "step": 4996, "time_per_iteration": 2.5686042308807373 }, { "auxiliary_loss_clip": 0.01165128, "auxiliary_loss_mlp": 0.01025824, "balance_loss_clip": 1.05268753, "balance_loss_mlp": 1.01835537, "epoch": 0.6008537245235376, "flos": 28324580780640.0, "grad_norm": 1.7097261815867597, "language_loss": 0.8179059, "learning_rate": 1.4509554055552022e-06, "loss": 0.83981544, "num_input_tokens_seen": 107738020, "step": 4997, "time_per_iteration": 2.5835039615631104 }, { "auxiliary_loss_clip": 0.01147889, "auxiliary_loss_mlp": 0.01033922, "balance_loss_clip": 1.04910398, "balance_loss_mlp": 1.02658415, "epoch": 0.6009739674141766, "flos": 20886116737440.0, "grad_norm": 2.636661013473203, "language_loss": 0.83858252, "learning_rate": 1.450206403090439e-06, "loss": 0.86040062, "num_input_tokens_seen": 107756215, "step": 4998, "time_per_iteration": 2.527914047241211 }, { "auxiliary_loss_clip": 0.01161483, "auxiliary_loss_mlp": 0.01028781, "balance_loss_clip": 1.05215561, "balance_loss_mlp": 1.02214694, "epoch": 0.6010942103048157, "flos": 20481007783200.0, "grad_norm": 1.9724206337016292, "language_loss": 0.86133546, "learning_rate": 1.4494574840289274e-06, "loss": 0.88323808, "num_input_tokens_seen": 107773330, "step": 4999, "time_per_iteration": 2.5221452713012695 }, { "auxiliary_loss_clip": 0.01169896, "auxiliary_loss_mlp": 0.01027166, "balance_loss_clip": 1.05151796, "balance_loss_mlp": 1.01951909, "epoch": 0.6012144531954549, "flos": 23805977772480.0, "grad_norm": 1.7755740076168618, "language_loss": 0.73951828, "learning_rate": 1.4487086484842782e-06, "loss": 0.76148891, "num_input_tokens_seen": 107791975, "step": 5000, "time_per_iteration": 2.5390117168426514 }, { "auxiliary_loss_clip": 0.01177293, "auxiliary_loss_mlp": 0.01024296, "balance_loss_clip": 1.05207491, "balance_loss_mlp": 1.01769435, "epoch": 0.6013346960860939, "flos": 18988381261920.0, "grad_norm": 2.160082821496231, "language_loss": 0.60236251, "learning_rate": 1.4479598965700878e-06, "loss": 0.62437838, "num_input_tokens_seen": 107809240, "step": 5001, "time_per_iteration": 2.448528528213501 }, { "auxiliary_loss_clip": 0.01134307, "auxiliary_loss_mlp": 0.01027831, "balance_loss_clip": 1.04670942, "balance_loss_mlp": 1.02068114, "epoch": 0.601454938976733, "flos": 24025320234240.0, "grad_norm": 2.2866832730001234, "language_loss": 0.69016016, "learning_rate": 1.4472112283999427e-06, "loss": 0.7117815, "num_input_tokens_seen": 107827895, "step": 5002, "time_per_iteration": 2.593841791152954 }, { "auxiliary_loss_clip": 0.01160063, "auxiliary_loss_mlp": 0.01028999, "balance_loss_clip": 1.05178428, "balance_loss_mlp": 1.02188277, "epoch": 0.6015751818673721, "flos": 26427132640800.0, "grad_norm": 2.4134201068121444, "language_loss": 0.69431573, "learning_rate": 1.4464626440874143e-06, "loss": 0.71620637, "num_input_tokens_seen": 107847010, "step": 5003, "time_per_iteration": 3.3282055854797363 }, { "auxiliary_loss_clip": 0.01125819, "auxiliary_loss_mlp": 0.01029401, "balance_loss_clip": 1.04352224, "balance_loss_mlp": 1.02174735, "epoch": 0.6016954247580112, "flos": 13115262807360.0, "grad_norm": 2.591488749871486, "language_loss": 0.73777449, "learning_rate": 1.4457141437460636e-06, "loss": 0.7593267, "num_input_tokens_seen": 107864235, "step": 5004, "time_per_iteration": 2.5614585876464844 }, { "auxiliary_loss_clip": 0.0115156, "auxiliary_loss_mlp": 0.01031869, "balance_loss_clip": 1.05083585, "balance_loss_mlp": 1.02382791, "epoch": 0.6018156676486502, "flos": 23768450937600.0, "grad_norm": 1.8608785912153374, "language_loss": 0.73616827, "learning_rate": 1.444965727489436e-06, "loss": 0.75800252, "num_input_tokens_seen": 107883680, "step": 5005, "time_per_iteration": 3.3444159030914307 }, { "auxiliary_loss_clip": 0.01132675, "auxiliary_loss_mlp": 0.01025397, "balance_loss_clip": 1.04548228, "balance_loss_mlp": 1.01812184, "epoch": 0.6019359105392894, "flos": 26469364597440.0, "grad_norm": 1.85364471436192, "language_loss": 0.62983537, "learning_rate": 1.444217395431066e-06, "loss": 0.65141606, "num_input_tokens_seen": 107906220, "step": 5006, "time_per_iteration": 2.6142802238464355 }, { "auxiliary_loss_clip": 0.01041692, "auxiliary_loss_mlp": 0.01004977, "balance_loss_clip": 1.01539767, "balance_loss_mlp": 1.00352252, "epoch": 0.6020561534299285, "flos": 69190860896640.0, "grad_norm": 0.7940418626649266, "language_loss": 0.55821395, "learning_rate": 1.4434691476844755e-06, "loss": 0.57868063, "num_input_tokens_seen": 107967195, "step": 5007, "time_per_iteration": 3.836284637451172 }, { "auxiliary_loss_clip": 0.0114548, "auxiliary_loss_mlp": 0.01027353, "balance_loss_clip": 1.04948163, "balance_loss_mlp": 1.02087116, "epoch": 0.6021763963205675, "flos": 21835307727840.0, "grad_norm": 2.3192153173308343, "language_loss": 0.66925412, "learning_rate": 1.4427209843631729e-06, "loss": 0.69098246, "num_input_tokens_seen": 107984245, "step": 5008, "time_per_iteration": 2.5488369464874268 }, { "auxiliary_loss_clip": 0.01174824, "auxiliary_loss_mlp": 0.00762187, "balance_loss_clip": 1.05235863, "balance_loss_mlp": 1.00026369, "epoch": 0.6022966392112067, "flos": 26578640741760.0, "grad_norm": 3.7011045531612585, "language_loss": 0.81137049, "learning_rate": 1.4419729055806534e-06, "loss": 0.83074063, "num_input_tokens_seen": 108003680, "step": 5009, "time_per_iteration": 2.5197489261627197 }, { "auxiliary_loss_clip": 0.01146842, "auxiliary_loss_mlp": 0.0076162, "balance_loss_clip": 1.05282664, "balance_loss_mlp": 1.00024652, "epoch": 0.6024168821018457, "flos": 20703726439200.0, "grad_norm": 1.7654941593480296, "language_loss": 0.82277715, "learning_rate": 1.441224911450401e-06, "loss": 0.84186178, "num_input_tokens_seen": 108019635, "step": 5010, "time_per_iteration": 2.5601046085357666 }, { "auxiliary_loss_clip": 0.01169532, "auxiliary_loss_mlp": 0.01026771, "balance_loss_clip": 1.05354726, "balance_loss_mlp": 1.01948452, "epoch": 0.6025371249924848, "flos": 24680977100160.0, "grad_norm": 1.7266300678214146, "language_loss": 0.82208991, "learning_rate": 1.4404770020858851e-06, "loss": 0.84405297, "num_input_tokens_seen": 108039120, "step": 5011, "time_per_iteration": 3.296905755996704 }, { "auxiliary_loss_clip": 0.01156773, "auxiliary_loss_mlp": 0.01028459, "balance_loss_clip": 1.04992878, "balance_loss_mlp": 1.02134788, "epoch": 0.602657367883124, "flos": 25955805588960.0, "grad_norm": 1.8829012888348353, "language_loss": 0.85980964, "learning_rate": 1.439729177600563e-06, "loss": 0.88166195, "num_input_tokens_seen": 108059615, "step": 5012, "time_per_iteration": 2.556039571762085 }, { "auxiliary_loss_clip": 0.01163596, "auxiliary_loss_mlp": 0.01027618, "balance_loss_clip": 1.05328727, "balance_loss_mlp": 1.02058506, "epoch": 0.602777610773763, "flos": 16690637040480.0, "grad_norm": 1.7901325417620133, "language_loss": 0.73143709, "learning_rate": 1.4389814381078793e-06, "loss": 0.75334924, "num_input_tokens_seen": 108078855, "step": 5013, "time_per_iteration": 2.5052242279052734 }, { "auxiliary_loss_clip": 0.01067118, "auxiliary_loss_mlp": 0.01030963, "balance_loss_clip": 1.042346, "balance_loss_mlp": 1.02414465, "epoch": 0.6028978536644021, "flos": 13334246099520.0, "grad_norm": 1.950862268850263, "language_loss": 0.80401146, "learning_rate": 1.438233783721265e-06, "loss": 0.8249923, "num_input_tokens_seen": 108095020, "step": 5014, "time_per_iteration": 2.8825817108154297 }, { "auxiliary_loss_clip": 0.01145413, "auxiliary_loss_mlp": 0.01031531, "balance_loss_clip": 1.05083454, "balance_loss_mlp": 1.02488506, "epoch": 0.6030180965550412, "flos": 19644828300960.0, "grad_norm": 1.9381832119270914, "language_loss": 0.77660596, "learning_rate": 1.43748621455414e-06, "loss": 0.79837537, "num_input_tokens_seen": 108111455, "step": 5015, "time_per_iteration": 2.759855031967163 }, { "auxiliary_loss_clip": 0.01145971, "auxiliary_loss_mlp": 0.01027403, "balance_loss_clip": 1.04978943, "balance_loss_mlp": 1.02050066, "epoch": 0.6031383394456803, "flos": 14458392577440.0, "grad_norm": 2.2738552547920716, "language_loss": 0.80385268, "learning_rate": 1.4367387307199082e-06, "loss": 0.82558644, "num_input_tokens_seen": 108128305, "step": 5016, "time_per_iteration": 2.5040318965911865 }, { "auxiliary_loss_clip": 0.01157567, "auxiliary_loss_mlp": 0.0103242, "balance_loss_clip": 1.04852259, "balance_loss_mlp": 1.02464187, "epoch": 0.6032585823363193, "flos": 13917794486400.0, "grad_norm": 1.7858770100584145, "language_loss": 0.82391441, "learning_rate": 1.4359913323319632e-06, "loss": 0.84581429, "num_input_tokens_seen": 108145475, "step": 5017, "time_per_iteration": 2.488391399383545 }, { "auxiliary_loss_clip": 0.01095838, "auxiliary_loss_mlp": 0.01032687, "balance_loss_clip": 1.0414865, "balance_loss_mlp": 1.02505207, "epoch": 0.6033788252269584, "flos": 24353256418080.0, "grad_norm": 1.6955911357617819, "language_loss": 0.77631569, "learning_rate": 1.4352440195036847e-06, "loss": 0.79760087, "num_input_tokens_seen": 108165650, "step": 5018, "time_per_iteration": 2.669332265853882 }, { "auxiliary_loss_clip": 0.01094453, "auxiliary_loss_mlp": 0.0102771, "balance_loss_clip": 1.04217589, "balance_loss_mlp": 1.02046788, "epoch": 0.6034990681175976, "flos": 25521251115840.0, "grad_norm": 1.5484318857132353, "language_loss": 0.80027449, "learning_rate": 1.4344967923484395e-06, "loss": 0.82149613, "num_input_tokens_seen": 108187620, "step": 5019, "time_per_iteration": 2.711308240890503 }, { "auxiliary_loss_clip": 0.01160624, "auxiliary_loss_mlp": 0.01027188, "balance_loss_clip": 1.05103314, "balance_loss_mlp": 1.02016914, "epoch": 0.6036193110082366, "flos": 25958391610080.0, "grad_norm": 2.165116105533216, "language_loss": 0.72352231, "learning_rate": 1.433749650979581e-06, "loss": 0.74540043, "num_input_tokens_seen": 108207605, "step": 5020, "time_per_iteration": 2.549691915512085 }, { "auxiliary_loss_clip": 0.01135586, "auxiliary_loss_mlp": 0.01025828, "balance_loss_clip": 1.04680395, "balance_loss_mlp": 1.01807332, "epoch": 0.6037395538988757, "flos": 25593431428800.0, "grad_norm": 1.911853568979863, "language_loss": 0.68304181, "learning_rate": 1.433002595510451e-06, "loss": 0.70465595, "num_input_tokens_seen": 108226385, "step": 5021, "time_per_iteration": 2.6099960803985596 }, { "auxiliary_loss_clip": 0.01145564, "auxiliary_loss_mlp": 0.00762458, "balance_loss_clip": 1.04839802, "balance_loss_mlp": 1.00030625, "epoch": 0.6038597967895148, "flos": 17816256113760.0, "grad_norm": 1.847351096019737, "language_loss": 0.716919, "learning_rate": 1.4322556260543757e-06, "loss": 0.73599923, "num_input_tokens_seen": 108242960, "step": 5022, "time_per_iteration": 2.541027069091797 }, { "auxiliary_loss_clip": 0.01039128, "auxiliary_loss_mlp": 0.01002141, "balance_loss_clip": 1.01032257, "balance_loss_mlp": 1.00068092, "epoch": 0.6039800396801539, "flos": 65169223261440.0, "grad_norm": 0.8986548910671707, "language_loss": 0.62826049, "learning_rate": 1.4315087427246703e-06, "loss": 0.64867318, "num_input_tokens_seen": 108296785, "step": 5023, "time_per_iteration": 3.0749309062957764 }, { "auxiliary_loss_clip": 0.01070575, "auxiliary_loss_mlp": 0.01001754, "balance_loss_clip": 1.01090336, "balance_loss_mlp": 1.00036573, "epoch": 0.604100282570793, "flos": 67386418518240.0, "grad_norm": 0.8832232497534387, "language_loss": 0.58507979, "learning_rate": 1.4307619456346372e-06, "loss": 0.60580313, "num_input_tokens_seen": 108341090, "step": 5024, "time_per_iteration": 2.7849011421203613 }, { "auxiliary_loss_clip": 0.01162681, "auxiliary_loss_mlp": 0.01024326, "balance_loss_clip": 1.04746485, "balance_loss_mlp": 1.01702189, "epoch": 0.6042205254614321, "flos": 35297500402560.0, "grad_norm": 2.002416820731233, "language_loss": 0.74132061, "learning_rate": 1.430015234897564e-06, "loss": 0.76319075, "num_input_tokens_seen": 108364370, "step": 5025, "time_per_iteration": 2.6576759815216064 }, { "auxiliary_loss_clip": 0.0117625, "auxiliary_loss_mlp": 0.00762057, "balance_loss_clip": 1.05144167, "balance_loss_mlp": 1.00024199, "epoch": 0.6043407683520712, "flos": 45658267834080.0, "grad_norm": 1.7006670205208998, "language_loss": 0.66192997, "learning_rate": 1.4292686106267274e-06, "loss": 0.68131304, "num_input_tokens_seen": 108387220, "step": 5026, "time_per_iteration": 2.681972026824951 }, { "auxiliary_loss_clip": 0.01166168, "auxiliary_loss_mlp": 0.01026936, "balance_loss_clip": 1.05024803, "balance_loss_mlp": 1.01952171, "epoch": 0.6044610112427102, "flos": 16180023222720.0, "grad_norm": 2.1147330584995006, "language_loss": 0.76911694, "learning_rate": 1.4285220729353876e-06, "loss": 0.79104799, "num_input_tokens_seen": 108405760, "step": 5027, "time_per_iteration": 2.502760171890259 }, { "auxiliary_loss_clip": 0.01147062, "auxiliary_loss_mlp": 0.01024309, "balance_loss_clip": 1.04583931, "balance_loss_mlp": 1.01704907, "epoch": 0.6045812541333494, "flos": 13804064639040.0, "grad_norm": 3.0143960807716543, "language_loss": 0.78017527, "learning_rate": 1.4277756219367957e-06, "loss": 0.80188894, "num_input_tokens_seen": 108422785, "step": 5028, "time_per_iteration": 2.5049147605895996 }, { "auxiliary_loss_clip": 0.01142236, "auxiliary_loss_mlp": 0.01027586, "balance_loss_clip": 1.04952192, "balance_loss_mlp": 1.02026629, "epoch": 0.6047014970239885, "flos": 19975063170240.0, "grad_norm": 2.228764419510843, "language_loss": 0.79236388, "learning_rate": 1.4270292577441864e-06, "loss": 0.81406212, "num_input_tokens_seen": 108442290, "step": 5029, "time_per_iteration": 3.3179004192352295 }, { "auxiliary_loss_clip": 0.01165502, "auxiliary_loss_mlp": 0.01031504, "balance_loss_clip": 1.04862463, "balance_loss_mlp": 1.02434516, "epoch": 0.6048217399146275, "flos": 25337100886560.0, "grad_norm": 1.8854303756477955, "language_loss": 0.71853697, "learning_rate": 1.4262829804707836e-06, "loss": 0.74050707, "num_input_tokens_seen": 108464280, "step": 5030, "time_per_iteration": 2.5450100898742676 }, { "auxiliary_loss_clip": 0.01165323, "auxiliary_loss_mlp": 0.0102972, "balance_loss_clip": 1.04948628, "balance_loss_mlp": 1.02167296, "epoch": 0.6049419828052667, "flos": 26030823341760.0, "grad_norm": 1.5082746280296255, "language_loss": 0.69984907, "learning_rate": 1.4255367902297958e-06, "loss": 0.72179949, "num_input_tokens_seen": 108485610, "step": 5031, "time_per_iteration": 3.4087562561035156 }, { "auxiliary_loss_clip": 0.01175016, "auxiliary_loss_mlp": 0.01029461, "balance_loss_clip": 1.05192626, "balance_loss_mlp": 1.02254987, "epoch": 0.6050622256959057, "flos": 14648109935520.0, "grad_norm": 2.320260786062127, "language_loss": 0.78806931, "learning_rate": 1.4247906871344215e-06, "loss": 0.81011409, "num_input_tokens_seen": 108501005, "step": 5032, "time_per_iteration": 2.4546940326690674 }, { "auxiliary_loss_clip": 0.01140588, "auxiliary_loss_mlp": 0.01024603, "balance_loss_clip": 1.04390788, "balance_loss_mlp": 1.01779008, "epoch": 0.6051824685865448, "flos": 23331454111200.0, "grad_norm": 2.2552443337126955, "language_loss": 0.75345391, "learning_rate": 1.4240446712978415e-06, "loss": 0.77510583, "num_input_tokens_seen": 108519990, "step": 5033, "time_per_iteration": 3.270510673522949 }, { "auxiliary_loss_clip": 0.01166205, "auxiliary_loss_mlp": 0.01030886, "balance_loss_clip": 1.0506109, "balance_loss_mlp": 1.02398682, "epoch": 0.605302711477184, "flos": 27563311300320.0, "grad_norm": 1.827568521318529, "language_loss": 0.74714327, "learning_rate": 1.423298742833227e-06, "loss": 0.7691142, "num_input_tokens_seen": 108538650, "step": 5034, "time_per_iteration": 2.5622036457061768 }, { "auxiliary_loss_clip": 0.01139161, "auxiliary_loss_mlp": 0.01029038, "balance_loss_clip": 1.04684412, "balance_loss_mlp": 1.02182603, "epoch": 0.605422954367823, "flos": 15154701053760.0, "grad_norm": 2.2828201320439376, "language_loss": 0.71435893, "learning_rate": 1.4225529018537352e-06, "loss": 0.73604089, "num_input_tokens_seen": 108554155, "step": 5035, "time_per_iteration": 2.551840305328369 }, { "auxiliary_loss_clip": 0.01174736, "auxiliary_loss_mlp": 0.01027186, "balance_loss_clip": 1.05149412, "balance_loss_mlp": 1.02014697, "epoch": 0.6055431972584621, "flos": 27673916372160.0, "grad_norm": 1.4861771878662284, "language_loss": 0.77874863, "learning_rate": 1.4218071484725082e-06, "loss": 0.8007679, "num_input_tokens_seen": 108576275, "step": 5036, "time_per_iteration": 2.540121555328369 }, { "auxiliary_loss_clip": 0.01145362, "auxiliary_loss_mlp": 0.01024243, "balance_loss_clip": 1.05060065, "balance_loss_mlp": 1.01715922, "epoch": 0.6056634401491012, "flos": 19387492083840.0, "grad_norm": 1.9995683750749689, "language_loss": 0.76550937, "learning_rate": 1.4210614828026786e-06, "loss": 0.7872054, "num_input_tokens_seen": 108594125, "step": 5037, "time_per_iteration": 3.2967922687530518 }, { "auxiliary_loss_clip": 0.01171981, "auxiliary_loss_mlp": 0.01022646, "balance_loss_clip": 1.0494467, "balance_loss_mlp": 1.01580977, "epoch": 0.6057836830397403, "flos": 24789463071360.0, "grad_norm": 1.8772749537479174, "language_loss": 0.74587846, "learning_rate": 1.4203159049573605e-06, "loss": 0.76782477, "num_input_tokens_seen": 108615360, "step": 5038, "time_per_iteration": 2.5081779956817627 }, { "auxiliary_loss_clip": 0.01153134, "auxiliary_loss_mlp": 0.0102807, "balance_loss_clip": 1.04856479, "balance_loss_mlp": 1.02069724, "epoch": 0.6059039259303793, "flos": 20558252387520.0, "grad_norm": 2.2193729396765782, "language_loss": 0.86900049, "learning_rate": 1.4195704150496593e-06, "loss": 0.89081252, "num_input_tokens_seen": 108633075, "step": 5039, "time_per_iteration": 2.5240962505340576 }, { "auxiliary_loss_clip": 0.01150555, "auxiliary_loss_mlp": 0.01030439, "balance_loss_clip": 1.05181122, "balance_loss_mlp": 1.02337575, "epoch": 0.6060241688210185, "flos": 21069728212320.0, "grad_norm": 1.667530804265775, "language_loss": 0.73672426, "learning_rate": 1.4188250131926639e-06, "loss": 0.75853425, "num_input_tokens_seen": 108651875, "step": 5040, "time_per_iteration": 2.5317330360412598 }, { "auxiliary_loss_clip": 0.01148213, "auxiliary_loss_mlp": 0.01030354, "balance_loss_clip": 1.04740155, "balance_loss_mlp": 1.02272487, "epoch": 0.6061444117116576, "flos": 16361084593440.0, "grad_norm": 2.863346473361468, "language_loss": 0.80355656, "learning_rate": 1.4180796994994525e-06, "loss": 0.82534218, "num_input_tokens_seen": 108669290, "step": 5041, "time_per_iteration": 2.5014755725860596 }, { "auxiliary_loss_clip": 0.01146391, "auxiliary_loss_mlp": 0.01024381, "balance_loss_clip": 1.04716325, "balance_loss_mlp": 1.01745749, "epoch": 0.6062646546022966, "flos": 21507299710080.0, "grad_norm": 1.7804622646024435, "language_loss": 0.7211237, "learning_rate": 1.4173344740830877e-06, "loss": 0.74283135, "num_input_tokens_seen": 108688420, "step": 5042, "time_per_iteration": 2.545535087585449 }, { "auxiliary_loss_clip": 0.01141216, "auxiliary_loss_mlp": 0.01029231, "balance_loss_clip": 1.04963064, "balance_loss_mlp": 1.02184904, "epoch": 0.6063848974929358, "flos": 38983156454880.0, "grad_norm": 1.6742261238465979, "language_loss": 0.70542711, "learning_rate": 1.4165893370566206e-06, "loss": 0.72713155, "num_input_tokens_seen": 108712175, "step": 5043, "time_per_iteration": 2.7035396099090576 }, { "auxiliary_loss_clip": 0.01156548, "auxiliary_loss_mlp": 0.01027684, "balance_loss_clip": 1.04766297, "balance_loss_mlp": 1.01973033, "epoch": 0.6065051403835748, "flos": 19646588232000.0, "grad_norm": 1.7714764191950456, "language_loss": 0.77318776, "learning_rate": 1.4158442885330865e-06, "loss": 0.79503012, "num_input_tokens_seen": 108730745, "step": 5044, "time_per_iteration": 2.507781982421875 }, { "auxiliary_loss_clip": 0.01153814, "auxiliary_loss_mlp": 0.01027655, "balance_loss_clip": 1.04679847, "balance_loss_mlp": 1.02012134, "epoch": 0.6066253832742139, "flos": 23513090153280.0, "grad_norm": 1.9319689538274518, "language_loss": 0.78559822, "learning_rate": 1.4150993286255094e-06, "loss": 0.80741292, "num_input_tokens_seen": 108749995, "step": 5045, "time_per_iteration": 2.5055058002471924 }, { "auxiliary_loss_clip": 0.011771, "auxiliary_loss_mlp": 0.01029057, "balance_loss_clip": 1.05185664, "balance_loss_mlp": 1.02257836, "epoch": 0.6067456261648531, "flos": 19133711645760.0, "grad_norm": 1.80394594160717, "language_loss": 0.79707611, "learning_rate": 1.4143544574468993e-06, "loss": 0.81913769, "num_input_tokens_seen": 108768355, "step": 5046, "time_per_iteration": 2.452178955078125 }, { "auxiliary_loss_clip": 0.01159902, "auxiliary_loss_mlp": 0.01027028, "balance_loss_clip": 1.04993081, "balance_loss_mlp": 1.0202992, "epoch": 0.6068658690554921, "flos": 20520617801760.0, "grad_norm": 1.613545997424795, "language_loss": 0.8237226, "learning_rate": 1.4136096751102523e-06, "loss": 0.8455919, "num_input_tokens_seen": 108786685, "step": 5047, "time_per_iteration": 2.48578143119812 }, { "auxiliary_loss_clip": 0.01150951, "auxiliary_loss_mlp": 0.01028602, "balance_loss_clip": 1.048563, "balance_loss_mlp": 1.02164614, "epoch": 0.6069861119461312, "flos": 27374563700160.0, "grad_norm": 1.9623792216120557, "language_loss": 0.8303231, "learning_rate": 1.4128649817285516e-06, "loss": 0.85211867, "num_input_tokens_seen": 108804820, "step": 5048, "time_per_iteration": 2.570906639099121 }, { "auxiliary_loss_clip": 0.01150082, "auxiliary_loss_mlp": 0.01026725, "balance_loss_clip": 1.04736781, "balance_loss_mlp": 1.01953411, "epoch": 0.6071063548367702, "flos": 25626504560640.0, "grad_norm": 1.736559055923634, "language_loss": 0.63196969, "learning_rate": 1.412120377414766e-06, "loss": 0.65373778, "num_input_tokens_seen": 108825010, "step": 5049, "time_per_iteration": 2.5743041038513184 }, { "auxiliary_loss_clip": 0.01175691, "auxiliary_loss_mlp": 0.01026423, "balance_loss_clip": 1.05311704, "balance_loss_mlp": 1.0195806, "epoch": 0.6072265977274094, "flos": 24460521212640.0, "grad_norm": 1.4854605082305408, "language_loss": 0.71243924, "learning_rate": 1.4113758622818522e-06, "loss": 0.73446041, "num_input_tokens_seen": 108845075, "step": 5050, "time_per_iteration": 2.496579647064209 }, { "auxiliary_loss_clip": 0.01153011, "auxiliary_loss_mlp": 0.00761061, "balance_loss_clip": 1.04941976, "balance_loss_mlp": 1.00023866, "epoch": 0.6073468406180484, "flos": 18149256588960.0, "grad_norm": 2.156510687277369, "language_loss": 0.8307358, "learning_rate": 1.410631436442751e-06, "loss": 0.84987652, "num_input_tokens_seen": 108863870, "step": 5051, "time_per_iteration": 2.520833969116211 }, { "auxiliary_loss_clip": 0.01167809, "auxiliary_loss_mlp": 0.01023179, "balance_loss_clip": 1.05029082, "balance_loss_mlp": 1.01590776, "epoch": 0.6074670835086875, "flos": 20697620556000.0, "grad_norm": 2.052799624357449, "language_loss": 0.86382055, "learning_rate": 1.4098871000103936e-06, "loss": 0.8857305, "num_input_tokens_seen": 108882470, "step": 5052, "time_per_iteration": 2.499155044555664 }, { "auxiliary_loss_clip": 0.01145821, "auxiliary_loss_mlp": 0.01027441, "balance_loss_clip": 1.0483526, "balance_loss_mlp": 1.02002621, "epoch": 0.6075873263993267, "flos": 23769959449920.0, "grad_norm": 1.8619065764773788, "language_loss": 0.82582068, "learning_rate": 1.409142853097693e-06, "loss": 0.84755337, "num_input_tokens_seen": 108902710, "step": 5053, "time_per_iteration": 2.565490484237671 }, { "auxiliary_loss_clip": 0.01150173, "auxiliary_loss_mlp": 0.01027116, "balance_loss_clip": 1.04996002, "balance_loss_mlp": 1.01979041, "epoch": 0.6077075692899657, "flos": 24454487163360.0, "grad_norm": 2.6392403914260387, "language_loss": 0.79703248, "learning_rate": 1.408398695817553e-06, "loss": 0.8188054, "num_input_tokens_seen": 108919935, "step": 5054, "time_per_iteration": 2.5477452278137207 }, { "auxiliary_loss_clip": 0.0114799, "auxiliary_loss_mlp": 0.0104138, "balance_loss_clip": 1.04739785, "balance_loss_mlp": 1.03311884, "epoch": 0.6078278121806048, "flos": 27382106261760.0, "grad_norm": 1.7939895831974197, "language_loss": 0.70149529, "learning_rate": 1.4076546282828593e-06, "loss": 0.72338897, "num_input_tokens_seen": 108942790, "step": 5055, "time_per_iteration": 3.5948164463043213 }, { "auxiliary_loss_clip": 0.01152382, "auxiliary_loss_mlp": 0.01023151, "balance_loss_clip": 1.04676771, "balance_loss_mlp": 1.01597738, "epoch": 0.6079480550712439, "flos": 38436452480640.0, "grad_norm": 2.2629105648292387, "language_loss": 0.66272604, "learning_rate": 1.4069106506064874e-06, "loss": 0.68448138, "num_input_tokens_seen": 108964215, "step": 5056, "time_per_iteration": 2.7171974182128906 }, { "auxiliary_loss_clip": 0.01144757, "auxiliary_loss_mlp": 0.01031225, "balance_loss_clip": 1.04978001, "balance_loss_mlp": 1.02455544, "epoch": 0.608068297961883, "flos": 25336274796480.0, "grad_norm": 2.1263669768490723, "language_loss": 0.78298706, "learning_rate": 1.4061667629012989e-06, "loss": 0.80474687, "num_input_tokens_seen": 108984885, "step": 5057, "time_per_iteration": 3.356123685836792 }, { "auxiliary_loss_clip": 0.01138749, "auxiliary_loss_mlp": 0.01028716, "balance_loss_clip": 1.0482018, "balance_loss_mlp": 1.02115512, "epoch": 0.608188540852522, "flos": 24202466656320.0, "grad_norm": 1.5191149149271017, "language_loss": 0.83200699, "learning_rate": 1.40542296528014e-06, "loss": 0.85368162, "num_input_tokens_seen": 109004545, "step": 5058, "time_per_iteration": 2.5541763305664062 }, { "auxiliary_loss_clip": 0.01162119, "auxiliary_loss_mlp": 0.01024297, "balance_loss_clip": 1.04901314, "balance_loss_mlp": 1.01697111, "epoch": 0.6083087837431612, "flos": 21284150050560.0, "grad_norm": 1.843281101956526, "language_loss": 0.75810969, "learning_rate": 1.4046792578558452e-06, "loss": 0.7799738, "num_input_tokens_seen": 109022440, "step": 5059, "time_per_iteration": 3.2844090461730957 }, { "auxiliary_loss_clip": 0.01144131, "auxiliary_loss_mlp": 0.01031818, "balance_loss_clip": 1.04810464, "balance_loss_mlp": 1.02459443, "epoch": 0.6084290266338003, "flos": 16471438246560.0, "grad_norm": 2.078270358650264, "language_loss": 0.75806504, "learning_rate": 1.4039356407412325e-06, "loss": 0.77982455, "num_input_tokens_seen": 109035680, "step": 5060, "time_per_iteration": 2.5163145065307617 }, { "auxiliary_loss_clip": 0.01061465, "auxiliary_loss_mlp": 0.01004676, "balance_loss_clip": 1.01081252, "balance_loss_mlp": 1.00321567, "epoch": 0.6085492695244393, "flos": 66443584829760.0, "grad_norm": 0.791987291891068, "language_loss": 0.57122588, "learning_rate": 1.40319211404911e-06, "loss": 0.59188724, "num_input_tokens_seen": 109090680, "step": 5061, "time_per_iteration": 3.0597686767578125 }, { "auxiliary_loss_clip": 0.0117542, "auxiliary_loss_mlp": 0.01028706, "balance_loss_clip": 1.05118012, "balance_loss_mlp": 1.02198231, "epoch": 0.6086695124150785, "flos": 23618990103360.0, "grad_norm": 1.701193825279203, "language_loss": 0.90546447, "learning_rate": 1.4024486778922691e-06, "loss": 0.92750567, "num_input_tokens_seen": 109108995, "step": 5062, "time_per_iteration": 2.5007922649383545 }, { "auxiliary_loss_clip": 0.01152349, "auxiliary_loss_mlp": 0.0102827, "balance_loss_clip": 1.0461067, "balance_loss_mlp": 1.02101612, "epoch": 0.6087897553057176, "flos": 20157058381920.0, "grad_norm": 1.8627219430097608, "language_loss": 0.77790415, "learning_rate": 1.4017053323834884e-06, "loss": 0.79971033, "num_input_tokens_seen": 109128825, "step": 5063, "time_per_iteration": 3.306638240814209 }, { "auxiliary_loss_clip": 0.01149079, "auxiliary_loss_mlp": 0.01021879, "balance_loss_clip": 1.04566121, "balance_loss_mlp": 1.01508415, "epoch": 0.6089099981963566, "flos": 25482539021280.0, "grad_norm": 1.8677643112928144, "language_loss": 0.75880307, "learning_rate": 1.4009620776355333e-06, "loss": 0.78051269, "num_input_tokens_seen": 109150425, "step": 5064, "time_per_iteration": 2.5818421840667725 }, { "auxiliary_loss_clip": 0.01158198, "auxiliary_loss_mlp": 0.01026349, "balance_loss_clip": 1.04844117, "balance_loss_mlp": 1.01868415, "epoch": 0.6090302410869958, "flos": 25332898602240.0, "grad_norm": 1.6977709817271502, "language_loss": 0.78870422, "learning_rate": 1.4002189137611553e-06, "loss": 0.81054974, "num_input_tokens_seen": 109169765, "step": 5065, "time_per_iteration": 2.5179202556610107 }, { "auxiliary_loss_clip": 0.01161132, "auxiliary_loss_mlp": 0.01031172, "balance_loss_clip": 1.04961145, "balance_loss_mlp": 1.02434182, "epoch": 0.6091504839776348, "flos": 23987362395840.0, "grad_norm": 1.5923774158331803, "language_loss": 0.69690865, "learning_rate": 1.3994758408730901e-06, "loss": 0.71883166, "num_input_tokens_seen": 109188950, "step": 5066, "time_per_iteration": 2.521817445755005 }, { "auxiliary_loss_clip": 0.01150207, "auxiliary_loss_mlp": 0.01024014, "balance_loss_clip": 1.05064297, "balance_loss_mlp": 1.01619983, "epoch": 0.6092707268682739, "flos": 29643041987520.0, "grad_norm": 2.749898715984687, "language_loss": 0.76474476, "learning_rate": 1.3987328590840629e-06, "loss": 0.78648698, "num_input_tokens_seen": 109209895, "step": 5067, "time_per_iteration": 2.578019857406616 }, { "auxiliary_loss_clip": 0.01158543, "auxiliary_loss_mlp": 0.01028122, "balance_loss_clip": 1.04831219, "balance_loss_mlp": 1.02119589, "epoch": 0.609390969758913, "flos": 24024961064640.0, "grad_norm": 2.570986388934434, "language_loss": 0.86381024, "learning_rate": 1.397989968506783e-06, "loss": 0.88567686, "num_input_tokens_seen": 109228905, "step": 5068, "time_per_iteration": 2.5190227031707764 }, { "auxiliary_loss_clip": 0.01181549, "auxiliary_loss_mlp": 0.01027762, "balance_loss_clip": 1.0537442, "balance_loss_mlp": 1.02080333, "epoch": 0.6095112126495521, "flos": 11102145304320.0, "grad_norm": 2.0904014805674547, "language_loss": 0.72385001, "learning_rate": 1.3972471692539458e-06, "loss": 0.74594307, "num_input_tokens_seen": 109243620, "step": 5069, "time_per_iteration": 2.4350993633270264 }, { "auxiliary_loss_clip": 0.01142479, "auxiliary_loss_mlp": 0.01023063, "balance_loss_clip": 1.0470705, "balance_loss_mlp": 1.0163188, "epoch": 0.6096314555401912, "flos": 17265493523040.0, "grad_norm": 1.94468404241686, "language_loss": 0.75059879, "learning_rate": 1.3965044614382348e-06, "loss": 0.77225423, "num_input_tokens_seen": 109259070, "step": 5070, "time_per_iteration": 2.4801156520843506 }, { "auxiliary_loss_clip": 0.01179569, "auxiliary_loss_mlp": 0.01033414, "balance_loss_clip": 1.0516181, "balance_loss_mlp": 1.0253315, "epoch": 0.6097516984308303, "flos": 21645913622400.0, "grad_norm": 2.686480645507562, "language_loss": 0.75731248, "learning_rate": 1.3957618451723162e-06, "loss": 0.77944231, "num_input_tokens_seen": 109275100, "step": 5071, "time_per_iteration": 2.5035736560821533 }, { "auxiliary_loss_clip": 0.01147045, "auxiliary_loss_mlp": 0.01027925, "balance_loss_clip": 1.04664063, "balance_loss_mlp": 1.02037299, "epoch": 0.6098719413214694, "flos": 27199213126080.0, "grad_norm": 2.169981944537852, "language_loss": 0.71316612, "learning_rate": 1.3950193205688457e-06, "loss": 0.73491579, "num_input_tokens_seen": 109294825, "step": 5072, "time_per_iteration": 2.566723585128784 }, { "auxiliary_loss_clip": 0.01147841, "auxiliary_loss_mlp": 0.01026572, "balance_loss_clip": 1.05083108, "balance_loss_mlp": 1.01973855, "epoch": 0.6099921842121084, "flos": 20412957920640.0, "grad_norm": 2.329458848502446, "language_loss": 0.8360244, "learning_rate": 1.3942768877404627e-06, "loss": 0.85776854, "num_input_tokens_seen": 109313790, "step": 5073, "time_per_iteration": 2.517483711242676 }, { "auxiliary_loss_clip": 0.01173081, "auxiliary_loss_mlp": 0.01023783, "balance_loss_clip": 1.04904747, "balance_loss_mlp": 1.01677382, "epoch": 0.6101124271027476, "flos": 23366143506240.0, "grad_norm": 1.4718146805904238, "language_loss": 0.73870438, "learning_rate": 1.393534546799795e-06, "loss": 0.76067305, "num_input_tokens_seen": 109333490, "step": 5074, "time_per_iteration": 2.481212854385376 }, { "auxiliary_loss_clip": 0.01136591, "auxiliary_loss_mlp": 0.0102644, "balance_loss_clip": 1.04536402, "balance_loss_mlp": 1.01880491, "epoch": 0.6102326699933867, "flos": 26687845052160.0, "grad_norm": 1.7289589854685141, "language_loss": 0.67822242, "learning_rate": 1.3927922978594536e-06, "loss": 0.69985271, "num_input_tokens_seen": 109354575, "step": 5075, "time_per_iteration": 2.5550990104675293 }, { "auxiliary_loss_clip": 0.0105684, "auxiliary_loss_mlp": 0.01001785, "balance_loss_clip": 1.01063359, "balance_loss_mlp": 1.000319, "epoch": 0.6103529128840257, "flos": 60644622120960.0, "grad_norm": 0.7872338147703597, "language_loss": 0.57444668, "learning_rate": 1.3920501410320387e-06, "loss": 0.59503293, "num_input_tokens_seen": 109410690, "step": 5076, "time_per_iteration": 3.0562021732330322 }, { "auxiliary_loss_clip": 0.0114768, "auxiliary_loss_mlp": 0.01026571, "balance_loss_clip": 1.04600465, "balance_loss_mlp": 1.01921606, "epoch": 0.6104731557746649, "flos": 19021310725920.0, "grad_norm": 2.1798851350821202, "language_loss": 0.76084256, "learning_rate": 1.3913080764301333e-06, "loss": 0.78258502, "num_input_tokens_seen": 109427650, "step": 5077, "time_per_iteration": 2.5448286533355713 }, { "auxiliary_loss_clip": 0.01128446, "auxiliary_loss_mlp": 0.01029561, "balance_loss_clip": 1.04377937, "balance_loss_mlp": 1.02232826, "epoch": 0.6105933986653039, "flos": 23366897762400.0, "grad_norm": 1.8082160506618086, "language_loss": 0.71347708, "learning_rate": 1.3905661041663085e-06, "loss": 0.73505712, "num_input_tokens_seen": 109448835, "step": 5078, "time_per_iteration": 2.627227783203125 }, { "auxiliary_loss_clip": 0.01161487, "auxiliary_loss_mlp": 0.01029676, "balance_loss_clip": 1.04930568, "balance_loss_mlp": 1.02171326, "epoch": 0.610713641555943, "flos": 34637569418400.0, "grad_norm": 2.518839911578646, "language_loss": 0.65157264, "learning_rate": 1.389824224353122e-06, "loss": 0.67348427, "num_input_tokens_seen": 109470425, "step": 5079, "time_per_iteration": 2.6261935234069824 }, { "auxiliary_loss_clip": 0.01160274, "auxiliary_loss_mlp": 0.01020554, "balance_loss_clip": 1.05093503, "balance_loss_mlp": 1.01349068, "epoch": 0.610833884446582, "flos": 26646475102560.0, "grad_norm": 1.5353860811492395, "language_loss": 0.76685309, "learning_rate": 1.389082437103115e-06, "loss": 0.78866142, "num_input_tokens_seen": 109489695, "step": 5080, "time_per_iteration": 2.5386176109313965 }, { "auxiliary_loss_clip": 0.0113269, "auxiliary_loss_mlp": 0.01027257, "balance_loss_clip": 1.04497075, "balance_loss_mlp": 1.01940727, "epoch": 0.6109541273372212, "flos": 21215130430080.0, "grad_norm": 3.0803407690793754, "language_loss": 0.77933681, "learning_rate": 1.3883407425288172e-06, "loss": 0.80093628, "num_input_tokens_seen": 109510030, "step": 5081, "time_per_iteration": 3.3253724575042725 }, { "auxiliary_loss_clip": 0.01143892, "auxiliary_loss_mlp": 0.01026774, "balance_loss_clip": 1.04599535, "balance_loss_mlp": 1.01928186, "epoch": 0.6110743702278603, "flos": 20084087895840.0, "grad_norm": 3.434871817335991, "language_loss": 0.79995191, "learning_rate": 1.3875991407427417e-06, "loss": 0.82165861, "num_input_tokens_seen": 109528255, "step": 5082, "time_per_iteration": -0.1474132537841797 }, { "auxiliary_loss_clip": 0.01040179, "auxiliary_loss_mlp": 0.01002085, "balance_loss_clip": 1.00852025, "balance_loss_mlp": 1.00056553, "epoch": 0.6111946131184993, "flos": 68302967380320.0, "grad_norm": 0.7696932311066584, "language_loss": 0.5822587, "learning_rate": 1.38685763185739e-06, "loss": 0.6026814, "num_input_tokens_seen": 109581915, "step": 5083, "time_per_iteration": 3.893010139465332 }, { "auxiliary_loss_clip": 0.0117536, "auxiliary_loss_mlp": 0.01030333, "balance_loss_clip": 1.05069709, "balance_loss_mlp": 1.02291775, "epoch": 0.6113148560091385, "flos": 19937680003200.0, "grad_norm": 2.930109711345365, "language_loss": 0.68141079, "learning_rate": 1.3861162159852476e-06, "loss": 0.70346773, "num_input_tokens_seen": 109600050, "step": 5084, "time_per_iteration": 2.4808318614959717 }, { "auxiliary_loss_clip": 0.01154614, "auxiliary_loss_mlp": 0.01029957, "balance_loss_clip": 1.05090761, "balance_loss_mlp": 1.02216065, "epoch": 0.6114350988997775, "flos": 23731857943680.0, "grad_norm": 1.7582683894876596, "language_loss": 0.80209851, "learning_rate": 1.3853748932387875e-06, "loss": 0.82394421, "num_input_tokens_seen": 109620690, "step": 5085, "time_per_iteration": 3.2647674083709717 }, { "auxiliary_loss_clip": 0.01135936, "auxiliary_loss_mlp": 0.01024897, "balance_loss_clip": 1.04464912, "balance_loss_mlp": 1.0167129, "epoch": 0.6115553417904166, "flos": 24023703971040.0, "grad_norm": 2.5346001908095226, "language_loss": 0.74830836, "learning_rate": 1.3846336637304671e-06, "loss": 0.76991671, "num_input_tokens_seen": 109638960, "step": 5086, "time_per_iteration": 2.5332865715026855 }, { "auxiliary_loss_clip": 0.0114022, "auxiliary_loss_mlp": 0.01027597, "balance_loss_clip": 1.04697812, "balance_loss_mlp": 1.02075183, "epoch": 0.6116755846810558, "flos": 23733546040800.0, "grad_norm": 2.071573709108332, "language_loss": 0.83129823, "learning_rate": 1.3838925275727316e-06, "loss": 0.85297644, "num_input_tokens_seen": 109659700, "step": 5087, "time_per_iteration": 2.552314281463623 }, { "auxiliary_loss_clip": 0.01177134, "auxiliary_loss_mlp": 0.01026664, "balance_loss_clip": 1.05271554, "balance_loss_mlp": 1.01967192, "epoch": 0.6117958275716948, "flos": 18661630337760.0, "grad_norm": 1.909282647369632, "language_loss": 0.79267597, "learning_rate": 1.3831514848780089e-06, "loss": 0.81471395, "num_input_tokens_seen": 109679275, "step": 5088, "time_per_iteration": 2.4654033184051514 }, { "auxiliary_loss_clip": 0.01154608, "auxiliary_loss_mlp": 0.01032081, "balance_loss_clip": 1.04812634, "balance_loss_mlp": 1.0253427, "epoch": 0.6119160704623339, "flos": 16471186827840.0, "grad_norm": 2.0996996525461316, "language_loss": 0.91465843, "learning_rate": 1.3824105357587152e-06, "loss": 0.93652534, "num_input_tokens_seen": 109696380, "step": 5089, "time_per_iteration": 3.2352545261383057 }, { "auxiliary_loss_clip": 0.0114247, "auxiliary_loss_mlp": 0.01026061, "balance_loss_clip": 1.04566181, "balance_loss_mlp": 1.01913798, "epoch": 0.612036313352973, "flos": 23915469418560.0, "grad_norm": 1.482500551530955, "language_loss": 0.82528013, "learning_rate": 1.381669680327253e-06, "loss": 0.84696543, "num_input_tokens_seen": 109718060, "step": 5090, "time_per_iteration": 2.56117844581604 }, { "auxiliary_loss_clip": 0.01142258, "auxiliary_loss_mlp": 0.01028065, "balance_loss_clip": 1.04950869, "balance_loss_mlp": 1.02075481, "epoch": 0.6121565562436121, "flos": 26974770456000.0, "grad_norm": 1.8896260619107073, "language_loss": 0.70898199, "learning_rate": 1.380928918696008e-06, "loss": 0.73068523, "num_input_tokens_seen": 109736830, "step": 5091, "time_per_iteration": 2.5637755393981934 }, { "auxiliary_loss_clip": 0.01158442, "auxiliary_loss_mlp": 0.01024646, "balance_loss_clip": 1.04862738, "balance_loss_mlp": 1.01730859, "epoch": 0.6122767991342511, "flos": 15668870650560.0, "grad_norm": 2.3231772514500166, "language_loss": 0.71621835, "learning_rate": 1.3801882509773548e-06, "loss": 0.73804927, "num_input_tokens_seen": 109754690, "step": 5092, "time_per_iteration": 2.472471237182617 }, { "auxiliary_loss_clip": 0.01155591, "auxiliary_loss_mlp": 0.01024373, "balance_loss_clip": 1.0461551, "balance_loss_mlp": 1.01739931, "epoch": 0.6123970420248903, "flos": 27964325721120.0, "grad_norm": 1.8812428558668777, "language_loss": 0.81429052, "learning_rate": 1.3794476772836503e-06, "loss": 0.83609009, "num_input_tokens_seen": 109775790, "step": 5093, "time_per_iteration": 2.576965093612671 }, { "auxiliary_loss_clip": 0.01123959, "auxiliary_loss_mlp": 0.01022369, "balance_loss_clip": 1.04551876, "balance_loss_mlp": 1.014853, "epoch": 0.6125172849155294, "flos": 21468731283360.0, "grad_norm": 1.6776691703836828, "language_loss": 0.84580886, "learning_rate": 1.3787071977272402e-06, "loss": 0.86727214, "num_input_tokens_seen": 109795050, "step": 5094, "time_per_iteration": 2.5834555625915527 }, { "auxiliary_loss_clip": 0.01117411, "auxiliary_loss_mlp": 0.01023505, "balance_loss_clip": 1.04817438, "balance_loss_mlp": 1.01589966, "epoch": 0.6126375278061684, "flos": 16248324504000.0, "grad_norm": 2.5172277974561807, "language_loss": 0.71804035, "learning_rate": 1.3779668124204535e-06, "loss": 0.7394495, "num_input_tokens_seen": 109811465, "step": 5095, "time_per_iteration": 2.5265543460845947 }, { "auxiliary_loss_clip": 0.01140169, "auxiliary_loss_mlp": 0.0102962, "balance_loss_clip": 1.04761302, "balance_loss_mlp": 1.02253938, "epoch": 0.6127577706968076, "flos": 20448868492320.0, "grad_norm": 1.5066654686421912, "language_loss": 0.80445492, "learning_rate": 1.3772265214756074e-06, "loss": 0.82615286, "num_input_tokens_seen": 109831225, "step": 5096, "time_per_iteration": 2.5313243865966797 }, { "auxiliary_loss_clip": 0.01163664, "auxiliary_loss_mlp": 0.01024236, "balance_loss_clip": 1.04724789, "balance_loss_mlp": 1.01745319, "epoch": 0.6128780135874466, "flos": 18260400415200.0, "grad_norm": 2.0739416787252924, "language_loss": 0.75316572, "learning_rate": 1.3764863250050025e-06, "loss": 0.7750448, "num_input_tokens_seen": 109849465, "step": 5097, "time_per_iteration": 2.4750306606292725 }, { "auxiliary_loss_clip": 0.01132859, "auxiliary_loss_mlp": 0.01029509, "balance_loss_clip": 1.04624069, "balance_loss_mlp": 1.02233863, "epoch": 0.6129982564780857, "flos": 24937092140640.0, "grad_norm": 2.2117901655245236, "language_loss": 0.80729818, "learning_rate": 1.3757462231209272e-06, "loss": 0.82892185, "num_input_tokens_seen": 109869770, "step": 5098, "time_per_iteration": 2.6077804565429688 }, { "auxiliary_loss_clip": 0.01140036, "auxiliary_loss_mlp": 0.01032532, "balance_loss_clip": 1.04607344, "balance_loss_mlp": 1.02488816, "epoch": 0.6131184993687249, "flos": 22492042102560.0, "grad_norm": 1.9286963431763717, "language_loss": 0.88743371, "learning_rate": 1.3750062159356525e-06, "loss": 0.90915942, "num_input_tokens_seen": 109889120, "step": 5099, "time_per_iteration": 2.5553455352783203 }, { "auxiliary_loss_clip": 0.01123353, "auxiliary_loss_mlp": 0.01028239, "balance_loss_clip": 1.04513359, "balance_loss_mlp": 1.02127409, "epoch": 0.6132387422593639, "flos": 15885842592960.0, "grad_norm": 1.6822247977369467, "language_loss": 0.8261863, "learning_rate": 1.3742663035614382e-06, "loss": 0.84770226, "num_input_tokens_seen": 109906490, "step": 5100, "time_per_iteration": 2.5266244411468506 }, { "auxiliary_loss_clip": 0.0117734, "auxiliary_loss_mlp": 0.01024766, "balance_loss_clip": 1.05156612, "balance_loss_mlp": 1.01779771, "epoch": 0.613358985150003, "flos": 25411543968000.0, "grad_norm": 1.8294512719695437, "language_loss": 0.80081403, "learning_rate": 1.3735264861105283e-06, "loss": 0.82283509, "num_input_tokens_seen": 109927130, "step": 5101, "time_per_iteration": 2.5096242427825928 }, { "auxiliary_loss_clip": 0.01131809, "auxiliary_loss_mlp": 0.01024104, "balance_loss_clip": 1.04327989, "balance_loss_mlp": 1.01696396, "epoch": 0.6134792280406421, "flos": 21361286904000.0, "grad_norm": 1.9430605598188504, "language_loss": 0.78481114, "learning_rate": 1.372786763695152e-06, "loss": 0.80637032, "num_input_tokens_seen": 109945890, "step": 5102, "time_per_iteration": 2.5555219650268555 }, { "auxiliary_loss_clip": 0.01161419, "auxiliary_loss_mlp": 0.01029647, "balance_loss_clip": 1.04718423, "balance_loss_mlp": 1.02236366, "epoch": 0.6135994709312812, "flos": 21211251398400.0, "grad_norm": 2.4906196893232706, "language_loss": 0.77438021, "learning_rate": 1.3720471364275257e-06, "loss": 0.79629087, "num_input_tokens_seen": 109965535, "step": 5103, "time_per_iteration": 2.5211992263793945 }, { "auxiliary_loss_clip": 0.01130176, "auxiliary_loss_mlp": 0.00762361, "balance_loss_clip": 1.04595423, "balance_loss_mlp": 1.00030255, "epoch": 0.6137197138219203, "flos": 14794050907680.0, "grad_norm": 2.0884410395186244, "language_loss": 0.78009725, "learning_rate": 1.3713076044198486e-06, "loss": 0.79902267, "num_input_tokens_seen": 109982345, "step": 5104, "time_per_iteration": 2.5127480030059814 }, { "auxiliary_loss_clip": 0.01139501, "auxiliary_loss_mlp": 0.0102481, "balance_loss_clip": 1.0460608, "balance_loss_mlp": 1.0173955, "epoch": 0.6138399567125594, "flos": 20084518899360.0, "grad_norm": 3.2013325915441055, "language_loss": 0.80929351, "learning_rate": 1.3705681677843086e-06, "loss": 0.83093655, "num_input_tokens_seen": 110000940, "step": 5105, "time_per_iteration": 2.5179378986358643 }, { "auxiliary_loss_clip": 0.01066627, "auxiliary_loss_mlp": 0.01002331, "balance_loss_clip": 1.00756359, "balance_loss_mlp": 1.00091219, "epoch": 0.6139601996031985, "flos": 60123850191840.0, "grad_norm": 0.786882409501228, "language_loss": 0.60621798, "learning_rate": 1.3698288266330768e-06, "loss": 0.62690759, "num_input_tokens_seen": 110061565, "step": 5106, "time_per_iteration": 3.1461167335510254 }, { "auxiliary_loss_clip": 0.01144624, "auxiliary_loss_mlp": 0.01023271, "balance_loss_clip": 1.05172253, "balance_loss_mlp": 1.01694739, "epoch": 0.6140804424938375, "flos": 23586707144640.0, "grad_norm": 2.313622515807742, "language_loss": 0.72644544, "learning_rate": 1.3690895810783113e-06, "loss": 0.74812436, "num_input_tokens_seen": 110080360, "step": 5107, "time_per_iteration": 3.301497459411621 }, { "auxiliary_loss_clip": 0.01107164, "auxiliary_loss_mlp": 0.00762333, "balance_loss_clip": 1.04060614, "balance_loss_mlp": 1.00028801, "epoch": 0.6142006853844767, "flos": 21398203150560.0, "grad_norm": 3.004557175593251, "language_loss": 0.71517628, "learning_rate": 1.3683504312321543e-06, "loss": 0.73387122, "num_input_tokens_seen": 110100695, "step": 5108, "time_per_iteration": 2.6641292572021484 }, { "auxiliary_loss_clip": 0.01164431, "auxiliary_loss_mlp": 0.01023764, "balance_loss_clip": 1.0482285, "balance_loss_mlp": 1.01670372, "epoch": 0.6143209282751158, "flos": 12057370344000.0, "grad_norm": 1.8770636507463734, "language_loss": 0.80184495, "learning_rate": 1.3676113772067355e-06, "loss": 0.82372689, "num_input_tokens_seen": 110117750, "step": 5109, "time_per_iteration": 3.27534818649292 }, { "auxiliary_loss_clip": 0.01125124, "auxiliary_loss_mlp": 0.01026457, "balance_loss_clip": 1.04656613, "balance_loss_mlp": 1.01894712, "epoch": 0.6144411711657548, "flos": 25082278856640.0, "grad_norm": 1.9017210997371357, "language_loss": 0.72801667, "learning_rate": 1.3668724191141671e-06, "loss": 0.74953246, "num_input_tokens_seen": 110137020, "step": 5110, "time_per_iteration": 2.6766955852508545 }, { "auxiliary_loss_clip": 0.01133578, "auxiliary_loss_mlp": 0.01027573, "balance_loss_clip": 1.05387878, "balance_loss_mlp": 1.02023292, "epoch": 0.6145614140563939, "flos": 20114072169120.0, "grad_norm": 2.527951048417075, "language_loss": 0.6608538, "learning_rate": 1.3661335570665493e-06, "loss": 0.68246537, "num_input_tokens_seen": 110154930, "step": 5111, "time_per_iteration": 3.34250807762146 }, { "auxiliary_loss_clip": 0.01151305, "auxiliary_loss_mlp": 0.01033574, "balance_loss_clip": 1.0504396, "balance_loss_mlp": 1.02649307, "epoch": 0.614681656947033, "flos": 16800380105280.0, "grad_norm": 2.7714463836305145, "language_loss": 0.69772857, "learning_rate": 1.3653947911759676e-06, "loss": 0.71957737, "num_input_tokens_seen": 110172480, "step": 5112, "time_per_iteration": 2.504329204559326 }, { "auxiliary_loss_clip": 0.01113273, "auxiliary_loss_mlp": 0.01022617, "balance_loss_clip": 1.04438162, "balance_loss_mlp": 1.01532435, "epoch": 0.6148018998376721, "flos": 38801592246720.0, "grad_norm": 1.5445631653250365, "language_loss": 0.74351239, "learning_rate": 1.3646561215544904e-06, "loss": 0.7648713, "num_input_tokens_seen": 110197120, "step": 5113, "time_per_iteration": 2.749933958053589 }, { "auxiliary_loss_clip": 0.01159105, "auxiliary_loss_mlp": 0.01027612, "balance_loss_clip": 1.04848218, "balance_loss_mlp": 1.01999497, "epoch": 0.6149221427283111, "flos": 23327036325120.0, "grad_norm": 2.5678862959214617, "language_loss": 0.79589373, "learning_rate": 1.363917548314176e-06, "loss": 0.81776094, "num_input_tokens_seen": 110216385, "step": 5114, "time_per_iteration": 2.520686149597168 }, { "auxiliary_loss_clip": 0.01165743, "auxiliary_loss_mlp": 0.01027302, "balance_loss_clip": 1.05052471, "balance_loss_mlp": 1.019786, "epoch": 0.6150423856189503, "flos": 22379497514880.0, "grad_norm": 1.688134835088119, "language_loss": 0.72908193, "learning_rate": 1.3631790715670626e-06, "loss": 0.75101244, "num_input_tokens_seen": 110234790, "step": 5115, "time_per_iteration": 2.493579864501953 }, { "auxiliary_loss_clip": 0.01080981, "auxiliary_loss_mlp": 0.01021357, "balance_loss_clip": 1.04357588, "balance_loss_mlp": 1.01481807, "epoch": 0.6151626285095894, "flos": 18692081531520.0, "grad_norm": 1.8991123308486473, "language_loss": 0.85803431, "learning_rate": 1.3624406914251783e-06, "loss": 0.87905771, "num_input_tokens_seen": 110251910, "step": 5116, "time_per_iteration": 3.394636631011963 }, { "auxiliary_loss_clip": 0.01161951, "auxiliary_loss_mlp": 0.01024088, "balance_loss_clip": 1.04813004, "balance_loss_mlp": 1.01706064, "epoch": 0.6152828714002284, "flos": 15851691952320.0, "grad_norm": 1.9359519213831564, "language_loss": 0.88418633, "learning_rate": 1.3617024080005335e-06, "loss": 0.90604669, "num_input_tokens_seen": 110268810, "step": 5117, "time_per_iteration": 2.515165328979492 }, { "auxiliary_loss_clip": 0.01146144, "auxiliary_loss_mlp": 0.0076223, "balance_loss_clip": 1.04709172, "balance_loss_mlp": 1.0002979, "epoch": 0.6154031142908676, "flos": 24869796534240.0, "grad_norm": 1.5386053171470144, "language_loss": 0.74441051, "learning_rate": 1.3609642214051266e-06, "loss": 0.76349425, "num_input_tokens_seen": 110293035, "step": 5118, "time_per_iteration": 2.638821601867676 }, { "auxiliary_loss_clip": 0.01140585, "auxiliary_loss_mlp": 0.01028762, "balance_loss_clip": 1.04841113, "balance_loss_mlp": 1.02093017, "epoch": 0.6155233571815066, "flos": 19244747721120.0, "grad_norm": 1.8121094753966254, "language_loss": 0.66044444, "learning_rate": 1.3602261317509385e-06, "loss": 0.68213791, "num_input_tokens_seen": 110309695, "step": 5119, "time_per_iteration": 2.5248680114746094 }, { "auxiliary_loss_clip": 0.01163489, "auxiliary_loss_mlp": 0.01029738, "balance_loss_clip": 1.04886866, "balance_loss_mlp": 1.02184665, "epoch": 0.6156436000721457, "flos": 18770080392000.0, "grad_norm": 2.548485629421259, "language_loss": 0.8243463, "learning_rate": 1.3594881391499387e-06, "loss": 0.84627855, "num_input_tokens_seen": 110328610, "step": 5120, "time_per_iteration": 2.490475654602051 }, { "auxiliary_loss_clip": 0.01150148, "auxiliary_loss_mlp": 0.01026883, "balance_loss_clip": 1.04878557, "balance_loss_mlp": 1.01988578, "epoch": 0.6157638429627849, "flos": 18041201621280.0, "grad_norm": 1.7615123620946982, "language_loss": 0.79213202, "learning_rate": 1.3587502437140778e-06, "loss": 0.81390232, "num_input_tokens_seen": 110346775, "step": 5121, "time_per_iteration": 2.5392038822174072 }, { "auxiliary_loss_clip": 0.01149777, "auxiliary_loss_mlp": 0.01023588, "balance_loss_clip": 1.04656672, "balance_loss_mlp": 1.01605105, "epoch": 0.6158840858534239, "flos": 25556730684000.0, "grad_norm": 2.254693275481427, "language_loss": 0.84910911, "learning_rate": 1.3580124455552952e-06, "loss": 0.87084275, "num_input_tokens_seen": 110366140, "step": 5122, "time_per_iteration": 2.5722861289978027 }, { "auxiliary_loss_clip": 0.01161576, "auxiliary_loss_mlp": 0.00761706, "balance_loss_clip": 1.05039084, "balance_loss_mlp": 1.00028801, "epoch": 0.616004328744063, "flos": 24640792410240.0, "grad_norm": 1.7657283463642763, "language_loss": 0.87046194, "learning_rate": 1.3572747447855148e-06, "loss": 0.88969481, "num_input_tokens_seen": 110386550, "step": 5123, "time_per_iteration": 2.543165683746338 }, { "auxiliary_loss_clip": 0.01176599, "auxiliary_loss_mlp": 0.01020876, "balance_loss_clip": 1.05126286, "balance_loss_mlp": 1.01346993, "epoch": 0.6161245716347021, "flos": 21689690008320.0, "grad_norm": 1.855582539140864, "language_loss": 0.69317031, "learning_rate": 1.356537141516644e-06, "loss": 0.71514511, "num_input_tokens_seen": 110403970, "step": 5124, "time_per_iteration": 2.473790407180786 }, { "auxiliary_loss_clip": 0.01160564, "auxiliary_loss_mlp": 0.01027402, "balance_loss_clip": 1.05082273, "balance_loss_mlp": 1.01993334, "epoch": 0.6162448145253412, "flos": 35189229933120.0, "grad_norm": 1.921009562718812, "language_loss": 0.62025881, "learning_rate": 1.3557996358605775e-06, "loss": 0.64213848, "num_input_tokens_seen": 110423890, "step": 5125, "time_per_iteration": 2.617335796356201 }, { "auxiliary_loss_clip": 0.0115994, "auxiliary_loss_mlp": 0.01023799, "balance_loss_clip": 1.04779744, "balance_loss_mlp": 1.01731706, "epoch": 0.6163650574159802, "flos": 21615282843840.0, "grad_norm": 2.0867870406807607, "language_loss": 0.70054877, "learning_rate": 1.3550622279291941e-06, "loss": 0.72238618, "num_input_tokens_seen": 110442035, "step": 5126, "time_per_iteration": 2.4957029819488525 }, { "auxiliary_loss_clip": 0.01111633, "auxiliary_loss_mlp": 0.01028031, "balance_loss_clip": 1.04385281, "balance_loss_mlp": 1.02061641, "epoch": 0.6164853003066194, "flos": 24572167876320.0, "grad_norm": 1.4004073128337213, "language_loss": 0.83277267, "learning_rate": 1.354324917834358e-06, "loss": 0.85416937, "num_input_tokens_seen": 110463280, "step": 5127, "time_per_iteration": 2.6334779262542725 }, { "auxiliary_loss_clip": 0.01101061, "auxiliary_loss_mlp": 0.0076251, "balance_loss_clip": 1.04244375, "balance_loss_mlp": 1.00031281, "epoch": 0.6166055431972585, "flos": 21835990150080.0, "grad_norm": 1.7694270267493213, "language_loss": 0.76528466, "learning_rate": 1.353587705687918e-06, "loss": 0.78392035, "num_input_tokens_seen": 110481455, "step": 5128, "time_per_iteration": 2.6267638206481934 }, { "auxiliary_loss_clip": 0.01153097, "auxiliary_loss_mlp": 0.01028087, "balance_loss_clip": 1.04929876, "balance_loss_mlp": 1.02032709, "epoch": 0.6167257860878975, "flos": 17785266165600.0, "grad_norm": 2.790468368108481, "language_loss": 0.71945679, "learning_rate": 1.3528505916017096e-06, "loss": 0.74126858, "num_input_tokens_seen": 110499155, "step": 5129, "time_per_iteration": 2.5048043727874756 }, { "auxiliary_loss_clip": 0.01162325, "auxiliary_loss_mlp": 0.01023193, "balance_loss_clip": 1.04852366, "balance_loss_mlp": 1.01607013, "epoch": 0.6168460289785367, "flos": 23214814990080.0, "grad_norm": 2.0374971926669847, "language_loss": 0.88605964, "learning_rate": 1.3521135756875514e-06, "loss": 0.90791482, "num_input_tokens_seen": 110515470, "step": 5130, "time_per_iteration": 2.4898829460144043 }, { "auxiliary_loss_clip": 0.01095101, "auxiliary_loss_mlp": 0.01026616, "balance_loss_clip": 1.0421176, "balance_loss_mlp": 1.01967239, "epoch": 0.6169662718691757, "flos": 26213285473920.0, "grad_norm": 1.5292295682407848, "language_loss": 0.86088294, "learning_rate": 1.3513766580572496e-06, "loss": 0.88210011, "num_input_tokens_seen": 110538290, "step": 5131, "time_per_iteration": 2.6858532428741455 }, { "auxiliary_loss_clip": 0.01159721, "auxiliary_loss_mlp": 0.01027232, "balance_loss_clip": 1.04909825, "balance_loss_mlp": 1.02074659, "epoch": 0.6170865147598148, "flos": 19026123598560.0, "grad_norm": 2.399696999925704, "language_loss": 0.77339733, "learning_rate": 1.3506398388225924e-06, "loss": 0.79526687, "num_input_tokens_seen": 110555610, "step": 5132, "time_per_iteration": 2.4774069786071777 }, { "auxiliary_loss_clip": 0.01175747, "auxiliary_loss_mlp": 0.01025347, "balance_loss_clip": 1.05275202, "balance_loss_mlp": 1.01846886, "epoch": 0.617206757650454, "flos": 18260364498240.0, "grad_norm": 1.7600636561523528, "language_loss": 0.71928787, "learning_rate": 1.349903118095355e-06, "loss": 0.74129879, "num_input_tokens_seen": 110574745, "step": 5133, "time_per_iteration": 3.2238333225250244 }, { "auxiliary_loss_clip": 0.01166161, "auxiliary_loss_mlp": 0.01024474, "balance_loss_clip": 1.05145991, "balance_loss_mlp": 1.01738715, "epoch": 0.617327000541093, "flos": 18186963008640.0, "grad_norm": 1.6174723294660125, "language_loss": 0.73224056, "learning_rate": 1.349166495987298e-06, "loss": 0.75414693, "num_input_tokens_seen": 110593310, "step": 5134, "time_per_iteration": 3.2484490871429443 }, { "auxiliary_loss_clip": 0.01056629, "auxiliary_loss_mlp": 0.01004143, "balance_loss_clip": 1.02436161, "balance_loss_mlp": 1.00276053, "epoch": 0.6174472434317321, "flos": 61833807825120.0, "grad_norm": 0.8170155640708476, "language_loss": 0.60896766, "learning_rate": 1.348429972610166e-06, "loss": 0.62957537, "num_input_tokens_seen": 110657615, "step": 5135, "time_per_iteration": 3.1925806999206543 }, { "auxiliary_loss_clip": 0.01028848, "auxiliary_loss_mlp": 0.010078, "balance_loss_clip": 1.02165937, "balance_loss_mlp": 1.006387, "epoch": 0.6175674863223712, "flos": 71230981565280.0, "grad_norm": 0.9087108984574845, "language_loss": 0.57775301, "learning_rate": 1.3476935480756897e-06, "loss": 0.5981195, "num_input_tokens_seen": 110714365, "step": 5136, "time_per_iteration": 3.062098979949951 }, { "auxiliary_loss_clip": 0.01123604, "auxiliary_loss_mlp": 0.0103314, "balance_loss_clip": 1.0435214, "balance_loss_mlp": 1.02474129, "epoch": 0.6176877292130103, "flos": 21835451395680.0, "grad_norm": 2.2643868646587246, "language_loss": 0.75405777, "learning_rate": 1.346957222495583e-06, "loss": 0.77562511, "num_input_tokens_seen": 110732160, "step": 5137, "time_per_iteration": 3.323436737060547 }, { "auxiliary_loss_clip": 0.0114917, "auxiliary_loss_mlp": 0.00762095, "balance_loss_clip": 1.05007052, "balance_loss_mlp": 1.00039315, "epoch": 0.6178079721036493, "flos": 17741741198400.0, "grad_norm": 3.5134358183819705, "language_loss": 0.70951617, "learning_rate": 1.3462209959815466e-06, "loss": 0.72862887, "num_input_tokens_seen": 110746900, "step": 5138, "time_per_iteration": 2.4845147132873535 }, { "auxiliary_loss_clip": 0.01149653, "auxiliary_loss_mlp": 0.01027953, "balance_loss_clip": 1.04917407, "balance_loss_mlp": 1.02122998, "epoch": 0.6179282149942885, "flos": 22633134285120.0, "grad_norm": 1.993008940123491, "language_loss": 0.74107254, "learning_rate": 1.345484868645265e-06, "loss": 0.76284862, "num_input_tokens_seen": 110765710, "step": 5139, "time_per_iteration": 2.5509066581726074 }, { "auxiliary_loss_clip": 0.01137723, "auxiliary_loss_mlp": 0.01032881, "balance_loss_clip": 1.04700756, "balance_loss_mlp": 1.02565372, "epoch": 0.6180484578849276, "flos": 22310334226560.0, "grad_norm": 2.4942073667050644, "language_loss": 0.78823149, "learning_rate": 1.3447488405984088e-06, "loss": 0.8099376, "num_input_tokens_seen": 110783970, "step": 5140, "time_per_iteration": 2.5566532611846924 }, { "auxiliary_loss_clip": 0.01147665, "auxiliary_loss_mlp": 0.01027176, "balance_loss_clip": 1.04902542, "balance_loss_mlp": 1.02019644, "epoch": 0.6181687007755666, "flos": 35225463757440.0, "grad_norm": 1.8898332242611775, "language_loss": 0.69737035, "learning_rate": 1.3440129119526322e-06, "loss": 0.71911871, "num_input_tokens_seen": 110806395, "step": 5141, "time_per_iteration": 2.6415200233459473 }, { "auxiliary_loss_clip": 0.01070093, "auxiliary_loss_mlp": 0.01002265, "balance_loss_clip": 1.01106453, "balance_loss_mlp": 1.00087059, "epoch": 0.6182889436662057, "flos": 61547378870400.0, "grad_norm": 0.8091957498763354, "language_loss": 0.51234186, "learning_rate": 1.3432770828195762e-06, "loss": 0.53306544, "num_input_tokens_seen": 110867380, "step": 5142, "time_per_iteration": 3.993584394454956 }, { "auxiliary_loss_clip": 0.01123391, "auxiliary_loss_mlp": 0.01028306, "balance_loss_clip": 1.04312325, "balance_loss_mlp": 1.02089381, "epoch": 0.6184091865568448, "flos": 19609995238080.0, "grad_norm": 2.486843585875106, "language_loss": 0.70262337, "learning_rate": 1.3425413533108635e-06, "loss": 0.72414041, "num_input_tokens_seen": 110885980, "step": 5143, "time_per_iteration": 2.5576915740966797 }, { "auxiliary_loss_clip": 0.01122095, "auxiliary_loss_mlp": 0.01034136, "balance_loss_clip": 1.04866099, "balance_loss_mlp": 1.02676272, "epoch": 0.6185294294474839, "flos": 23586886729440.0, "grad_norm": 2.2008754535006028, "language_loss": 0.70487744, "learning_rate": 1.341805723538105e-06, "loss": 0.72643971, "num_input_tokens_seen": 110906085, "step": 5144, "time_per_iteration": 2.6253435611724854 }, { "auxiliary_loss_clip": 0.01153728, "auxiliary_loss_mlp": 0.01026116, "balance_loss_clip": 1.05056727, "balance_loss_mlp": 1.01870716, "epoch": 0.618649672338123, "flos": 26762036714880.0, "grad_norm": 1.5552865049050995, "language_loss": 0.776842, "learning_rate": 1.3410701936128948e-06, "loss": 0.79864037, "num_input_tokens_seen": 110928865, "step": 5145, "time_per_iteration": 2.580479145050049 }, { "auxiliary_loss_clip": 0.01163433, "auxiliary_loss_mlp": 0.01027045, "balance_loss_clip": 1.05260706, "balance_loss_mlp": 1.02019978, "epoch": 0.6187699152287621, "flos": 14456632646400.0, "grad_norm": 2.6757844473229464, "language_loss": 0.84892249, "learning_rate": 1.340334763646812e-06, "loss": 0.87082726, "num_input_tokens_seen": 110943000, "step": 5146, "time_per_iteration": 2.4408745765686035 }, { "auxiliary_loss_clip": 0.01177953, "auxiliary_loss_mlp": 0.01024339, "balance_loss_clip": 1.0520395, "balance_loss_mlp": 1.01724052, "epoch": 0.6188901581194012, "flos": 20084770318080.0, "grad_norm": 1.635840783587502, "language_loss": 0.7436589, "learning_rate": 1.3395994337514218e-06, "loss": 0.7656818, "num_input_tokens_seen": 110963170, "step": 5147, "time_per_iteration": 2.4791259765625 }, { "auxiliary_loss_clip": 0.0115596, "auxiliary_loss_mlp": 0.01027418, "balance_loss_clip": 1.04912448, "balance_loss_mlp": 1.02046251, "epoch": 0.6190104010100402, "flos": 25700732140320.0, "grad_norm": 1.6207122187788896, "language_loss": 0.78487432, "learning_rate": 1.3388642040382725e-06, "loss": 0.8067081, "num_input_tokens_seen": 110983595, "step": 5148, "time_per_iteration": 2.5301201343536377 }, { "auxiliary_loss_clip": 0.01137014, "auxiliary_loss_mlp": 0.01024794, "balance_loss_clip": 1.04437149, "balance_loss_mlp": 1.01748013, "epoch": 0.6191306439006794, "flos": 30442377057120.0, "grad_norm": 1.8638113120966122, "language_loss": 0.84250963, "learning_rate": 1.3381290746188975e-06, "loss": 0.86412776, "num_input_tokens_seen": 111002965, "step": 5149, "time_per_iteration": 2.6468818187713623 }, { "auxiliary_loss_clip": 0.01165188, "auxiliary_loss_mlp": 0.01036421, "balance_loss_clip": 1.05419075, "balance_loss_mlp": 1.0287981, "epoch": 0.6192508867913185, "flos": 26685797785440.0, "grad_norm": 1.6443909712136164, "language_loss": 0.67081571, "learning_rate": 1.3373940456048152e-06, "loss": 0.69283175, "num_input_tokens_seen": 111022990, "step": 5150, "time_per_iteration": 2.534470796585083 }, { "auxiliary_loss_clip": 0.01177933, "auxiliary_loss_mlp": 0.01028269, "balance_loss_clip": 1.05274367, "balance_loss_mlp": 1.02120256, "epoch": 0.6193711296819575, "flos": 36722041144320.0, "grad_norm": 1.626762891377852, "language_loss": 0.59407228, "learning_rate": 1.3366591171075299e-06, "loss": 0.61613429, "num_input_tokens_seen": 111046495, "step": 5151, "time_per_iteration": 2.61529541015625 }, { "auxiliary_loss_clip": 0.01149669, "auxiliary_loss_mlp": 0.01032086, "balance_loss_clip": 1.05134046, "balance_loss_mlp": 1.02487123, "epoch": 0.6194913725725967, "flos": 25192560675840.0, "grad_norm": 1.9745436854915803, "language_loss": 0.9106369, "learning_rate": 1.335924289238529e-06, "loss": 0.93245447, "num_input_tokens_seen": 111065705, "step": 5152, "time_per_iteration": 2.5606956481933594 }, { "auxiliary_loss_clip": 0.01164051, "auxiliary_loss_mlp": 0.00763017, "balance_loss_clip": 1.05392504, "balance_loss_mlp": 1.00043428, "epoch": 0.6196116154632357, "flos": 21178824771840.0, "grad_norm": 1.7340631547117384, "language_loss": 0.76911706, "learning_rate": 1.3351895621092859e-06, "loss": 0.78838772, "num_input_tokens_seen": 111086050, "step": 5153, "time_per_iteration": 2.514832019805908 }, { "auxiliary_loss_clip": 0.01060719, "auxiliary_loss_mlp": 0.01029077, "balance_loss_clip": 1.032125, "balance_loss_mlp": 1.02157319, "epoch": 0.6197318583538748, "flos": 16253747964960.0, "grad_norm": 2.0136402786148033, "language_loss": 0.76587236, "learning_rate": 1.3344549358312567e-06, "loss": 0.78677034, "num_input_tokens_seen": 111104450, "step": 5154, "time_per_iteration": 2.7892115116119385 }, { "auxiliary_loss_clip": 0.01165438, "auxiliary_loss_mlp": 0.01025537, "balance_loss_clip": 1.05131638, "balance_loss_mlp": 1.01810682, "epoch": 0.619852101244514, "flos": 24425616315840.0, "grad_norm": 1.8674947497869956, "language_loss": 0.78179049, "learning_rate": 1.3337204105158852e-06, "loss": 0.80370021, "num_input_tokens_seen": 111123320, "step": 5155, "time_per_iteration": 2.6374175548553467 }, { "auxiliary_loss_clip": 0.0111858, "auxiliary_loss_mlp": 0.01025295, "balance_loss_clip": 1.03828418, "balance_loss_mlp": 1.01826131, "epoch": 0.619972344135153, "flos": 16727301868320.0, "grad_norm": 1.9511550628611742, "language_loss": 0.72944993, "learning_rate": 1.332985986274597e-06, "loss": 0.75088871, "num_input_tokens_seen": 111140950, "step": 5156, "time_per_iteration": 2.55111026763916 }, { "auxiliary_loss_clip": 0.01100901, "auxiliary_loss_mlp": 0.00761837, "balance_loss_clip": 1.04644012, "balance_loss_mlp": 1.00050116, "epoch": 0.6200925870257921, "flos": 12495193260480.0, "grad_norm": 1.9117237271562368, "language_loss": 0.75053269, "learning_rate": 1.3322516632188047e-06, "loss": 0.76916009, "num_input_tokens_seen": 111157845, "step": 5157, "time_per_iteration": 2.626483678817749 }, { "auxiliary_loss_clip": 0.0113162, "auxiliary_loss_mlp": 0.01021094, "balance_loss_clip": 1.04642272, "balance_loss_mlp": 1.01394701, "epoch": 0.6202128299164312, "flos": 26539353975840.0, "grad_norm": 1.734533554262632, "language_loss": 0.66921961, "learning_rate": 1.3315174414599045e-06, "loss": 0.69074678, "num_input_tokens_seen": 111179165, "step": 5158, "time_per_iteration": 2.608776807785034 }, { "auxiliary_loss_clip": 0.01156973, "auxiliary_loss_mlp": 0.01028051, "balance_loss_clip": 1.04889345, "balance_loss_mlp": 1.02018023, "epoch": 0.6203330728070703, "flos": 18770511395520.0, "grad_norm": 1.7682423389393507, "language_loss": 0.75197709, "learning_rate": 1.3307833211092768e-06, "loss": 0.77382731, "num_input_tokens_seen": 111197830, "step": 5159, "time_per_iteration": 3.352786064147949 }, { "auxiliary_loss_clip": 0.01182842, "auxiliary_loss_mlp": 0.01032095, "balance_loss_clip": 1.056638, "balance_loss_mlp": 1.02490687, "epoch": 0.6204533156977093, "flos": 20629786195200.0, "grad_norm": 1.5702892697417883, "language_loss": 0.75229067, "learning_rate": 1.3300493022782873e-06, "loss": 0.77444005, "num_input_tokens_seen": 111218400, "step": 5160, "time_per_iteration": 3.31372332572937 }, { "auxiliary_loss_clip": 0.01111403, "auxiliary_loss_mlp": 0.00762545, "balance_loss_clip": 1.04512084, "balance_loss_mlp": 1.0004636, "epoch": 0.6205735585883485, "flos": 17348053837440.0, "grad_norm": 1.8226871150403812, "language_loss": 0.72293961, "learning_rate": 1.3293153850782855e-06, "loss": 0.74167913, "num_input_tokens_seen": 111236720, "step": 5161, "time_per_iteration": 2.580244302749634 }, { "auxiliary_loss_clip": 0.0112684, "auxiliary_loss_mlp": 0.01026276, "balance_loss_clip": 1.04719496, "balance_loss_mlp": 1.01857495, "epoch": 0.6206938014789876, "flos": 22965021334560.0, "grad_norm": 1.7100892146102808, "language_loss": 0.71117222, "learning_rate": 1.3285815696206069e-06, "loss": 0.73270339, "num_input_tokens_seen": 111258265, "step": 5162, "time_per_iteration": 2.5944392681121826 }, { "auxiliary_loss_clip": 0.01138047, "auxiliary_loss_mlp": 0.01027706, "balance_loss_clip": 1.04654503, "balance_loss_mlp": 1.01977217, "epoch": 0.6208140443696266, "flos": 23983196028480.0, "grad_norm": 2.9064052801782907, "language_loss": 0.77001345, "learning_rate": 1.32784785601657e-06, "loss": 0.79167104, "num_input_tokens_seen": 111277675, "step": 5163, "time_per_iteration": 3.337707996368408 }, { "auxiliary_loss_clip": 0.01149099, "auxiliary_loss_mlp": 0.01020879, "balance_loss_clip": 1.04624915, "balance_loss_mlp": 1.01335704, "epoch": 0.6209342872602658, "flos": 35077295933760.0, "grad_norm": 1.6700776897271594, "language_loss": 0.73722076, "learning_rate": 1.3271142443774798e-06, "loss": 0.75892055, "num_input_tokens_seen": 111299910, "step": 5164, "time_per_iteration": 2.666945695877075 }, { "auxiliary_loss_clip": 0.01144782, "auxiliary_loss_mlp": 0.01022363, "balance_loss_clip": 1.04979122, "balance_loss_mlp": 1.01538348, "epoch": 0.6210545301509048, "flos": 26979331909920.0, "grad_norm": 1.7188524798031746, "language_loss": 0.81580794, "learning_rate": 1.3263807348146228e-06, "loss": 0.83747935, "num_input_tokens_seen": 111319765, "step": 5165, "time_per_iteration": 2.5917551517486572 }, { "auxiliary_loss_clip": 0.01145428, "auxiliary_loss_mlp": 0.01033195, "balance_loss_clip": 1.04439437, "balance_loss_mlp": 1.02510047, "epoch": 0.6211747730415439, "flos": 33618245381760.0, "grad_norm": 2.268323644383704, "language_loss": 0.73217738, "learning_rate": 1.3256473274392733e-06, "loss": 0.75396359, "num_input_tokens_seen": 111341110, "step": 5166, "time_per_iteration": 2.6478288173675537 }, { "auxiliary_loss_clip": 0.01178665, "auxiliary_loss_mlp": 0.01028251, "balance_loss_clip": 1.05358672, "balance_loss_mlp": 1.02072239, "epoch": 0.6212950159321831, "flos": 34167104373600.0, "grad_norm": 1.7942739751784698, "language_loss": 0.70259857, "learning_rate": 1.3249140223626873e-06, "loss": 0.72466773, "num_input_tokens_seen": 111362730, "step": 5167, "time_per_iteration": 2.607034206390381 }, { "auxiliary_loss_clip": 0.01160307, "auxiliary_loss_mlp": 0.01025849, "balance_loss_clip": 1.05112243, "balance_loss_mlp": 1.0194087, "epoch": 0.6214152588228221, "flos": 27965762399520.0, "grad_norm": 1.7983936574044987, "language_loss": 0.75473791, "learning_rate": 1.3241808196961077e-06, "loss": 0.77659947, "num_input_tokens_seen": 111383855, "step": 5168, "time_per_iteration": 3.2862675189971924 }, { "auxiliary_loss_clip": 0.01137433, "auxiliary_loss_mlp": 0.01023842, "balance_loss_clip": 1.04855943, "balance_loss_mlp": 1.01664495, "epoch": 0.6215355017134612, "flos": 20230208452800.0, "grad_norm": 1.743927076244528, "language_loss": 0.70551431, "learning_rate": 1.3234477195507608e-06, "loss": 0.72712702, "num_input_tokens_seen": 111402685, "step": 5169, "time_per_iteration": 2.544933795928955 }, { "auxiliary_loss_clip": 0.01134717, "auxiliary_loss_mlp": 0.01023784, "balance_loss_clip": 1.04961312, "balance_loss_mlp": 1.01636291, "epoch": 0.6216557446041003, "flos": 41428134659040.0, "grad_norm": 2.6065733761960383, "language_loss": 0.62622374, "learning_rate": 1.322714722037857e-06, "loss": 0.64780873, "num_input_tokens_seen": 111424130, "step": 5170, "time_per_iteration": 2.7281999588012695 }, { "auxiliary_loss_clip": 0.01141283, "auxiliary_loss_mlp": 0.01032709, "balance_loss_clip": 1.0464927, "balance_loss_mlp": 1.02539515, "epoch": 0.6217759874947394, "flos": 27928774319040.0, "grad_norm": 1.866185225067516, "language_loss": 0.77359641, "learning_rate": 1.321981827268591e-06, "loss": 0.79533625, "num_input_tokens_seen": 111444785, "step": 5171, "time_per_iteration": 2.6246533393859863 }, { "auxiliary_loss_clip": 0.01151018, "auxiliary_loss_mlp": 0.0102262, "balance_loss_clip": 1.04767966, "balance_loss_mlp": 1.0156579, "epoch": 0.6218962303853784, "flos": 21765677519040.0, "grad_norm": 1.65790167828245, "language_loss": 0.81425381, "learning_rate": 1.3212490353541426e-06, "loss": 0.83599025, "num_input_tokens_seen": 111467045, "step": 5172, "time_per_iteration": 2.555621862411499 }, { "auxiliary_loss_clip": 0.01177808, "auxiliary_loss_mlp": 0.01027078, "balance_loss_clip": 1.05158067, "balance_loss_mlp": 1.01949012, "epoch": 0.6220164732760175, "flos": 21246264046080.0, "grad_norm": 1.8003711825054938, "language_loss": 0.80236757, "learning_rate": 1.3205163464056762e-06, "loss": 0.8244164, "num_input_tokens_seen": 111483650, "step": 5173, "time_per_iteration": 2.469287395477295 }, { "auxiliary_loss_clip": 0.01159424, "auxiliary_loss_mlp": 0.01026589, "balance_loss_clip": 1.04912257, "balance_loss_mlp": 1.01963043, "epoch": 0.6221367161666567, "flos": 26136364122240.0, "grad_norm": 1.9587707410277424, "language_loss": 0.73021221, "learning_rate": 1.319783760534339e-06, "loss": 0.75207233, "num_input_tokens_seen": 111502895, "step": 5174, "time_per_iteration": 2.5286448001861572 }, { "auxiliary_loss_clip": 0.01164496, "auxiliary_loss_mlp": 0.01025687, "balance_loss_clip": 1.05301094, "balance_loss_mlp": 1.01797116, "epoch": 0.6222569590572957, "flos": 16284199158720.0, "grad_norm": 2.1030383291685517, "language_loss": 0.75124902, "learning_rate": 1.319051277851266e-06, "loss": 0.7731508, "num_input_tokens_seen": 111519180, "step": 5175, "time_per_iteration": 2.46807861328125 }, { "auxiliary_loss_clip": 0.01162502, "auxiliary_loss_mlp": 0.01024125, "balance_loss_clip": 1.05184615, "balance_loss_mlp": 1.01737809, "epoch": 0.6223772019479348, "flos": 18223843338240.0, "grad_norm": 1.9359563221142095, "language_loss": 0.83933014, "learning_rate": 1.3183188984675716e-06, "loss": 0.8611964, "num_input_tokens_seen": 111537545, "step": 5176, "time_per_iteration": 2.4606449604034424 }, { "auxiliary_loss_clip": 0.01149986, "auxiliary_loss_mlp": 0.01027076, "balance_loss_clip": 1.05127788, "balance_loss_mlp": 1.01939905, "epoch": 0.6224974448385739, "flos": 27489802059840.0, "grad_norm": 2.2143315303606634, "language_loss": 0.71079248, "learning_rate": 1.3175866224943586e-06, "loss": 0.73256314, "num_input_tokens_seen": 111556265, "step": 5177, "time_per_iteration": 2.581632614135742 }, { "auxiliary_loss_clip": 0.01152664, "auxiliary_loss_mlp": 0.01032808, "balance_loss_clip": 1.05059981, "balance_loss_mlp": 1.02495539, "epoch": 0.622617687729213, "flos": 19791954532800.0, "grad_norm": 2.181215997965316, "language_loss": 0.73703682, "learning_rate": 1.316854450042712e-06, "loss": 0.75889152, "num_input_tokens_seen": 111574205, "step": 5178, "time_per_iteration": 2.506304979324341 }, { "auxiliary_loss_clip": 0.01166278, "auxiliary_loss_mlp": 0.01034541, "balance_loss_clip": 1.05087483, "balance_loss_mlp": 1.02667916, "epoch": 0.622737930619852, "flos": 23038889744640.0, "grad_norm": 2.555269156085605, "language_loss": 0.74422574, "learning_rate": 1.3161223812237024e-06, "loss": 0.76623392, "num_input_tokens_seen": 111593560, "step": 5179, "time_per_iteration": 2.5121009349823 }, { "auxiliary_loss_clip": 0.01176456, "auxiliary_loss_mlp": 0.01027182, "balance_loss_clip": 1.05076814, "balance_loss_mlp": 1.01945782, "epoch": 0.6228581735104912, "flos": 12634274093280.0, "grad_norm": 5.808727328035773, "language_loss": 0.84785348, "learning_rate": 1.3153904161483842e-06, "loss": 0.86988986, "num_input_tokens_seen": 111608860, "step": 5180, "time_per_iteration": 2.405924081802368 }, { "auxiliary_loss_clip": 0.0113035, "auxiliary_loss_mlp": 0.0102507, "balance_loss_clip": 1.04479194, "balance_loss_mlp": 1.01753044, "epoch": 0.6229784164011303, "flos": 23802817080000.0, "grad_norm": 2.092921099655151, "language_loss": 0.8546524, "learning_rate": 1.3146585549277953e-06, "loss": 0.87620664, "num_input_tokens_seen": 111627500, "step": 5181, "time_per_iteration": 2.6005451679229736 }, { "auxiliary_loss_clip": 0.01154655, "auxiliary_loss_mlp": 0.01023553, "balance_loss_clip": 1.05106235, "balance_loss_mlp": 1.01599514, "epoch": 0.6230986592917693, "flos": 22414222826880.0, "grad_norm": 2.5264279716219153, "language_loss": 0.78349555, "learning_rate": 1.3139267976729591e-06, "loss": 0.80527765, "num_input_tokens_seen": 111647690, "step": 5182, "time_per_iteration": 2.5396244525909424 }, { "auxiliary_loss_clip": 0.01167017, "auxiliary_loss_mlp": 0.0102483, "balance_loss_clip": 1.05210495, "balance_loss_mlp": 1.01725161, "epoch": 0.6232189021824085, "flos": 34528221440160.0, "grad_norm": 1.8323142361570526, "language_loss": 0.71963167, "learning_rate": 1.3131951444948815e-06, "loss": 0.74155015, "num_input_tokens_seen": 111667090, "step": 5183, "time_per_iteration": 2.6053199768066406 }, { "auxiliary_loss_clip": 0.01154573, "auxiliary_loss_mlp": 0.01029787, "balance_loss_clip": 1.05252862, "balance_loss_mlp": 1.02255988, "epoch": 0.6233391450730476, "flos": 22237004570880.0, "grad_norm": 1.8556916149390859, "language_loss": 0.76239276, "learning_rate": 1.3124635955045546e-06, "loss": 0.78423637, "num_input_tokens_seen": 111686905, "step": 5184, "time_per_iteration": 2.525334358215332 }, { "auxiliary_loss_clip": 0.0110363, "auxiliary_loss_mlp": 0.00762216, "balance_loss_clip": 1.04109812, "balance_loss_mlp": 1.00044954, "epoch": 0.6234593879636866, "flos": 20332696291680.0, "grad_norm": 2.4770670159721817, "language_loss": 0.84368813, "learning_rate": 1.3117321508129537e-06, "loss": 0.86234665, "num_input_tokens_seen": 111704985, "step": 5185, "time_per_iteration": 3.3454854488372803 }, { "auxiliary_loss_clip": 0.01151358, "auxiliary_loss_mlp": 0.01025567, "balance_loss_clip": 1.05091929, "balance_loss_mlp": 1.01838756, "epoch": 0.6235796308543258, "flos": 20664906593760.0, "grad_norm": 1.962289826958264, "language_loss": 0.76263988, "learning_rate": 1.3110008105310388e-06, "loss": 0.78440911, "num_input_tokens_seen": 111724805, "step": 5186, "time_per_iteration": 3.3134069442749023 }, { "auxiliary_loss_clip": 0.01176201, "auxiliary_loss_mlp": 0.01026753, "balance_loss_clip": 1.0499264, "balance_loss_mlp": 1.0192163, "epoch": 0.6236998737449648, "flos": 26618645846880.0, "grad_norm": 1.6316910414407635, "language_loss": 0.77929783, "learning_rate": 1.3102695747697526e-06, "loss": 0.80132741, "num_input_tokens_seen": 111747675, "step": 5187, "time_per_iteration": 2.5263819694519043 }, { "auxiliary_loss_clip": 0.01109856, "auxiliary_loss_mlp": 0.01027261, "balance_loss_clip": 1.04954278, "balance_loss_mlp": 1.01967955, "epoch": 0.6238201166356039, "flos": 12674602451040.0, "grad_norm": 2.2257466343799126, "language_loss": 0.9034282, "learning_rate": 1.3095384436400237e-06, "loss": 0.92479932, "num_input_tokens_seen": 111759205, "step": 5188, "time_per_iteration": 2.56612491607666 }, { "auxiliary_loss_clip": 0.01152992, "auxiliary_loss_mlp": 0.01027034, "balance_loss_clip": 1.04927742, "balance_loss_mlp": 1.01994085, "epoch": 0.623940359526243, "flos": 10452163318080.0, "grad_norm": 2.1000228436020714, "language_loss": 0.82218385, "learning_rate": 1.3088074172527633e-06, "loss": 0.84398413, "num_input_tokens_seen": 111776335, "step": 5189, "time_per_iteration": 3.2286503314971924 }, { "auxiliary_loss_clip": 0.01153638, "auxiliary_loss_mlp": 0.01024113, "balance_loss_clip": 1.04826629, "balance_loss_mlp": 1.01628709, "epoch": 0.6240606024168821, "flos": 29059529517600.0, "grad_norm": 1.8513962903769459, "language_loss": 0.71470946, "learning_rate": 1.3080764957188684e-06, "loss": 0.73648703, "num_input_tokens_seen": 111796580, "step": 5190, "time_per_iteration": 2.6015195846557617 }, { "auxiliary_loss_clip": 0.01120408, "auxiliary_loss_mlp": 0.01023491, "balance_loss_clip": 1.04554927, "balance_loss_mlp": 1.01595759, "epoch": 0.6241808453075212, "flos": 22018093112640.0, "grad_norm": 2.1873770035789155, "language_loss": 0.71024245, "learning_rate": 1.3073456791492192e-06, "loss": 0.73168147, "num_input_tokens_seen": 111816290, "step": 5191, "time_per_iteration": 2.599382162094116 }, { "auxiliary_loss_clip": 0.01147889, "auxiliary_loss_mlp": 0.01029942, "balance_loss_clip": 1.04643536, "balance_loss_mlp": 1.02311981, "epoch": 0.6243010881981603, "flos": 21138711915840.0, "grad_norm": 2.011283750953758, "language_loss": 0.78283322, "learning_rate": 1.3066149676546801e-06, "loss": 0.80461156, "num_input_tokens_seen": 111834470, "step": 5192, "time_per_iteration": 2.531705379486084 }, { "auxiliary_loss_clip": 0.01147235, "auxiliary_loss_mlp": 0.01024218, "balance_loss_clip": 1.05161595, "balance_loss_mlp": 1.01722634, "epoch": 0.6244213310887994, "flos": 22344951787680.0, "grad_norm": 1.763015268288322, "language_loss": 0.66092616, "learning_rate": 1.3058843613460985e-06, "loss": 0.68264067, "num_input_tokens_seen": 111852410, "step": 5193, "time_per_iteration": 2.540518045425415 }, { "auxiliary_loss_clip": 0.01140418, "auxiliary_loss_mlp": 0.01024849, "balance_loss_clip": 1.04820108, "balance_loss_mlp": 1.01651585, "epoch": 0.6245415739794384, "flos": 15231981575040.0, "grad_norm": 8.049996584272952, "language_loss": 0.74246097, "learning_rate": 1.3051538603343075e-06, "loss": 0.76411366, "num_input_tokens_seen": 111870340, "step": 5194, "time_per_iteration": 3.2838854789733887 }, { "auxiliary_loss_clip": 0.0116353, "auxiliary_loss_mlp": 0.01033024, "balance_loss_clip": 1.05327392, "balance_loss_mlp": 1.02626169, "epoch": 0.6246618168700776, "flos": 18879895290720.0, "grad_norm": 1.7315587636070922, "language_loss": 0.68038392, "learning_rate": 1.3044234647301235e-06, "loss": 0.70234948, "num_input_tokens_seen": 111888365, "step": 5195, "time_per_iteration": 2.502189874649048 }, { "auxiliary_loss_clip": 0.0115802, "auxiliary_loss_mlp": 0.01026788, "balance_loss_clip": 1.04972243, "balance_loss_mlp": 1.01943839, "epoch": 0.6247820597607167, "flos": 14319203993760.0, "grad_norm": 1.795417118505714, "language_loss": 0.72456288, "learning_rate": 1.303693174644347e-06, "loss": 0.74641091, "num_input_tokens_seen": 111905840, "step": 5196, "time_per_iteration": 2.4684486389160156 }, { "auxiliary_loss_clip": 0.0114457, "auxiliary_loss_mlp": 0.01028036, "balance_loss_clip": 1.04779816, "balance_loss_mlp": 1.02009094, "epoch": 0.6249023026513557, "flos": 22637983074720.0, "grad_norm": 2.30499999904653, "language_loss": 0.80435771, "learning_rate": 1.3029629901877625e-06, "loss": 0.82608378, "num_input_tokens_seen": 111925215, "step": 5197, "time_per_iteration": 2.5359795093536377 }, { "auxiliary_loss_clip": 0.01168336, "auxiliary_loss_mlp": 0.01031085, "balance_loss_clip": 1.05243134, "balance_loss_mlp": 1.02310348, "epoch": 0.6250225455419949, "flos": 20266693695840.0, "grad_norm": 3.4205596977604587, "language_loss": 0.77459908, "learning_rate": 1.3022329114711376e-06, "loss": 0.79659331, "num_input_tokens_seen": 111943925, "step": 5198, "time_per_iteration": 2.487225294113159 }, { "auxiliary_loss_clip": 0.01144733, "auxiliary_loss_mlp": 0.01025684, "balance_loss_clip": 1.04813159, "balance_loss_mlp": 1.01832891, "epoch": 0.6251427884326339, "flos": 23437856898720.0, "grad_norm": 1.7899983468576466, "language_loss": 0.69768238, "learning_rate": 1.3015029386052256e-06, "loss": 0.71938658, "num_input_tokens_seen": 111964095, "step": 5199, "time_per_iteration": 2.5603878498077393 }, { "auxiliary_loss_clip": 0.01143739, "auxiliary_loss_mlp": 0.01030814, "balance_loss_clip": 1.04948926, "balance_loss_mlp": 1.0230062, "epoch": 0.625263031323273, "flos": 31723060010400.0, "grad_norm": 1.8140636785343298, "language_loss": 0.72694403, "learning_rate": 1.3007730717007622e-06, "loss": 0.74868953, "num_input_tokens_seen": 111984910, "step": 5200, "time_per_iteration": 2.6243255138397217 }, { "auxiliary_loss_clip": 0.01178809, "auxiliary_loss_mlp": 0.01026529, "balance_loss_clip": 1.0521816, "balance_loss_mlp": 1.01879823, "epoch": 0.6253832742139122, "flos": 24134344959840.0, "grad_norm": 1.8746981500565223, "language_loss": 0.75304747, "learning_rate": 1.3000433108684676e-06, "loss": 0.77510083, "num_input_tokens_seen": 112005410, "step": 5201, "time_per_iteration": 2.504838228225708 }, { "auxiliary_loss_clip": 0.01161948, "auxiliary_loss_mlp": 0.01029355, "balance_loss_clip": 1.05196238, "balance_loss_mlp": 1.02194059, "epoch": 0.6255035171045512, "flos": 27668816163840.0, "grad_norm": 2.3934109712567553, "language_loss": 0.80403274, "learning_rate": 1.2993136562190467e-06, "loss": 0.82594579, "num_input_tokens_seen": 112024530, "step": 5202, "time_per_iteration": 2.546877861022949 }, { "auxiliary_loss_clip": 0.01154718, "auxiliary_loss_mlp": 0.01024397, "balance_loss_clip": 1.05042601, "balance_loss_mlp": 1.01742649, "epoch": 0.6256237599951903, "flos": 20227802016480.0, "grad_norm": 1.591554129261925, "language_loss": 0.706788, "learning_rate": 1.2985841078631871e-06, "loss": 0.72857916, "num_input_tokens_seen": 112043850, "step": 5203, "time_per_iteration": 2.5379087924957275 }, { "auxiliary_loss_clip": 0.01101137, "auxiliary_loss_mlp": 0.01028473, "balance_loss_clip": 1.04042804, "balance_loss_mlp": 1.02115977, "epoch": 0.6257440028858293, "flos": 24170578784160.0, "grad_norm": 1.7044080107337325, "language_loss": 0.78180844, "learning_rate": 1.2978546659115608e-06, "loss": 0.80310452, "num_input_tokens_seen": 112061930, "step": 5204, "time_per_iteration": 2.6476337909698486 }, { "auxiliary_loss_clip": 0.01152811, "auxiliary_loss_mlp": 0.0102758, "balance_loss_clip": 1.05133975, "balance_loss_mlp": 1.02067506, "epoch": 0.6258642457764685, "flos": 15851943371040.0, "grad_norm": 2.1745093874634027, "language_loss": 0.85310805, "learning_rate": 1.2971253304748228e-06, "loss": 0.8749119, "num_input_tokens_seen": 112079645, "step": 5205, "time_per_iteration": 2.515815258026123 }, { "auxiliary_loss_clip": 0.01170377, "auxiliary_loss_mlp": 0.01031123, "balance_loss_clip": 1.05455494, "balance_loss_mlp": 1.0233624, "epoch": 0.6259844886671075, "flos": 11911357537920.0, "grad_norm": 1.6923158813107855, "language_loss": 0.75069338, "learning_rate": 1.296396101663614e-06, "loss": 0.77270836, "num_input_tokens_seen": 112096205, "step": 5206, "time_per_iteration": 2.470533609390259 }, { "auxiliary_loss_clip": 0.01166587, "auxiliary_loss_mlp": 0.01027557, "balance_loss_clip": 1.05246067, "balance_loss_mlp": 1.020473, "epoch": 0.6261047315577466, "flos": 15887961693600.0, "grad_norm": 2.1727378696669777, "language_loss": 0.84065175, "learning_rate": 1.2956669795885565e-06, "loss": 0.86259317, "num_input_tokens_seen": 112112835, "step": 5207, "time_per_iteration": 2.499798536300659 }, { "auxiliary_loss_clip": 0.01128386, "auxiliary_loss_mlp": 0.0102781, "balance_loss_clip": 1.04819798, "balance_loss_mlp": 1.02011549, "epoch": 0.6262249744483858, "flos": 31248931435680.0, "grad_norm": 1.757449110871488, "language_loss": 0.68175459, "learning_rate": 1.294937964360259e-06, "loss": 0.70331657, "num_input_tokens_seen": 112133105, "step": 5208, "time_per_iteration": 2.723273277282715 }, { "auxiliary_loss_clip": 0.01154966, "auxiliary_loss_mlp": 0.0102956, "balance_loss_clip": 1.0495615, "balance_loss_mlp": 1.02153158, "epoch": 0.6263452173390248, "flos": 27198602537760.0, "grad_norm": 2.31758855450262, "language_loss": 0.70917773, "learning_rate": 1.2942090560893108e-06, "loss": 0.73102307, "num_input_tokens_seen": 112152510, "step": 5209, "time_per_iteration": 2.603559970855713 }, { "auxiliary_loss_clip": 0.01176999, "auxiliary_loss_mlp": 0.01032825, "balance_loss_clip": 1.05223131, "balance_loss_mlp": 1.02587855, "epoch": 0.6264654602296639, "flos": 37342074774240.0, "grad_norm": 2.0502376639886126, "language_loss": 0.60604799, "learning_rate": 1.2934802548862882e-06, "loss": 0.62814623, "num_input_tokens_seen": 112175295, "step": 5210, "time_per_iteration": 2.5988030433654785 }, { "auxiliary_loss_clip": 0.01145373, "auxiliary_loss_mlp": 0.01027829, "balance_loss_clip": 1.04805636, "balance_loss_mlp": 1.02032518, "epoch": 0.626585703120303, "flos": 14756955076320.0, "grad_norm": 1.9477054008714063, "language_loss": 0.82546961, "learning_rate": 1.292751560861749e-06, "loss": 0.84720159, "num_input_tokens_seen": 112190200, "step": 5211, "time_per_iteration": 3.246917247772217 }, { "auxiliary_loss_clip": 0.0118149, "auxiliary_loss_mlp": 0.01022869, "balance_loss_clip": 1.05345452, "balance_loss_mlp": 1.01510262, "epoch": 0.6267059460109421, "flos": 22347322307040.0, "grad_norm": 1.9191981126448672, "language_loss": 0.79669309, "learning_rate": 1.2920229741262354e-06, "loss": 0.81873667, "num_input_tokens_seen": 112208205, "step": 5212, "time_per_iteration": 3.2788259983062744 }, { "auxiliary_loss_clip": 0.01150501, "auxiliary_loss_mlp": 0.0102945, "balance_loss_clip": 1.04907846, "balance_loss_mlp": 1.02231264, "epoch": 0.6268261889015811, "flos": 17748816839520.0, "grad_norm": 2.1710967677833013, "language_loss": 0.75435269, "learning_rate": 1.2912944947902739e-06, "loss": 0.77615219, "num_input_tokens_seen": 112224690, "step": 5213, "time_per_iteration": 2.5390939712524414 }, { "auxiliary_loss_clip": 0.01154351, "auxiliary_loss_mlp": 0.01031939, "balance_loss_clip": 1.04880321, "balance_loss_mlp": 1.02426744, "epoch": 0.6269464317922203, "flos": 32846488149120.0, "grad_norm": 1.9891259608906549, "language_loss": 0.71394968, "learning_rate": 1.2905661229643742e-06, "loss": 0.7358126, "num_input_tokens_seen": 112244450, "step": 5214, "time_per_iteration": 2.639147996902466 }, { "auxiliary_loss_clip": 0.01180266, "auxiliary_loss_mlp": 0.01027498, "balance_loss_clip": 1.05263901, "balance_loss_mlp": 1.01971984, "epoch": 0.6270666746828594, "flos": 17929195788000.0, "grad_norm": 2.599875426515764, "language_loss": 0.84352654, "learning_rate": 1.2898378587590299e-06, "loss": 0.86560416, "num_input_tokens_seen": 112261050, "step": 5215, "time_per_iteration": 3.1369807720184326 }, { "auxiliary_loss_clip": 0.0115973, "auxiliary_loss_mlp": 0.01030403, "balance_loss_clip": 1.04971492, "balance_loss_mlp": 1.02351844, "epoch": 0.6271869175734984, "flos": 17457329981760.0, "grad_norm": 3.4974061255865583, "language_loss": 0.87227869, "learning_rate": 1.2891097022847173e-06, "loss": 0.89418006, "num_input_tokens_seen": 112278395, "step": 5216, "time_per_iteration": 2.4760818481445312 }, { "auxiliary_loss_clip": 0.01149765, "auxiliary_loss_mlp": 0.01028449, "balance_loss_clip": 1.049371, "balance_loss_mlp": 1.02033734, "epoch": 0.6273071604641376, "flos": 26868619087200.0, "grad_norm": 2.630545684314663, "language_loss": 0.66563654, "learning_rate": 1.2883816536518978e-06, "loss": 0.6874187, "num_input_tokens_seen": 112299535, "step": 5217, "time_per_iteration": 2.567777395248413 }, { "auxiliary_loss_clip": 0.01160293, "auxiliary_loss_mlp": 0.01023777, "balance_loss_clip": 1.05042541, "balance_loss_mlp": 1.01651692, "epoch": 0.6274274033547766, "flos": 26062387961280.0, "grad_norm": 1.8465773059833237, "language_loss": 0.81636125, "learning_rate": 1.2876537129710155e-06, "loss": 0.83820188, "num_input_tokens_seen": 112317265, "step": 5218, "time_per_iteration": 2.5493698120117188 }, { "auxiliary_loss_clip": 0.01148712, "auxiliary_loss_mlp": 0.01025275, "balance_loss_clip": 1.05340219, "balance_loss_mlp": 1.01682854, "epoch": 0.6275476462454157, "flos": 20266262692320.0, "grad_norm": 1.926692069881183, "language_loss": 0.75119591, "learning_rate": 1.286925880352499e-06, "loss": 0.77293575, "num_input_tokens_seen": 112336125, "step": 5219, "time_per_iteration": 2.5188045501708984 }, { "auxiliary_loss_clip": 0.0114587, "auxiliary_loss_mlp": 0.01028933, "balance_loss_clip": 1.04803693, "balance_loss_mlp": 1.02135086, "epoch": 0.6276678891360549, "flos": 26320406600640.0, "grad_norm": 1.6490234693479588, "language_loss": 0.71172237, "learning_rate": 1.2861981559067592e-06, "loss": 0.73347032, "num_input_tokens_seen": 112356730, "step": 5220, "time_per_iteration": 3.330314874649048 }, { "auxiliary_loss_clip": 0.0111355, "auxiliary_loss_mlp": 0.01023069, "balance_loss_clip": 1.04606843, "balance_loss_mlp": 1.01529074, "epoch": 0.6277881320266939, "flos": 13912514693280.0, "grad_norm": 2.218608921609596, "language_loss": 0.80283564, "learning_rate": 1.2854705397441917e-06, "loss": 0.82420182, "num_input_tokens_seen": 112372270, "step": 5221, "time_per_iteration": 2.5596272945404053 }, { "auxiliary_loss_clip": 0.01130994, "auxiliary_loss_mlp": 0.01029665, "balance_loss_clip": 1.04626203, "balance_loss_mlp": 1.02244115, "epoch": 0.627908374917333, "flos": 27048926201760.0, "grad_norm": 2.460302006717503, "language_loss": 0.77301145, "learning_rate": 1.2847430319751747e-06, "loss": 0.79461807, "num_input_tokens_seen": 112390365, "step": 5222, "time_per_iteration": 2.598839044570923 }, { "auxiliary_loss_clip": 0.01160235, "auxiliary_loss_mlp": 0.01025872, "balance_loss_clip": 1.05231404, "balance_loss_mlp": 1.0188055, "epoch": 0.6280286178079721, "flos": 23769205193760.0, "grad_norm": 2.501775406006134, "language_loss": 0.67132795, "learning_rate": 1.2840156327100712e-06, "loss": 0.69318902, "num_input_tokens_seen": 112407490, "step": 5223, "time_per_iteration": 2.4995522499084473 }, { "auxiliary_loss_clip": 0.01178756, "auxiliary_loss_mlp": 0.01023358, "balance_loss_clip": 1.05462027, "balance_loss_mlp": 1.01558018, "epoch": 0.6281488606986112, "flos": 26359154612160.0, "grad_norm": 2.2675949990112976, "language_loss": 0.72162837, "learning_rate": 1.2832883420592272e-06, "loss": 0.74364948, "num_input_tokens_seen": 112426385, "step": 5224, "time_per_iteration": 2.5190396308898926 }, { "auxiliary_loss_clip": 0.01145759, "auxiliary_loss_mlp": 0.01027393, "balance_loss_clip": 1.04850078, "balance_loss_mlp": 1.01980853, "epoch": 0.6282691035892503, "flos": 36137199746880.0, "grad_norm": 2.4906514202347534, "language_loss": 0.6421454, "learning_rate": 1.282561160132972e-06, "loss": 0.66387689, "num_input_tokens_seen": 112446905, "step": 5225, "time_per_iteration": 2.658783435821533 }, { "auxiliary_loss_clip": 0.0115263, "auxiliary_loss_mlp": 0.01026374, "balance_loss_clip": 1.04510689, "balance_loss_mlp": 1.01888728, "epoch": 0.6283893464798894, "flos": 26537234875200.0, "grad_norm": 1.9374448212694326, "language_loss": 0.80651212, "learning_rate": 1.2818340870416186e-06, "loss": 0.82830215, "num_input_tokens_seen": 112468040, "step": 5226, "time_per_iteration": 2.589906930923462 }, { "auxiliary_loss_clip": 0.01142777, "auxiliary_loss_mlp": 0.01027717, "balance_loss_clip": 1.04600632, "balance_loss_mlp": 1.02001619, "epoch": 0.6285095893705285, "flos": 22237220072640.0, "grad_norm": 1.7604256800663722, "language_loss": 0.75584388, "learning_rate": 1.2811071228954626e-06, "loss": 0.77754879, "num_input_tokens_seen": 112486675, "step": 5227, "time_per_iteration": 2.589961051940918 }, { "auxiliary_loss_clip": 0.01149766, "auxiliary_loss_mlp": 0.01026125, "balance_loss_clip": 1.05118585, "balance_loss_mlp": 1.01930571, "epoch": 0.6286298322611675, "flos": 26542263249600.0, "grad_norm": 1.7959231121515598, "language_loss": 0.81152028, "learning_rate": 1.2803802678047846e-06, "loss": 0.83327919, "num_input_tokens_seen": 112506825, "step": 5228, "time_per_iteration": 2.572219133377075 }, { "auxiliary_loss_clip": 0.01155769, "auxiliary_loss_mlp": 0.01036165, "balance_loss_clip": 1.0516448, "balance_loss_mlp": 1.02752233, "epoch": 0.6287500751518067, "flos": 21795230788800.0, "grad_norm": 2.3646768279226076, "language_loss": 0.74141711, "learning_rate": 1.279653521879848e-06, "loss": 0.76333642, "num_input_tokens_seen": 112526890, "step": 5229, "time_per_iteration": 2.55823016166687 }, { "auxiliary_loss_clip": 0.0108241, "auxiliary_loss_mlp": 0.01027879, "balance_loss_clip": 1.03899908, "balance_loss_mlp": 1.02043486, "epoch": 0.6288703180424458, "flos": 20009608897440.0, "grad_norm": 1.91524855586513, "language_loss": 0.84224534, "learning_rate": 1.2789268852308997e-06, "loss": 0.86334825, "num_input_tokens_seen": 112542100, "step": 5230, "time_per_iteration": 2.619903326034546 }, { "auxiliary_loss_clip": 0.01158039, "auxiliary_loss_mlp": 0.01023368, "balance_loss_clip": 1.05072773, "balance_loss_mlp": 1.01616156, "epoch": 0.6289905609330848, "flos": 22124926903680.0, "grad_norm": 1.7644468623306957, "language_loss": 0.70458382, "learning_rate": 1.2782003579681688e-06, "loss": 0.72639787, "num_input_tokens_seen": 112561630, "step": 5231, "time_per_iteration": 2.5347821712493896 }, { "auxiliary_loss_clip": 0.01178148, "auxiliary_loss_mlp": 0.01027479, "balance_loss_clip": 1.05171788, "balance_loss_mlp": 1.02017736, "epoch": 0.629110803823724, "flos": 25518485509920.0, "grad_norm": 1.6670015573811097, "language_loss": 0.74301928, "learning_rate": 1.2774739402018701e-06, "loss": 0.76507556, "num_input_tokens_seen": 112582465, "step": 5232, "time_per_iteration": 2.5097248554229736 }, { "auxiliary_loss_clip": 0.01164262, "auxiliary_loss_mlp": 0.0102978, "balance_loss_clip": 1.05229759, "balance_loss_mlp": 1.02211738, "epoch": 0.629231046714363, "flos": 20886619574880.0, "grad_norm": 1.5937710405573893, "language_loss": 0.72946674, "learning_rate": 1.2767476320422002e-06, "loss": 0.75140715, "num_input_tokens_seen": 112602390, "step": 5233, "time_per_iteration": 2.5181171894073486 }, { "auxiliary_loss_clip": 0.0104489, "auxiliary_loss_mlp": 0.01005765, "balance_loss_clip": 1.01218486, "balance_loss_mlp": 1.00457883, "epoch": 0.6293512896050021, "flos": 65050040424480.0, "grad_norm": 0.6776321309424007, "language_loss": 0.57259202, "learning_rate": 1.2760214335993392e-06, "loss": 0.59309864, "num_input_tokens_seen": 112669035, "step": 5234, "time_per_iteration": 3.2167468070983887 }, { "auxiliary_loss_clip": 0.0115305, "auxiliary_loss_mlp": 0.01025649, "balance_loss_clip": 1.04665184, "balance_loss_mlp": 1.01846123, "epoch": 0.6294715324956413, "flos": 34677861859200.0, "grad_norm": 1.9238096912359124, "language_loss": 0.58633947, "learning_rate": 1.2752953449834514e-06, "loss": 0.60812652, "num_input_tokens_seen": 112691485, "step": 5235, "time_per_iteration": 2.6178977489471436 }, { "auxiliary_loss_clip": 0.01174837, "auxiliary_loss_mlp": 0.01029393, "balance_loss_clip": 1.05063593, "balance_loss_mlp": 1.0224905, "epoch": 0.6295917753862803, "flos": 22784211382560.0, "grad_norm": 1.611881884534392, "language_loss": 0.80402946, "learning_rate": 1.2745693663046836e-06, "loss": 0.82607174, "num_input_tokens_seen": 112710555, "step": 5236, "time_per_iteration": 2.4964325428009033 }, { "auxiliary_loss_clip": 0.01156965, "auxiliary_loss_mlp": 0.01031541, "balance_loss_clip": 1.04862559, "balance_loss_mlp": 1.02463555, "epoch": 0.6297120182769194, "flos": 20850457584480.0, "grad_norm": 1.7352367778899944, "language_loss": 0.80456531, "learning_rate": 1.2738434976731662e-06, "loss": 0.82645035, "num_input_tokens_seen": 112728740, "step": 5237, "time_per_iteration": 3.225046157836914 }, { "auxiliary_loss_clip": 0.01151401, "auxiliary_loss_mlp": 0.0102441, "balance_loss_clip": 1.05140328, "balance_loss_mlp": 1.01681376, "epoch": 0.6298322611675584, "flos": 19497666152160.0, "grad_norm": 2.148960416963303, "language_loss": 0.75088018, "learning_rate": 1.2731177391990125e-06, "loss": 0.77263832, "num_input_tokens_seen": 112748665, "step": 5238, "time_per_iteration": 3.2886016368865967 }, { "auxiliary_loss_clip": 0.01149411, "auxiliary_loss_mlp": 0.01025138, "balance_loss_clip": 1.04664087, "balance_loss_mlp": 1.017851, "epoch": 0.6299525040581976, "flos": 12604469404800.0, "grad_norm": 1.9569137662414284, "language_loss": 0.81513983, "learning_rate": 1.2723920909923203e-06, "loss": 0.83688533, "num_input_tokens_seen": 112764410, "step": 5239, "time_per_iteration": 2.494974136352539 }, { "auxiliary_loss_clip": 0.01071865, "auxiliary_loss_mlp": 0.01004804, "balance_loss_clip": 1.01157212, "balance_loss_mlp": 1.00349891, "epoch": 0.6300727469488366, "flos": 57725694926880.0, "grad_norm": 0.8618859025849783, "language_loss": 0.6046387, "learning_rate": 1.2716665531631688e-06, "loss": 0.62540543, "num_input_tokens_seen": 112818695, "step": 5240, "time_per_iteration": 3.020927906036377 }, { "auxiliary_loss_clip": 0.01164607, "auxiliary_loss_mlp": 0.01030359, "balance_loss_clip": 1.05056059, "balance_loss_mlp": 1.0228188, "epoch": 0.6301929898394757, "flos": 22527306168960.0, "grad_norm": 1.801638098364895, "language_loss": 0.76931655, "learning_rate": 1.270941125821623e-06, "loss": 0.79126626, "num_input_tokens_seen": 112839120, "step": 5241, "time_per_iteration": 3.211376667022705 }, { "auxiliary_loss_clip": 0.01156721, "auxiliary_loss_mlp": 0.0102533, "balance_loss_clip": 1.04577243, "balance_loss_mlp": 1.01801682, "epoch": 0.6303132327301149, "flos": 28293554915520.0, "grad_norm": 1.5915385877694863, "language_loss": 0.75438833, "learning_rate": 1.2702158090777278e-06, "loss": 0.77620888, "num_input_tokens_seen": 112860210, "step": 5242, "time_per_iteration": 2.5556352138519287 }, { "auxiliary_loss_clip": 0.01128802, "auxiliary_loss_mlp": 0.01028441, "balance_loss_clip": 1.04498363, "balance_loss_mlp": 1.02110314, "epoch": 0.6304334756207539, "flos": 25264525487040.0, "grad_norm": 1.840096735881501, "language_loss": 0.74759001, "learning_rate": 1.2694906030415148e-06, "loss": 0.76916248, "num_input_tokens_seen": 112877955, "step": 5243, "time_per_iteration": 2.598081350326538 }, { "auxiliary_loss_clip": 0.0115604, "auxiliary_loss_mlp": 0.01025533, "balance_loss_clip": 1.04841757, "balance_loss_mlp": 1.01769841, "epoch": 0.630553718511393, "flos": 18033551308800.0, "grad_norm": 3.878796607105861, "language_loss": 0.82306576, "learning_rate": 1.2687655078229958e-06, "loss": 0.84488153, "num_input_tokens_seen": 112892285, "step": 5244, "time_per_iteration": 2.5089733600616455 }, { "auxiliary_loss_clip": 0.01148544, "auxiliary_loss_mlp": 0.01029804, "balance_loss_clip": 1.05101728, "balance_loss_mlp": 1.02226138, "epoch": 0.6306739614020321, "flos": 27304107401280.0, "grad_norm": 2.032471578735141, "language_loss": 0.69371057, "learning_rate": 1.2680405235321678e-06, "loss": 0.71549404, "num_input_tokens_seen": 112913620, "step": 5245, "time_per_iteration": 2.573258876800537 }, { "auxiliary_loss_clip": 0.01151945, "auxiliary_loss_mlp": 0.00762603, "balance_loss_clip": 1.05174518, "balance_loss_mlp": 1.00069082, "epoch": 0.6307942042926712, "flos": 15341437304160.0, "grad_norm": 4.155155162740504, "language_loss": 0.78729606, "learning_rate": 1.267315650279011e-06, "loss": 0.80644155, "num_input_tokens_seen": 112932090, "step": 5246, "time_per_iteration": 3.285975217819214 }, { "auxiliary_loss_clip": 0.0112663, "auxiliary_loss_mlp": 0.01024034, "balance_loss_clip": 1.04696107, "balance_loss_mlp": 1.01686954, "epoch": 0.6309144471833102, "flos": 19606403542080.0, "grad_norm": 1.8375013655987655, "language_loss": 0.73776472, "learning_rate": 1.2665908881734874e-06, "loss": 0.75927138, "num_input_tokens_seen": 112950925, "step": 5247, "time_per_iteration": 2.5375051498413086 }, { "auxiliary_loss_clip": 0.0116429, "auxiliary_loss_mlp": 0.01031108, "balance_loss_clip": 1.05199695, "balance_loss_mlp": 1.02395535, "epoch": 0.6310346900739494, "flos": 17493348304320.0, "grad_norm": 2.0536200964812092, "language_loss": 0.84617853, "learning_rate": 1.2658662373255432e-06, "loss": 0.86813253, "num_input_tokens_seen": 112969315, "step": 5248, "time_per_iteration": 2.4848875999450684 }, { "auxiliary_loss_clip": 0.01053666, "auxiliary_loss_mlp": 0.01002376, "balance_loss_clip": 1.01316345, "balance_loss_mlp": 1.00106478, "epoch": 0.6311549329645885, "flos": 55070173916160.0, "grad_norm": 0.7105484093376471, "language_loss": 0.52318627, "learning_rate": 1.2651416978451063e-06, "loss": 0.54374671, "num_input_tokens_seen": 113034700, "step": 5249, "time_per_iteration": 3.2228870391845703 }, { "auxiliary_loss_clip": 0.01181326, "auxiliary_loss_mlp": 0.01029253, "balance_loss_clip": 1.05291438, "balance_loss_mlp": 1.02095914, "epoch": 0.6312751758552275, "flos": 41902550569440.0, "grad_norm": 1.7543072043005525, "language_loss": 0.65120554, "learning_rate": 1.2644172698420903e-06, "loss": 0.67331129, "num_input_tokens_seen": 113056805, "step": 5250, "time_per_iteration": 2.647573947906494 }, { "auxiliary_loss_clip": 0.01137336, "auxiliary_loss_mlp": 0.01036107, "balance_loss_clip": 1.04807127, "balance_loss_mlp": 1.02858472, "epoch": 0.6313954187458667, "flos": 19646803733760.0, "grad_norm": 1.8137872253572562, "language_loss": 0.84776747, "learning_rate": 1.2636929534263892e-06, "loss": 0.86950183, "num_input_tokens_seen": 113075790, "step": 5251, "time_per_iteration": 2.566572666168213 }, { "auxiliary_loss_clip": 0.01135691, "auxiliary_loss_mlp": 0.01021666, "balance_loss_clip": 1.0440042, "balance_loss_mlp": 1.01429331, "epoch": 0.6315156616365057, "flos": 22894277700000.0, "grad_norm": 2.6834017578239653, "language_loss": 0.77628279, "learning_rate": 1.2629687487078821e-06, "loss": 0.79785633, "num_input_tokens_seen": 113094600, "step": 5252, "time_per_iteration": 2.577584743499756 }, { "auxiliary_loss_clip": 0.0116377, "auxiliary_loss_mlp": 0.01026474, "balance_loss_clip": 1.04718292, "balance_loss_mlp": 1.01857662, "epoch": 0.6316359045271448, "flos": 23726254897920.0, "grad_norm": 1.9512168930791636, "language_loss": 0.76056516, "learning_rate": 1.2622446557964293e-06, "loss": 0.7824676, "num_input_tokens_seen": 113112605, "step": 5253, "time_per_iteration": 2.5350940227508545 }, { "auxiliary_loss_clip": 0.01145444, "auxiliary_loss_mlp": 0.01027919, "balance_loss_clip": 1.04349613, "balance_loss_mlp": 1.02076626, "epoch": 0.631756147417784, "flos": 33108421737120.0, "grad_norm": 1.8358475808131292, "language_loss": 0.71279341, "learning_rate": 1.261520674801876e-06, "loss": 0.73452699, "num_input_tokens_seen": 113133200, "step": 5254, "time_per_iteration": 2.624713659286499 }, { "auxiliary_loss_clip": 0.0114837, "auxiliary_loss_mlp": 0.01027255, "balance_loss_clip": 1.05166769, "balance_loss_mlp": 1.01969683, "epoch": 0.631876390308423, "flos": 31248428598240.0, "grad_norm": 1.9552673899459945, "language_loss": 0.72459596, "learning_rate": 1.2607968058340488e-06, "loss": 0.7463522, "num_input_tokens_seen": 113152895, "step": 5255, "time_per_iteration": 2.624894142150879 }, { "auxiliary_loss_clip": 0.01145674, "auxiliary_loss_mlp": 0.01030602, "balance_loss_clip": 1.0477823, "balance_loss_mlp": 1.02333641, "epoch": 0.6319966331990621, "flos": 24681156684960.0, "grad_norm": 1.7113097198498954, "language_loss": 0.7311787, "learning_rate": 1.2600730490027583e-06, "loss": 0.75294149, "num_input_tokens_seen": 113173135, "step": 5256, "time_per_iteration": 2.561002254486084 }, { "auxiliary_loss_clip": 0.01133238, "auxiliary_loss_mlp": 0.01025781, "balance_loss_clip": 1.04599893, "balance_loss_mlp": 1.01807141, "epoch": 0.6321168760897012, "flos": 17491767958080.0, "grad_norm": 1.5793546184436051, "language_loss": 0.80501199, "learning_rate": 1.2593494044177984e-06, "loss": 0.82660222, "num_input_tokens_seen": 113191440, "step": 5257, "time_per_iteration": 2.584794759750366 }, { "auxiliary_loss_clip": 0.01180692, "auxiliary_loss_mlp": 0.01028004, "balance_loss_clip": 1.05068874, "balance_loss_mlp": 1.01917672, "epoch": 0.6322371189803403, "flos": 18295377145920.0, "grad_norm": 2.3211168730518743, "language_loss": 0.8018747, "learning_rate": 1.2586258721889448e-06, "loss": 0.82396168, "num_input_tokens_seen": 113208790, "step": 5258, "time_per_iteration": 2.4354374408721924 }, { "auxiliary_loss_clip": 0.01114255, "auxiliary_loss_mlp": 0.01026329, "balance_loss_clip": 1.04773235, "balance_loss_mlp": 1.01866663, "epoch": 0.6323573618709794, "flos": 20157273883680.0, "grad_norm": 1.7648799388307517, "language_loss": 0.8137815, "learning_rate": 1.2579024524259573e-06, "loss": 0.83518732, "num_input_tokens_seen": 113225050, "step": 5259, "time_per_iteration": 2.587308883666992 }, { "auxiliary_loss_clip": 0.01143259, "auxiliary_loss_mlp": 0.01036155, "balance_loss_clip": 1.04446113, "balance_loss_mlp": 1.0286684, "epoch": 0.6324776047616185, "flos": 20042394693600.0, "grad_norm": 1.8895756789965423, "language_loss": 0.91174054, "learning_rate": 1.2571791452385768e-06, "loss": 0.93353468, "num_input_tokens_seen": 113242315, "step": 5260, "time_per_iteration": 2.517681837081909 }, { "auxiliary_loss_clip": 0.01149193, "auxiliary_loss_mlp": 0.01030957, "balance_loss_clip": 1.04999304, "balance_loss_mlp": 1.02330709, "epoch": 0.6325978476522576, "flos": 30848240267520.0, "grad_norm": 1.5334346507739847, "language_loss": 0.76961708, "learning_rate": 1.2564559507365301e-06, "loss": 0.79141855, "num_input_tokens_seen": 113264720, "step": 5261, "time_per_iteration": 2.605372667312622 }, { "auxiliary_loss_clip": 0.01150633, "auxiliary_loss_mlp": 0.01029447, "balance_loss_clip": 1.04917479, "balance_loss_mlp": 1.02167153, "epoch": 0.6327180905428966, "flos": 24535107961920.0, "grad_norm": 4.594066930413629, "language_loss": 0.78935802, "learning_rate": 1.2557328690295244e-06, "loss": 0.81115878, "num_input_tokens_seen": 113282910, "step": 5262, "time_per_iteration": 2.5467851161956787 }, { "auxiliary_loss_clip": 0.01134924, "auxiliary_loss_mlp": 0.01027266, "balance_loss_clip": 1.04752183, "balance_loss_mlp": 1.01994693, "epoch": 0.6328383334335358, "flos": 21575277738720.0, "grad_norm": 1.8197347681557725, "language_loss": 0.76108855, "learning_rate": 1.255009900227251e-06, "loss": 0.78271043, "num_input_tokens_seen": 113301935, "step": 5263, "time_per_iteration": 3.387404203414917 }, { "auxiliary_loss_clip": 0.01172894, "auxiliary_loss_mlp": 0.01029963, "balance_loss_clip": 1.05107296, "balance_loss_mlp": 1.02296209, "epoch": 0.6329585763241748, "flos": 22929865019040.0, "grad_norm": 1.8018275325463196, "language_loss": 0.79066706, "learning_rate": 1.254287044439383e-06, "loss": 0.81269562, "num_input_tokens_seen": 113321540, "step": 5264, "time_per_iteration": 3.2653117179870605 }, { "auxiliary_loss_clip": 0.01070781, "auxiliary_loss_mlp": 0.01002493, "balance_loss_clip": 1.01078939, "balance_loss_mlp": 1.0011456, "epoch": 0.6330788192148139, "flos": 70936908686400.0, "grad_norm": 0.7682545116031633, "language_loss": 0.54449582, "learning_rate": 1.2535643017755776e-06, "loss": 0.56522846, "num_input_tokens_seen": 113383730, "step": 5265, "time_per_iteration": 3.139801263809204 }, { "auxiliary_loss_clip": 0.01129777, "auxiliary_loss_mlp": 0.01030158, "balance_loss_clip": 1.04506087, "balance_loss_mlp": 1.02248073, "epoch": 0.6331990621054531, "flos": 21244504115040.0, "grad_norm": 2.8600488183238686, "language_loss": 0.7210623, "learning_rate": 1.2528416723454737e-06, "loss": 0.74266165, "num_input_tokens_seen": 113400400, "step": 5266, "time_per_iteration": 2.5706799030303955 }, { "auxiliary_loss_clip": 0.01174154, "auxiliary_loss_mlp": 0.01024693, "balance_loss_clip": 1.05169237, "balance_loss_mlp": 1.01829791, "epoch": 0.6333193049960921, "flos": 34459417321440.0, "grad_norm": 1.4703154414758728, "language_loss": 0.71098787, "learning_rate": 1.2521191562586945e-06, "loss": 0.73297638, "num_input_tokens_seen": 113424050, "step": 5267, "time_per_iteration": 3.3181700706481934 }, { "auxiliary_loss_clip": 0.01175406, "auxiliary_loss_mlp": 0.0076259, "balance_loss_clip": 1.05139947, "balance_loss_mlp": 1.00063872, "epoch": 0.6334395478867312, "flos": 18329886956160.0, "grad_norm": 2.391187026714124, "language_loss": 0.76751614, "learning_rate": 1.2513967536248445e-06, "loss": 0.78689611, "num_input_tokens_seen": 113440370, "step": 5268, "time_per_iteration": 2.4536430835723877 }, { "auxiliary_loss_clip": 0.01159577, "auxiliary_loss_mlp": 0.01029067, "balance_loss_clip": 1.05200219, "balance_loss_mlp": 1.02141345, "epoch": 0.6335597907773702, "flos": 23623156470720.0, "grad_norm": 1.611004186213366, "language_loss": 0.81188238, "learning_rate": 1.2506744645535117e-06, "loss": 0.83376884, "num_input_tokens_seen": 113460800, "step": 5269, "time_per_iteration": 2.5217185020446777 }, { "auxiliary_loss_clip": 0.01138263, "auxiliary_loss_mlp": 0.01023034, "balance_loss_clip": 1.04200649, "balance_loss_mlp": 1.01508284, "epoch": 0.6336800336680094, "flos": 22710917643840.0, "grad_norm": 1.8372250786463826, "language_loss": 0.6047672, "learning_rate": 1.249952289154267e-06, "loss": 0.62638021, "num_input_tokens_seen": 113480840, "step": 5270, "time_per_iteration": 2.5473690032958984 }, { "auxiliary_loss_clip": 0.01089897, "auxiliary_loss_mlp": 0.01025932, "balance_loss_clip": 1.03993142, "balance_loss_mlp": 1.01939297, "epoch": 0.6338002765586485, "flos": 23622761384160.0, "grad_norm": 1.8411085285033522, "language_loss": 0.76422226, "learning_rate": 1.2492302275366635e-06, "loss": 0.7853806, "num_input_tokens_seen": 113500515, "step": 5271, "time_per_iteration": 2.6533308029174805 }, { "auxiliary_loss_clip": 0.01155177, "auxiliary_loss_mlp": 0.01027218, "balance_loss_clip": 1.0470624, "balance_loss_mlp": 1.01940334, "epoch": 0.6339205194492875, "flos": 26505454753920.0, "grad_norm": 2.339529442320665, "language_loss": 0.65301394, "learning_rate": 1.2485082798102377e-06, "loss": 0.67483795, "num_input_tokens_seen": 113520930, "step": 5272, "time_per_iteration": 2.541137933731079 }, { "auxiliary_loss_clip": 0.0114027, "auxiliary_loss_mlp": 0.01029852, "balance_loss_clip": 1.04558706, "balance_loss_mlp": 1.02228177, "epoch": 0.6340407623399267, "flos": 18544308794400.0, "grad_norm": 2.022747512362297, "language_loss": 0.68609059, "learning_rate": 1.2477864460845084e-06, "loss": 0.70779181, "num_input_tokens_seen": 113537330, "step": 5273, "time_per_iteration": 3.29921555519104 }, { "auxiliary_loss_clip": 0.01146093, "auxiliary_loss_mlp": 0.01027206, "balance_loss_clip": 1.04740143, "balance_loss_mlp": 1.01942158, "epoch": 0.6341610052305657, "flos": 17712583015200.0, "grad_norm": 2.7002032941542864, "language_loss": 0.73417264, "learning_rate": 1.2470647264689776e-06, "loss": 0.75590563, "num_input_tokens_seen": 113555810, "step": 5274, "time_per_iteration": 2.5055294036865234 }, { "auxiliary_loss_clip": 0.01112267, "auxiliary_loss_mlp": 0.01027264, "balance_loss_clip": 1.04179263, "balance_loss_mlp": 1.01995611, "epoch": 0.6342812481212048, "flos": 23587030397280.0, "grad_norm": 1.8792464676290725, "language_loss": 0.71253061, "learning_rate": 1.2463431210731282e-06, "loss": 0.73392594, "num_input_tokens_seen": 113575395, "step": 5275, "time_per_iteration": 2.664672613143921 }, { "auxiliary_loss_clip": 0.0112486, "auxiliary_loss_mlp": 0.01027339, "balance_loss_clip": 1.0432061, "balance_loss_mlp": 1.01967359, "epoch": 0.634401491011844, "flos": 17821930993440.0, "grad_norm": 2.420173433401722, "language_loss": 0.75785172, "learning_rate": 1.2456216300064289e-06, "loss": 0.77937371, "num_input_tokens_seen": 113592945, "step": 5276, "time_per_iteration": 2.562534809112549 }, { "auxiliary_loss_clip": 0.0114224, "auxiliary_loss_mlp": 0.0102574, "balance_loss_clip": 1.04695797, "balance_loss_mlp": 1.01779175, "epoch": 0.634521733902483, "flos": 21358162128480.0, "grad_norm": 1.596929809533522, "language_loss": 0.78068477, "learning_rate": 1.244900253378328e-06, "loss": 0.80236459, "num_input_tokens_seen": 113613000, "step": 5277, "time_per_iteration": 2.5891873836517334 }, { "auxiliary_loss_clip": 0.01075752, "auxiliary_loss_mlp": 0.0102583, "balance_loss_clip": 1.04309118, "balance_loss_mlp": 1.01902914, "epoch": 0.6346419767931221, "flos": 16545055237920.0, "grad_norm": 2.4980979484541694, "language_loss": 0.68986589, "learning_rate": 1.2441789912982583e-06, "loss": 0.71088171, "num_input_tokens_seen": 113630085, "step": 5278, "time_per_iteration": 2.79818058013916 }, { "auxiliary_loss_clip": 0.01166819, "auxiliary_loss_mlp": 0.01030945, "balance_loss_clip": 1.05162549, "balance_loss_mlp": 1.02268052, "epoch": 0.6347622196837612, "flos": 24350993649600.0, "grad_norm": 1.8868492218589792, "language_loss": 0.64658219, "learning_rate": 1.2434578438756346e-06, "loss": 0.66855979, "num_input_tokens_seen": 113650515, "step": 5279, "time_per_iteration": 2.8151473999023438 }, { "auxiliary_loss_clip": 0.01163634, "auxiliary_loss_mlp": 0.0102302, "balance_loss_clip": 1.04853678, "balance_loss_mlp": 1.01621044, "epoch": 0.6348824625744003, "flos": 64523185752480.0, "grad_norm": 1.8111891593825085, "language_loss": 0.78101397, "learning_rate": 1.242736811219855e-06, "loss": 0.80288053, "num_input_tokens_seen": 113676475, "step": 5280, "time_per_iteration": 2.901911973953247 }, { "auxiliary_loss_clip": 0.01156649, "auxiliary_loss_mlp": 0.01032508, "balance_loss_clip": 1.04906619, "balance_loss_mlp": 1.02477407, "epoch": 0.6350027054650393, "flos": 28622137604640.0, "grad_norm": 1.8861163770961078, "language_loss": 0.81954676, "learning_rate": 1.2420158934402988e-06, "loss": 0.84143829, "num_input_tokens_seen": 113697090, "step": 5281, "time_per_iteration": 2.5703442096710205 }, { "auxiliary_loss_clip": 0.01119214, "auxiliary_loss_mlp": 0.01023114, "balance_loss_clip": 1.04190552, "balance_loss_mlp": 1.01555645, "epoch": 0.6351229483556785, "flos": 23002548169440.0, "grad_norm": 1.9507554331310126, "language_loss": 0.84509522, "learning_rate": 1.2412950906463286e-06, "loss": 0.86651844, "num_input_tokens_seen": 113714395, "step": 5282, "time_per_iteration": 2.5554540157318115 }, { "auxiliary_loss_clip": 0.01118814, "auxiliary_loss_mlp": 0.01024276, "balance_loss_clip": 1.04692173, "balance_loss_mlp": 1.01773453, "epoch": 0.6352431912463176, "flos": 21939304079040.0, "grad_norm": 1.9469566691482114, "language_loss": 0.89805305, "learning_rate": 1.2405744029472902e-06, "loss": 0.91948402, "num_input_tokens_seen": 113733880, "step": 5283, "time_per_iteration": 2.635162115097046 }, { "auxiliary_loss_clip": 0.0114641, "auxiliary_loss_mlp": 0.01023968, "balance_loss_clip": 1.04815769, "balance_loss_mlp": 1.01694608, "epoch": 0.6353634341369566, "flos": 13735260520320.0, "grad_norm": 2.506286777282064, "language_loss": 0.75818646, "learning_rate": 1.2398538304525108e-06, "loss": 0.7798903, "num_input_tokens_seen": 113752505, "step": 5284, "time_per_iteration": 2.5208218097686768 }, { "auxiliary_loss_clip": 0.01129231, "auxiliary_loss_mlp": 0.01025344, "balance_loss_clip": 1.0469588, "balance_loss_mlp": 1.01685011, "epoch": 0.6354836770275958, "flos": 19316173777920.0, "grad_norm": 2.04727886898862, "language_loss": 0.75547934, "learning_rate": 1.2391333732713016e-06, "loss": 0.7770251, "num_input_tokens_seen": 113770310, "step": 5285, "time_per_iteration": 2.528259754180908 }, { "auxiliary_loss_clip": 0.0113377, "auxiliary_loss_mlp": 0.01027196, "balance_loss_clip": 1.04475701, "balance_loss_mlp": 1.01933455, "epoch": 0.6356039199182348, "flos": 21613379244960.0, "grad_norm": 2.0382881432428723, "language_loss": 0.78572953, "learning_rate": 1.2384130315129543e-06, "loss": 0.80733919, "num_input_tokens_seen": 113788635, "step": 5286, "time_per_iteration": 2.587996244430542 }, { "auxiliary_loss_clip": 0.01073189, "auxiliary_loss_mlp": 0.01026841, "balance_loss_clip": 1.04069018, "balance_loss_mlp": 1.0195514, "epoch": 0.6357241628088739, "flos": 18111981172800.0, "grad_norm": 3.2535124489883285, "language_loss": 0.7364189, "learning_rate": 1.2376928052867447e-06, "loss": 0.75741917, "num_input_tokens_seen": 113807755, "step": 5287, "time_per_iteration": 2.867892265319824 }, { "auxiliary_loss_clip": 0.01148888, "auxiliary_loss_mlp": 0.01031367, "balance_loss_clip": 1.05049479, "balance_loss_mlp": 1.02384508, "epoch": 0.6358444056995131, "flos": 24935260375680.0, "grad_norm": 2.141422078144442, "language_loss": 0.77814603, "learning_rate": 1.2369726947019299e-06, "loss": 0.79994857, "num_input_tokens_seen": 113828230, "step": 5288, "time_per_iteration": 2.9394240379333496 }, { "auxiliary_loss_clip": 0.01159899, "auxiliary_loss_mlp": 0.01028437, "balance_loss_clip": 1.04767191, "balance_loss_mlp": 1.02160931, "epoch": 0.6359646485901521, "flos": 23293352604960.0, "grad_norm": 2.3097675011892034, "language_loss": 0.66995585, "learning_rate": 1.2362526998677511e-06, "loss": 0.69183922, "num_input_tokens_seen": 113844595, "step": 5289, "time_per_iteration": 3.6559786796569824 }, { "auxiliary_loss_clip": 0.01151375, "auxiliary_loss_mlp": 0.01028131, "balance_loss_clip": 1.04772854, "balance_loss_mlp": 1.02092171, "epoch": 0.6360848914807912, "flos": 20887445664960.0, "grad_norm": 2.7241556902157247, "language_loss": 0.84211588, "learning_rate": 1.2355328208934301e-06, "loss": 0.86391085, "num_input_tokens_seen": 113863470, "step": 5290, "time_per_iteration": 3.605844020843506 }, { "auxiliary_loss_clip": 0.01162154, "auxiliary_loss_mlp": 0.00762511, "balance_loss_clip": 1.04798889, "balance_loss_mlp": 1.00067508, "epoch": 0.6362051343714303, "flos": 18479778793920.0, "grad_norm": 1.8918673723838697, "language_loss": 0.72200775, "learning_rate": 1.2348130578881728e-06, "loss": 0.74125439, "num_input_tokens_seen": 113881690, "step": 5291, "time_per_iteration": 2.54237961769104 }, { "auxiliary_loss_clip": 0.01179967, "auxiliary_loss_mlp": 0.01025111, "balance_loss_clip": 1.05267119, "balance_loss_mlp": 1.01728225, "epoch": 0.6363253772620694, "flos": 24389597993280.0, "grad_norm": 2.0661770109258586, "language_loss": 0.76080894, "learning_rate": 1.2340934109611664e-06, "loss": 0.78285968, "num_input_tokens_seen": 113902450, "step": 5292, "time_per_iteration": 2.4958596229553223 }, { "auxiliary_loss_clip": 0.01154302, "auxiliary_loss_mlp": 0.0102656, "balance_loss_clip": 1.04956365, "balance_loss_mlp": 1.01840043, "epoch": 0.6364456201527084, "flos": 25958247942240.0, "grad_norm": 2.0536577632868864, "language_loss": 0.68789524, "learning_rate": 1.2333738802215798e-06, "loss": 0.70970386, "num_input_tokens_seen": 113922670, "step": 5293, "time_per_iteration": 3.315990686416626 }, { "auxiliary_loss_clip": 0.01112449, "auxiliary_loss_mlp": 0.01028504, "balance_loss_clip": 1.04165602, "balance_loss_mlp": 1.02115452, "epoch": 0.6365658630433476, "flos": 20740714519680.0, "grad_norm": 1.9846672755281982, "language_loss": 0.80993211, "learning_rate": 1.2326544657785668e-06, "loss": 0.83134156, "num_input_tokens_seen": 113942360, "step": 5294, "time_per_iteration": 2.607780694961548 }, { "auxiliary_loss_clip": 0.01123015, "auxiliary_loss_mlp": 0.01031972, "balance_loss_clip": 1.04307723, "balance_loss_mlp": 1.0249567, "epoch": 0.6366861059339867, "flos": 21434149639200.0, "grad_norm": 2.728707457152839, "language_loss": 0.74556327, "learning_rate": 1.2319351677412608e-06, "loss": 0.76711321, "num_input_tokens_seen": 113959405, "step": 5295, "time_per_iteration": 2.564258575439453 }, { "auxiliary_loss_clip": 0.01145018, "auxiliary_loss_mlp": 0.01030348, "balance_loss_clip": 1.05021667, "balance_loss_mlp": 1.02297807, "epoch": 0.6368063488246257, "flos": 22267096595040.0, "grad_norm": 1.8127008044095188, "language_loss": 0.73883915, "learning_rate": 1.2312159862187796e-06, "loss": 0.76059282, "num_input_tokens_seen": 113977815, "step": 5296, "time_per_iteration": 2.5519468784332275 }, { "auxiliary_loss_clip": 0.01181307, "auxiliary_loss_mlp": 0.01028123, "balance_loss_clip": 1.05315375, "balance_loss_mlp": 1.0203743, "epoch": 0.6369265917152649, "flos": 22420723796640.0, "grad_norm": 1.6551545706783188, "language_loss": 0.75959027, "learning_rate": 1.2304969213202217e-06, "loss": 0.78168458, "num_input_tokens_seen": 113999075, "step": 5297, "time_per_iteration": 2.499636650085449 }, { "auxiliary_loss_clip": 0.01142463, "auxiliary_loss_mlp": 0.01027476, "balance_loss_clip": 1.04671311, "balance_loss_mlp": 1.02011466, "epoch": 0.6370468346059039, "flos": 24718180682400.0, "grad_norm": 2.8341738542865973, "language_loss": 0.7894811, "learning_rate": 1.2297779731546692e-06, "loss": 0.81118047, "num_input_tokens_seen": 114018170, "step": 5298, "time_per_iteration": 3.3058247566223145 }, { "auxiliary_loss_clip": 0.01146043, "auxiliary_loss_mlp": 0.01028631, "balance_loss_clip": 1.05049038, "balance_loss_mlp": 1.02097809, "epoch": 0.637167077496543, "flos": 25296592944000.0, "grad_norm": 1.898620769423474, "language_loss": 0.77663088, "learning_rate": 1.2290591418311853e-06, "loss": 0.79837769, "num_input_tokens_seen": 114035565, "step": 5299, "time_per_iteration": 2.548685073852539 }, { "auxiliary_loss_clip": 0.01161875, "auxiliary_loss_mlp": 0.01029176, "balance_loss_clip": 1.05050957, "balance_loss_mlp": 1.02159786, "epoch": 0.637287320387182, "flos": 27671114849280.0, "grad_norm": 1.5983034888438503, "language_loss": 0.72069663, "learning_rate": 1.2283404274588172e-06, "loss": 0.74260712, "num_input_tokens_seen": 114054510, "step": 5300, "time_per_iteration": 2.560880184173584 }, { "auxiliary_loss_clip": 0.00995687, "auxiliary_loss_mlp": 0.01008143, "balance_loss_clip": 1.00835383, "balance_loss_mlp": 1.00666475, "epoch": 0.6374075632778212, "flos": 63173417733120.0, "grad_norm": 0.7385632980503931, "language_loss": 0.52813458, "learning_rate": 1.227621830146592e-06, "loss": 0.54817289, "num_input_tokens_seen": 114109875, "step": 5301, "time_per_iteration": 3.2318148612976074 }, { "auxiliary_loss_clip": 0.01140093, "auxiliary_loss_mlp": 0.01027864, "balance_loss_clip": 1.04986262, "balance_loss_mlp": 1.02061915, "epoch": 0.6375278061684603, "flos": 25558131445440.0, "grad_norm": 1.8688794474097339, "language_loss": 0.78942001, "learning_rate": 1.2269033500035217e-06, "loss": 0.81109953, "num_input_tokens_seen": 114130010, "step": 5302, "time_per_iteration": 2.815134286880493 }, { "auxiliary_loss_clip": 0.01132876, "auxiliary_loss_mlp": 0.01026447, "balance_loss_clip": 1.04972577, "balance_loss_mlp": 1.019521, "epoch": 0.6376480490590993, "flos": 25666365997920.0, "grad_norm": 1.7955841520455207, "language_loss": 0.73203361, "learning_rate": 1.2261849871385988e-06, "loss": 0.75362688, "num_input_tokens_seen": 114151115, "step": 5303, "time_per_iteration": 2.5947883129119873 }, { "auxiliary_loss_clip": 0.01176357, "auxiliary_loss_mlp": 0.01025013, "balance_loss_clip": 1.05018246, "balance_loss_mlp": 1.01706791, "epoch": 0.6377682919497385, "flos": 31537688604480.0, "grad_norm": 2.0511277190896315, "language_loss": 0.62554812, "learning_rate": 1.2254667416607972e-06, "loss": 0.64756179, "num_input_tokens_seen": 114172715, "step": 5304, "time_per_iteration": 2.568621873855591 }, { "auxiliary_loss_clip": 0.0116228, "auxiliary_loss_mlp": 0.01026242, "balance_loss_clip": 1.05116379, "balance_loss_mlp": 1.01906848, "epoch": 0.6378885348403776, "flos": 23039212997280.0, "grad_norm": 1.7438288900893053, "language_loss": 0.82869756, "learning_rate": 1.2247486136790756e-06, "loss": 0.85058272, "num_input_tokens_seen": 114192195, "step": 5305, "time_per_iteration": 2.520615816116333 }, { "auxiliary_loss_clip": 0.01165138, "auxiliary_loss_mlp": 0.01029419, "balance_loss_clip": 1.0505774, "balance_loss_mlp": 1.0226357, "epoch": 0.6380087777310166, "flos": 18697073988960.0, "grad_norm": 2.2041255222112808, "language_loss": 0.80229747, "learning_rate": 1.2240306033023726e-06, "loss": 0.82424307, "num_input_tokens_seen": 114210020, "step": 5306, "time_per_iteration": 2.5116498470306396 }, { "auxiliary_loss_clip": 0.01134927, "auxiliary_loss_mlp": 0.01028213, "balance_loss_clip": 1.04216468, "balance_loss_mlp": 1.02105451, "epoch": 0.6381290206216558, "flos": 23331561862080.0, "grad_norm": 1.7137266562368179, "language_loss": 0.72117782, "learning_rate": 1.223312710639611e-06, "loss": 0.74280918, "num_input_tokens_seen": 114228740, "step": 5307, "time_per_iteration": 2.5699870586395264 }, { "auxiliary_loss_clip": 0.01146532, "auxiliary_loss_mlp": 0.01026696, "balance_loss_clip": 1.04805744, "balance_loss_mlp": 1.01934111, "epoch": 0.6382492635122948, "flos": 18880469962080.0, "grad_norm": 2.1251717177276683, "language_loss": 0.86526775, "learning_rate": 1.2225949357996928e-06, "loss": 0.88700002, "num_input_tokens_seen": 114246865, "step": 5308, "time_per_iteration": 2.5166263580322266 }, { "auxiliary_loss_clip": 0.01159613, "auxiliary_loss_mlp": 0.01026244, "balance_loss_clip": 1.05074143, "balance_loss_mlp": 1.01953864, "epoch": 0.6383695064029339, "flos": 27819139005120.0, "grad_norm": 1.517615996198121, "language_loss": 0.79895258, "learning_rate": 1.221877278891505e-06, "loss": 0.82081121, "num_input_tokens_seen": 114266120, "step": 5309, "time_per_iteration": 2.534383773803711 }, { "auxiliary_loss_clip": 0.01164568, "auxiliary_loss_mlp": 0.01032915, "balance_loss_clip": 1.05029798, "balance_loss_mlp": 1.02519941, "epoch": 0.638489749293573, "flos": 26395639855200.0, "grad_norm": 2.7028194391983824, "language_loss": 0.7148205, "learning_rate": 1.221159740023915e-06, "loss": 0.73679531, "num_input_tokens_seen": 114285950, "step": 5310, "time_per_iteration": 2.5286216735839844 }, { "auxiliary_loss_clip": 0.01145201, "auxiliary_loss_mlp": 0.00762575, "balance_loss_clip": 1.04973316, "balance_loss_mlp": 1.00068533, "epoch": 0.6386099921842121, "flos": 23988332153760.0, "grad_norm": 2.0517437578492874, "language_loss": 0.72375047, "learning_rate": 1.2204423193057735e-06, "loss": 0.74282813, "num_input_tokens_seen": 114304780, "step": 5311, "time_per_iteration": 2.591526746749878 }, { "auxiliary_loss_clip": 0.01050001, "auxiliary_loss_mlp": 0.01002776, "balance_loss_clip": 1.0107286, "balance_loss_mlp": 1.00142312, "epoch": 0.6387302350748512, "flos": 71731179464640.0, "grad_norm": 0.8467832196219471, "language_loss": 0.63313442, "learning_rate": 1.2197250168459122e-06, "loss": 0.6536622, "num_input_tokens_seen": 114361180, "step": 5312, "time_per_iteration": 3.1368250846862793 }, { "auxiliary_loss_clip": 0.01164877, "auxiliary_loss_mlp": 0.01026554, "balance_loss_clip": 1.049124, "balance_loss_mlp": 1.01938963, "epoch": 0.6388504779654903, "flos": 14535780849600.0, "grad_norm": 1.9802138780841108, "language_loss": 0.74485284, "learning_rate": 1.2190078327531454e-06, "loss": 0.76676714, "num_input_tokens_seen": 114377425, "step": 5313, "time_per_iteration": 2.4936699867248535 }, { "auxiliary_loss_clip": 0.01162016, "auxiliary_loss_mlp": 0.01029794, "balance_loss_clip": 1.04874325, "balance_loss_mlp": 1.02309442, "epoch": 0.6389707208561294, "flos": 22346137047360.0, "grad_norm": 1.5198117031661633, "language_loss": 0.72737426, "learning_rate": 1.2182907671362697e-06, "loss": 0.74929237, "num_input_tokens_seen": 114398120, "step": 5314, "time_per_iteration": 2.5068750381469727 }, { "auxiliary_loss_clip": 0.01163304, "auxiliary_loss_mlp": 0.01025149, "balance_loss_clip": 1.05187774, "balance_loss_mlp": 1.01769257, "epoch": 0.6390909637467684, "flos": 19426886600640.0, "grad_norm": 2.0253353908763088, "language_loss": 0.78972745, "learning_rate": 1.2175738201040626e-06, "loss": 0.81161201, "num_input_tokens_seen": 114415160, "step": 5315, "time_per_iteration": 3.2553977966308594 }, { "auxiliary_loss_clip": 0.01161535, "auxiliary_loss_mlp": 0.01034514, "balance_loss_clip": 1.04936433, "balance_loss_mlp": 1.02686405, "epoch": 0.6392112066374076, "flos": 24090855909600.0, "grad_norm": 2.064983207562518, "language_loss": 0.78332889, "learning_rate": 1.2168569917652855e-06, "loss": 0.80528939, "num_input_tokens_seen": 114435015, "step": 5316, "time_per_iteration": 3.2617080211639404 }, { "auxiliary_loss_clip": 0.01165227, "auxiliary_loss_mlp": 0.01027056, "balance_loss_clip": 1.05266786, "balance_loss_mlp": 1.01952803, "epoch": 0.6393314495280467, "flos": 26795145763680.0, "grad_norm": 1.4685704781958404, "language_loss": 0.63792008, "learning_rate": 1.2161402822286797e-06, "loss": 0.65984291, "num_input_tokens_seen": 114455700, "step": 5317, "time_per_iteration": 2.532890558242798 }, { "auxiliary_loss_clip": 0.01132578, "auxiliary_loss_mlp": 0.01026102, "balance_loss_clip": 1.04627156, "balance_loss_mlp": 1.01874721, "epoch": 0.6394516924186857, "flos": 20260695563520.0, "grad_norm": 2.0417356440849517, "language_loss": 0.78777903, "learning_rate": 1.2154236916029703e-06, "loss": 0.80936587, "num_input_tokens_seen": 114473675, "step": 5318, "time_per_iteration": 2.535743474960327 }, { "auxiliary_loss_clip": 0.0111919, "auxiliary_loss_mlp": 0.01029124, "balance_loss_clip": 1.04168379, "balance_loss_mlp": 1.02134001, "epoch": 0.6395719353093249, "flos": 18368850469440.0, "grad_norm": 2.585516293852112, "language_loss": 0.73611253, "learning_rate": 1.2147072199968627e-06, "loss": 0.75759566, "num_input_tokens_seen": 114492310, "step": 5319, "time_per_iteration": 3.3510243892669678 }, { "auxiliary_loss_clip": 0.01159624, "auxiliary_loss_mlp": 0.01026339, "balance_loss_clip": 1.04955149, "balance_loss_mlp": 1.0192759, "epoch": 0.6396921781999639, "flos": 17566318790400.0, "grad_norm": 1.9911347958269854, "language_loss": 0.71664888, "learning_rate": 1.2139908675190454e-06, "loss": 0.73850846, "num_input_tokens_seen": 114511520, "step": 5320, "time_per_iteration": 2.4671308994293213 }, { "auxiliary_loss_clip": 0.01097221, "auxiliary_loss_mlp": 0.01030517, "balance_loss_clip": 1.04096341, "balance_loss_mlp": 1.02320957, "epoch": 0.639812421090603, "flos": 21251256503520.0, "grad_norm": 2.8908077447719305, "language_loss": 0.74946773, "learning_rate": 1.2132746342781883e-06, "loss": 0.7707451, "num_input_tokens_seen": 114532680, "step": 5321, "time_per_iteration": 2.663301706314087 }, { "auxiliary_loss_clip": 0.0117888, "auxiliary_loss_mlp": 0.01031375, "balance_loss_clip": 1.05275726, "balance_loss_mlp": 1.02390051, "epoch": 0.6399326639812422, "flos": 11180970254880.0, "grad_norm": 2.3731399489628426, "language_loss": 0.79629916, "learning_rate": 1.2125585203829442e-06, "loss": 0.81840169, "num_input_tokens_seen": 114548320, "step": 5322, "time_per_iteration": 2.5157852172851562 }, { "auxiliary_loss_clip": 0.01122354, "auxiliary_loss_mlp": 0.01029605, "balance_loss_clip": 1.04652429, "balance_loss_mlp": 1.02217841, "epoch": 0.6400529068718812, "flos": 23911051632480.0, "grad_norm": 1.69973144865105, "language_loss": 0.74186158, "learning_rate": 1.211842525941946e-06, "loss": 0.76338124, "num_input_tokens_seen": 114568115, "step": 5323, "time_per_iteration": 2.5718400478363037 }, { "auxiliary_loss_clip": 0.01115602, "auxiliary_loss_mlp": 0.01027412, "balance_loss_clip": 1.0454706, "balance_loss_mlp": 1.01999676, "epoch": 0.6401731497625203, "flos": 44018730582720.0, "grad_norm": 1.8136037624849215, "language_loss": 0.78929603, "learning_rate": 1.2111266510638105e-06, "loss": 0.81072617, "num_input_tokens_seen": 114591040, "step": 5324, "time_per_iteration": 3.516815662384033 }, { "auxiliary_loss_clip": 0.01099687, "auxiliary_loss_mlp": 0.0103068, "balance_loss_clip": 1.04452121, "balance_loss_mlp": 1.02320862, "epoch": 0.6402933926531594, "flos": 20662212821760.0, "grad_norm": 1.710889651710043, "language_loss": 0.80008394, "learning_rate": 1.2104108958571346e-06, "loss": 0.82138765, "num_input_tokens_seen": 114609310, "step": 5325, "time_per_iteration": 2.603668689727783 }, { "auxiliary_loss_clip": 0.0115989, "auxiliary_loss_mlp": 0.01027447, "balance_loss_clip": 1.05083323, "balance_loss_mlp": 1.02022886, "epoch": 0.6404136355437985, "flos": 24863331481440.0, "grad_norm": 2.0456486905617353, "language_loss": 0.75924647, "learning_rate": 1.2096952604304975e-06, "loss": 0.78111988, "num_input_tokens_seen": 114629740, "step": 5326, "time_per_iteration": 2.5393903255462646 }, { "auxiliary_loss_clip": 0.01163726, "auxiliary_loss_mlp": 0.01026349, "balance_loss_clip": 1.04933286, "balance_loss_mlp": 1.01871407, "epoch": 0.6405338784344375, "flos": 40479554256960.0, "grad_norm": 2.12448765139832, "language_loss": 0.70149714, "learning_rate": 1.2089797448924616e-06, "loss": 0.72339791, "num_input_tokens_seen": 114653615, "step": 5327, "time_per_iteration": 2.6454436779022217 }, { "auxiliary_loss_clip": 0.01122417, "auxiliary_loss_mlp": 0.01023604, "balance_loss_clip": 1.04410672, "balance_loss_mlp": 1.01654708, "epoch": 0.6406541213250767, "flos": 20886044903520.0, "grad_norm": 2.094804020151483, "language_loss": 0.6588276, "learning_rate": 1.2082643493515692e-06, "loss": 0.68028784, "num_input_tokens_seen": 114671935, "step": 5328, "time_per_iteration": 2.596745729446411 }, { "auxiliary_loss_clip": 0.01160775, "auxiliary_loss_mlp": 0.01026295, "balance_loss_clip": 1.0496217, "balance_loss_mlp": 1.01929164, "epoch": 0.6407743642157158, "flos": 23295974543040.0, "grad_norm": 1.841772118628798, "language_loss": 0.81513417, "learning_rate": 1.207549073916346e-06, "loss": 0.83700484, "num_input_tokens_seen": 114692870, "step": 5329, "time_per_iteration": 2.5058538913726807 }, { "auxiliary_loss_clip": 0.01138733, "auxiliary_loss_mlp": 0.01029865, "balance_loss_clip": 1.04674959, "balance_loss_mlp": 1.02287304, "epoch": 0.6408946071063548, "flos": 15012639113280.0, "grad_norm": 2.101056891193528, "language_loss": 0.77822596, "learning_rate": 1.2068339186952976e-06, "loss": 0.79991204, "num_input_tokens_seen": 114710410, "step": 5330, "time_per_iteration": 2.504913091659546 }, { "auxiliary_loss_clip": 0.01166966, "auxiliary_loss_mlp": 0.01031657, "balance_loss_clip": 1.05297709, "balance_loss_mlp": 1.02453995, "epoch": 0.6410148499969939, "flos": 22528599179520.0, "grad_norm": 2.1335099778790654, "language_loss": 0.73397374, "learning_rate": 1.2061188837969136e-06, "loss": 0.75595987, "num_input_tokens_seen": 114730020, "step": 5331, "time_per_iteration": 2.4990596771240234 }, { "auxiliary_loss_clip": 0.01126984, "auxiliary_loss_mlp": 0.01026789, "balance_loss_clip": 1.0422256, "balance_loss_mlp": 1.01900458, "epoch": 0.641135092887633, "flos": 12422007272640.0, "grad_norm": 2.396946921287346, "language_loss": 0.84068537, "learning_rate": 1.2054039693296631e-06, "loss": 0.86222303, "num_input_tokens_seen": 114748015, "step": 5332, "time_per_iteration": 2.5436813831329346 }, { "auxiliary_loss_clip": 0.0112766, "auxiliary_loss_mlp": 0.01032339, "balance_loss_clip": 1.04347396, "balance_loss_mlp": 1.02490067, "epoch": 0.6412553357782721, "flos": 22127333340000.0, "grad_norm": 1.6813379330903424, "language_loss": 0.81513762, "learning_rate": 1.2046891754019992e-06, "loss": 0.83673763, "num_input_tokens_seen": 114768625, "step": 5333, "time_per_iteration": 2.576310396194458 }, { "auxiliary_loss_clip": 0.01166331, "auxiliary_loss_mlp": 0.01031432, "balance_loss_clip": 1.05165601, "balance_loss_mlp": 1.02418733, "epoch": 0.6413755786689112, "flos": 15888608198880.0, "grad_norm": 1.9916179645254366, "language_loss": 0.82585204, "learning_rate": 1.2039745021223548e-06, "loss": 0.8478297, "num_input_tokens_seen": 114786045, "step": 5334, "time_per_iteration": 2.483793020248413 }, { "auxiliary_loss_clip": 0.01027556, "auxiliary_loss_mlp": 0.01000503, "balance_loss_clip": 1.01319551, "balance_loss_mlp": 0.99931067, "epoch": 0.6414958215595503, "flos": 68039172498720.0, "grad_norm": 0.7948704030201739, "language_loss": 0.57142282, "learning_rate": 1.2032599495991456e-06, "loss": 0.59170341, "num_input_tokens_seen": 114850785, "step": 5335, "time_per_iteration": 3.244870185852051 }, { "auxiliary_loss_clip": 0.01164692, "auxiliary_loss_mlp": 0.01026903, "balance_loss_clip": 1.05252826, "balance_loss_mlp": 1.01915455, "epoch": 0.6416160644501894, "flos": 44091305982240.0, "grad_norm": 1.8118481935200277, "language_loss": 0.69561833, "learning_rate": 1.2025455179407685e-06, "loss": 0.7175343, "num_input_tokens_seen": 114871945, "step": 5336, "time_per_iteration": 2.7080156803131104 }, { "auxiliary_loss_clip": 0.01158963, "auxiliary_loss_mlp": 0.00762763, "balance_loss_clip": 1.04878616, "balance_loss_mlp": 1.00067449, "epoch": 0.6417363073408284, "flos": 20959841479680.0, "grad_norm": 2.364969698104862, "language_loss": 0.74308568, "learning_rate": 1.2018312072556022e-06, "loss": 0.76230294, "num_input_tokens_seen": 114890445, "step": 5337, "time_per_iteration": 2.508988618850708 }, { "auxiliary_loss_clip": 0.01173333, "auxiliary_loss_mlp": 0.00762756, "balance_loss_clip": 1.05017829, "balance_loss_mlp": 1.00071144, "epoch": 0.6418565502314676, "flos": 22455125856000.0, "grad_norm": 1.811725474520481, "language_loss": 0.74531603, "learning_rate": 1.2011170176520077e-06, "loss": 0.76467693, "num_input_tokens_seen": 114911360, "step": 5338, "time_per_iteration": 2.4762682914733887 }, { "auxiliary_loss_clip": 0.01087534, "auxiliary_loss_mlp": 0.01026592, "balance_loss_clip": 1.04060078, "balance_loss_mlp": 1.01920414, "epoch": 0.6419767931221066, "flos": 25045506277920.0, "grad_norm": 1.5116665214160068, "language_loss": 0.81343913, "learning_rate": 1.2004029492383256e-06, "loss": 0.83458042, "num_input_tokens_seen": 114932700, "step": 5339, "time_per_iteration": 2.6690828800201416 }, { "auxiliary_loss_clip": 0.01161627, "auxiliary_loss_mlp": 0.01024174, "balance_loss_clip": 1.05159402, "balance_loss_mlp": 1.01707554, "epoch": 0.6420970360127457, "flos": 19463695096320.0, "grad_norm": 1.8944630365295212, "language_loss": 0.73662418, "learning_rate": 1.1996890021228814e-06, "loss": 0.75848222, "num_input_tokens_seen": 114949475, "step": 5340, "time_per_iteration": 3.2225236892700195 }, { "auxiliary_loss_clip": 0.01142457, "auxiliary_loss_mlp": 0.01026298, "balance_loss_clip": 1.04553556, "balance_loss_mlp": 1.01874018, "epoch": 0.6422172789033849, "flos": 40406152767360.0, "grad_norm": 1.8346172469486308, "language_loss": 0.69815141, "learning_rate": 1.1989751764139785e-06, "loss": 0.71983898, "num_input_tokens_seen": 114973125, "step": 5341, "time_per_iteration": 2.703253746032715 }, { "auxiliary_loss_clip": 0.01114114, "auxiliary_loss_mlp": 0.01035525, "balance_loss_clip": 1.04058099, "balance_loss_mlp": 1.02819657, "epoch": 0.6423375217940239, "flos": 27672874780320.0, "grad_norm": 2.5103445762581282, "language_loss": 0.83263534, "learning_rate": 1.1982614722199044e-06, "loss": 0.8541317, "num_input_tokens_seen": 114994300, "step": 5342, "time_per_iteration": 2.628816604614258 }, { "auxiliary_loss_clip": 0.01150854, "auxiliary_loss_mlp": 0.01026165, "balance_loss_clip": 1.04623199, "balance_loss_mlp": 1.01950097, "epoch": 0.642457764684663, "flos": 18369245556000.0, "grad_norm": 2.2093514349396473, "language_loss": 0.78023583, "learning_rate": 1.1975478896489276e-06, "loss": 0.80200601, "num_input_tokens_seen": 115012135, "step": 5343, "time_per_iteration": 3.325920343399048 }, { "auxiliary_loss_clip": 0.01173272, "auxiliary_loss_mlp": 0.01023998, "balance_loss_clip": 1.05004954, "balance_loss_mlp": 1.01676798, "epoch": 0.6425780075753021, "flos": 19750512749280.0, "grad_norm": 1.8853758235404487, "language_loss": 0.76195097, "learning_rate": 1.1968344288092981e-06, "loss": 0.78392369, "num_input_tokens_seen": 115028715, "step": 5344, "time_per_iteration": 2.4490017890930176 }, { "auxiliary_loss_clip": 0.01164284, "auxiliary_loss_mlp": 0.00762503, "balance_loss_clip": 1.05191207, "balance_loss_mlp": 1.0007025, "epoch": 0.6426982504659412, "flos": 20558539723200.0, "grad_norm": 1.5853746774116377, "language_loss": 0.64537632, "learning_rate": 1.1961210898092468e-06, "loss": 0.66464424, "num_input_tokens_seen": 115047665, "step": 5345, "time_per_iteration": 3.2916128635406494 }, { "auxiliary_loss_clip": 0.01154085, "auxiliary_loss_mlp": 0.01025033, "balance_loss_clip": 1.04940259, "balance_loss_mlp": 1.01811588, "epoch": 0.6428184933565803, "flos": 17851987100640.0, "grad_norm": 1.9746687683039132, "language_loss": 0.787498, "learning_rate": 1.1954078727569874e-06, "loss": 0.80928922, "num_input_tokens_seen": 115064965, "step": 5346, "time_per_iteration": 2.4926257133483887 }, { "auxiliary_loss_clip": 0.01137168, "auxiliary_loss_mlp": 0.00762154, "balance_loss_clip": 1.04560196, "balance_loss_mlp": 1.00066471, "epoch": 0.6429387362472194, "flos": 22456957620960.0, "grad_norm": 1.622778736769174, "language_loss": 0.77909172, "learning_rate": 1.1946947777607141e-06, "loss": 0.79808497, "num_input_tokens_seen": 115086100, "step": 5347, "time_per_iteration": 2.5773117542266846 }, { "auxiliary_loss_clip": 0.01111258, "auxiliary_loss_mlp": 0.01024477, "balance_loss_clip": 1.0419414, "balance_loss_mlp": 1.01635885, "epoch": 0.6430589791378585, "flos": 24752582741760.0, "grad_norm": 2.339160503327391, "language_loss": 0.8025831, "learning_rate": 1.1939818049286024e-06, "loss": 0.82394052, "num_input_tokens_seen": 115104260, "step": 5348, "time_per_iteration": 2.6038877964019775 }, { "auxiliary_loss_clip": 0.01092809, "auxiliary_loss_mlp": 0.01027689, "balance_loss_clip": 1.04268026, "balance_loss_mlp": 1.0205183, "epoch": 0.6431792220284975, "flos": 24901253402880.0, "grad_norm": 1.6297321781451828, "language_loss": 0.75596464, "learning_rate": 1.1932689543688101e-06, "loss": 0.77716959, "num_input_tokens_seen": 115125365, "step": 5349, "time_per_iteration": 2.6454763412475586 }, { "auxiliary_loss_clip": 0.01147574, "auxiliary_loss_mlp": 0.01033485, "balance_loss_clip": 1.05009019, "balance_loss_mlp": 1.02576947, "epoch": 0.6432994649191367, "flos": 21032309128320.0, "grad_norm": 2.2261077241147387, "language_loss": 0.72526258, "learning_rate": 1.1925562261894756e-06, "loss": 0.74707317, "num_input_tokens_seen": 115144445, "step": 5350, "time_per_iteration": 2.5333526134490967 }, { "auxiliary_loss_clip": 0.01143046, "auxiliary_loss_mlp": 0.01026308, "balance_loss_clip": 1.04639876, "balance_loss_mlp": 1.01942348, "epoch": 0.6434197078097758, "flos": 30884438174880.0, "grad_norm": 1.7993871908613785, "language_loss": 0.77463078, "learning_rate": 1.1918436204987207e-06, "loss": 0.79632437, "num_input_tokens_seen": 115166305, "step": 5351, "time_per_iteration": 3.305356025695801 }, { "auxiliary_loss_clip": 0.01159094, "auxiliary_loss_mlp": 0.01027868, "balance_loss_clip": 1.05058551, "balance_loss_mlp": 1.02037525, "epoch": 0.6435399507004148, "flos": 15012495445440.0, "grad_norm": 3.1747908495574433, "language_loss": 0.81503308, "learning_rate": 1.191131137404645e-06, "loss": 0.83690268, "num_input_tokens_seen": 115183045, "step": 5352, "time_per_iteration": 2.4903481006622314 }, { "auxiliary_loss_clip": 0.01120676, "auxiliary_loss_mlp": 0.01026023, "balance_loss_clip": 1.04378927, "balance_loss_mlp": 1.01918292, "epoch": 0.643660193591054, "flos": 19901984933280.0, "grad_norm": 1.9878349992702544, "language_loss": 0.77407551, "learning_rate": 1.190418777015333e-06, "loss": 0.79554248, "num_input_tokens_seen": 115201955, "step": 5353, "time_per_iteration": 2.5524709224700928 }, { "auxiliary_loss_clip": 0.01145693, "auxiliary_loss_mlp": 0.01022076, "balance_loss_clip": 1.04724503, "balance_loss_mlp": 1.01553154, "epoch": 0.643780436481693, "flos": 24133626620640.0, "grad_norm": 1.4102727614101227, "language_loss": 0.73300296, "learning_rate": 1.1897065394388487e-06, "loss": 0.75468069, "num_input_tokens_seen": 115222395, "step": 5354, "time_per_iteration": 2.569164514541626 }, { "auxiliary_loss_clip": 0.01148732, "auxiliary_loss_mlp": 0.01031058, "balance_loss_clip": 1.05229902, "balance_loss_mlp": 1.02380466, "epoch": 0.6439006793723321, "flos": 23148812394240.0, "grad_norm": 1.5512578835228799, "language_loss": 0.76264638, "learning_rate": 1.1889944247832385e-06, "loss": 0.78444427, "num_input_tokens_seen": 115242635, "step": 5355, "time_per_iteration": 2.5414326190948486 }, { "auxiliary_loss_clip": 0.01162767, "auxiliary_loss_mlp": 0.01028331, "balance_loss_clip": 1.04762554, "balance_loss_mlp": 1.02041578, "epoch": 0.6440209222629713, "flos": 23617912594560.0, "grad_norm": 1.8114218032363922, "language_loss": 0.70681024, "learning_rate": 1.1882824331565283e-06, "loss": 0.72872126, "num_input_tokens_seen": 115262095, "step": 5356, "time_per_iteration": 2.5505459308624268 }, { "auxiliary_loss_clip": 0.01128232, "auxiliary_loss_mlp": 0.01025101, "balance_loss_clip": 1.04387724, "balance_loss_mlp": 1.01788926, "epoch": 0.6441411651536103, "flos": 16544875653120.0, "grad_norm": 2.1288822319063656, "language_loss": 0.89216924, "learning_rate": 1.1875705646667287e-06, "loss": 0.91370255, "num_input_tokens_seen": 115279985, "step": 5357, "time_per_iteration": 2.5298638343811035 }, { "auxiliary_loss_clip": 0.01157268, "auxiliary_loss_mlp": 0.01023063, "balance_loss_clip": 1.04536986, "balance_loss_mlp": 1.01604414, "epoch": 0.6442614080442494, "flos": 25410969296640.0, "grad_norm": 6.530787865159659, "language_loss": 0.75618756, "learning_rate": 1.1868588194218282e-06, "loss": 0.77799088, "num_input_tokens_seen": 115300365, "step": 5358, "time_per_iteration": 2.5325706005096436 }, { "auxiliary_loss_clip": 0.0115157, "auxiliary_loss_mlp": 0.01029226, "balance_loss_clip": 1.04618692, "balance_loss_mlp": 1.02181673, "epoch": 0.6443816509348885, "flos": 28294021836000.0, "grad_norm": 1.7879201536377896, "language_loss": 0.74043596, "learning_rate": 1.1861471975297979e-06, "loss": 0.76224387, "num_input_tokens_seen": 115322060, "step": 5359, "time_per_iteration": 2.6273436546325684 }, { "auxiliary_loss_clip": 0.01132589, "auxiliary_loss_mlp": 0.01030246, "balance_loss_clip": 1.04977536, "balance_loss_mlp": 1.02311087, "epoch": 0.6445018938255276, "flos": 36690081438240.0, "grad_norm": 1.4692948756919415, "language_loss": 0.70878536, "learning_rate": 1.185435699098591e-06, "loss": 0.73041368, "num_input_tokens_seen": 115348255, "step": 5360, "time_per_iteration": 2.7219295501708984 }, { "auxiliary_loss_clip": 0.01150701, "auxiliary_loss_mlp": 0.01028698, "balance_loss_clip": 1.04768777, "balance_loss_mlp": 1.02137589, "epoch": 0.6446221367161666, "flos": 14501414707200.0, "grad_norm": 3.2666074069427116, "language_loss": 0.78393853, "learning_rate": 1.1847243242361403e-06, "loss": 0.80573249, "num_input_tokens_seen": 115366845, "step": 5361, "time_per_iteration": 2.507594108581543 }, { "auxiliary_loss_clip": 0.01148864, "auxiliary_loss_mlp": 0.01029759, "balance_loss_clip": 1.04684496, "balance_loss_mlp": 1.02204585, "epoch": 0.6447423796068057, "flos": 24609371458560.0, "grad_norm": 1.7348688933762553, "language_loss": 0.78013432, "learning_rate": 1.1840130730503624e-06, "loss": 0.80192053, "num_input_tokens_seen": 115388125, "step": 5362, "time_per_iteration": 2.6267025470733643 }, { "auxiliary_loss_clip": 0.01176389, "auxiliary_loss_mlp": 0.01026492, "balance_loss_clip": 1.05084062, "balance_loss_mlp": 1.01881218, "epoch": 0.6448626224974449, "flos": 25047302125920.0, "grad_norm": 1.7309242978879202, "language_loss": 0.74811321, "learning_rate": 1.1833019456491518e-06, "loss": 0.77014202, "num_input_tokens_seen": 115409655, "step": 5363, "time_per_iteration": 2.4944682121276855 }, { "auxiliary_loss_clip": 0.01161451, "auxiliary_loss_mlp": 0.01029268, "balance_loss_clip": 1.04880548, "balance_loss_mlp": 1.02190709, "epoch": 0.6449828653880839, "flos": 22530359110560.0, "grad_norm": 2.1063369580343547, "language_loss": 0.78971386, "learning_rate": 1.1825909421403871e-06, "loss": 0.81162101, "num_input_tokens_seen": 115428750, "step": 5364, "time_per_iteration": 2.535090684890747 }, { "auxiliary_loss_clip": 0.01162861, "auxiliary_loss_mlp": 0.01023531, "balance_loss_clip": 1.05096078, "balance_loss_mlp": 1.01621723, "epoch": 0.645103108278723, "flos": 25695739682880.0, "grad_norm": 2.028373087326373, "language_loss": 0.76481205, "learning_rate": 1.181880062631926e-06, "loss": 0.78667599, "num_input_tokens_seen": 115448085, "step": 5365, "time_per_iteration": 2.5260472297668457 }, { "auxiliary_loss_clip": 0.0113928, "auxiliary_loss_mlp": 0.01026876, "balance_loss_clip": 1.04570985, "balance_loss_mlp": 1.01910949, "epoch": 0.6452233511693621, "flos": 27450335709120.0, "grad_norm": 2.033900160174446, "language_loss": 0.84910649, "learning_rate": 1.1811693072316093e-06, "loss": 0.87076807, "num_input_tokens_seen": 115465765, "step": 5366, "time_per_iteration": 3.32149076461792 }, { "auxiliary_loss_clip": 0.01175324, "auxiliary_loss_mlp": 0.007622, "balance_loss_clip": 1.05002511, "balance_loss_mlp": 1.00059283, "epoch": 0.6453435940600012, "flos": 19208621647680.0, "grad_norm": 10.57766692801126, "language_loss": 0.8427521, "learning_rate": 1.1804586760472574e-06, "loss": 0.86212736, "num_input_tokens_seen": 115482230, "step": 5367, "time_per_iteration": 2.4665040969848633 }, { "auxiliary_loss_clip": 0.01128298, "auxiliary_loss_mlp": 0.01029111, "balance_loss_clip": 1.04446864, "balance_loss_mlp": 1.0220747, "epoch": 0.6454638369506402, "flos": 25737684303840.0, "grad_norm": 4.289970935153369, "language_loss": 0.80367041, "learning_rate": 1.1797481691866736e-06, "loss": 0.82524449, "num_input_tokens_seen": 115499455, "step": 5368, "time_per_iteration": 3.3565268516540527 }, { "auxiliary_loss_clip": 0.01138041, "auxiliary_loss_mlp": 0.01030418, "balance_loss_clip": 1.04797173, "balance_loss_mlp": 1.02337587, "epoch": 0.6455840798412794, "flos": 20989179247680.0, "grad_norm": 2.0632471328343414, "language_loss": 0.82958543, "learning_rate": 1.1790377867576393e-06, "loss": 0.85127002, "num_input_tokens_seen": 115517205, "step": 5369, "time_per_iteration": 2.5452182292938232 }, { "auxiliary_loss_clip": 0.01152073, "auxiliary_loss_mlp": 0.01025824, "balance_loss_clip": 1.0489738, "balance_loss_mlp": 1.01890421, "epoch": 0.6457043227319185, "flos": 26067560003520.0, "grad_norm": 1.6279445997712783, "language_loss": 0.76320463, "learning_rate": 1.1783275288679203e-06, "loss": 0.78498363, "num_input_tokens_seen": 115534370, "step": 5370, "time_per_iteration": 2.5615971088409424 }, { "auxiliary_loss_clip": 0.01060608, "auxiliary_loss_mlp": 0.01004849, "balance_loss_clip": 1.00997734, "balance_loss_mlp": 1.00341284, "epoch": 0.6458245656225575, "flos": 60370842324480.0, "grad_norm": 0.8522849067721571, "language_loss": 0.57162094, "learning_rate": 1.177617395625262e-06, "loss": 0.5922755, "num_input_tokens_seen": 115592345, "step": 5371, "time_per_iteration": 3.725372314453125 }, { "auxiliary_loss_clip": 0.01162407, "auxiliary_loss_mlp": 0.01025705, "balance_loss_clip": 1.05058491, "balance_loss_mlp": 1.01799226, "epoch": 0.6459448085131967, "flos": 23076775749120.0, "grad_norm": 2.0126540184866646, "language_loss": 0.75422078, "learning_rate": 1.1769073871373908e-06, "loss": 0.77610189, "num_input_tokens_seen": 115612550, "step": 5372, "time_per_iteration": 2.5374906063079834 }, { "auxiliary_loss_clip": 0.01129716, "auxiliary_loss_mlp": 0.01024963, "balance_loss_clip": 1.04418468, "balance_loss_mlp": 1.01803136, "epoch": 0.6460650514038357, "flos": 22598193471360.0, "grad_norm": 1.791307678152373, "language_loss": 0.83988088, "learning_rate": 1.176197503512015e-06, "loss": 0.86142766, "num_input_tokens_seen": 115632265, "step": 5373, "time_per_iteration": 2.566713333129883 }, { "auxiliary_loss_clip": 0.01143386, "auxiliary_loss_mlp": 0.01027525, "balance_loss_clip": 1.0472486, "balance_loss_mlp": 1.02044368, "epoch": 0.6461852942944748, "flos": 20266729612800.0, "grad_norm": 2.012692546776806, "language_loss": 0.82472372, "learning_rate": 1.1754877448568223e-06, "loss": 0.84643281, "num_input_tokens_seen": 115651720, "step": 5374, "time_per_iteration": 2.534977912902832 }, { "auxiliary_loss_clip": 0.01146942, "auxiliary_loss_mlp": 0.01029969, "balance_loss_clip": 1.04703665, "balance_loss_mlp": 1.02314436, "epoch": 0.646305537185114, "flos": 23367113264160.0, "grad_norm": 1.9992309801352313, "language_loss": 0.905177, "learning_rate": 1.1747781112794837e-06, "loss": 0.9269461, "num_input_tokens_seen": 115668215, "step": 5375, "time_per_iteration": 2.5401456356048584 }, { "auxiliary_loss_clip": 0.01126863, "auxiliary_loss_mlp": 0.01033001, "balance_loss_clip": 1.04451585, "balance_loss_mlp": 1.02589035, "epoch": 0.646425780075753, "flos": 24277484409120.0, "grad_norm": 1.8665387207087916, "language_loss": 0.8305499, "learning_rate": 1.1740686028876487e-06, "loss": 0.85214853, "num_input_tokens_seen": 115687080, "step": 5376, "time_per_iteration": 3.378526210784912 }, { "auxiliary_loss_clip": 0.0115783, "auxiliary_loss_mlp": 0.0102934, "balance_loss_clip": 1.04971695, "balance_loss_mlp": 1.02207136, "epoch": 0.6465460229663921, "flos": 20813972341440.0, "grad_norm": 3.05463837190277, "language_loss": 0.74807894, "learning_rate": 1.1733592197889507e-06, "loss": 0.76995063, "num_input_tokens_seen": 115703990, "step": 5377, "time_per_iteration": 2.4851906299591064 }, { "auxiliary_loss_clip": 0.01155706, "auxiliary_loss_mlp": 0.01022713, "balance_loss_clip": 1.05078912, "balance_loss_mlp": 1.0157721, "epoch": 0.6466662658570312, "flos": 22853302836960.0, "grad_norm": 2.101761085195783, "language_loss": 0.72348166, "learning_rate": 1.1726499620910014e-06, "loss": 0.74526584, "num_input_tokens_seen": 115724270, "step": 5378, "time_per_iteration": 2.5503573417663574 }, { "auxiliary_loss_clip": 0.01157939, "auxiliary_loss_mlp": 0.01024251, "balance_loss_clip": 1.04791212, "balance_loss_mlp": 1.0167675, "epoch": 0.6467865087476703, "flos": 15304556974560.0, "grad_norm": 2.3030255094267256, "language_loss": 0.77598834, "learning_rate": 1.1719408299013955e-06, "loss": 0.7978102, "num_input_tokens_seen": 115742995, "step": 5379, "time_per_iteration": 2.4610023498535156 }, { "auxiliary_loss_clip": 0.01175742, "auxiliary_loss_mlp": 0.01025836, "balance_loss_clip": 1.0525651, "balance_loss_mlp": 1.01870155, "epoch": 0.6469067516383094, "flos": 19573653662880.0, "grad_norm": 2.5309332783624745, "language_loss": 0.75631481, "learning_rate": 1.1712318233277067e-06, "loss": 0.77833062, "num_input_tokens_seen": 115762015, "step": 5380, "time_per_iteration": 2.4882404804229736 }, { "auxiliary_loss_clip": 0.01059483, "auxiliary_loss_mlp": 0.0100301, "balance_loss_clip": 1.01004016, "balance_loss_mlp": 1.00162172, "epoch": 0.6470269945289485, "flos": 65098012706400.0, "grad_norm": 0.7586217862426111, "language_loss": 0.57942873, "learning_rate": 1.1705229424774916e-06, "loss": 0.60005367, "num_input_tokens_seen": 115816285, "step": 5381, "time_per_iteration": 2.9693028926849365 }, { "auxiliary_loss_clip": 0.01142342, "auxiliary_loss_mlp": 0.01020591, "balance_loss_clip": 1.04536688, "balance_loss_mlp": 1.01360214, "epoch": 0.6471472374195876, "flos": 30696947668320.0, "grad_norm": 1.7485684659608862, "language_loss": 0.64426959, "learning_rate": 1.1698141874582867e-06, "loss": 0.66589892, "num_input_tokens_seen": 115837330, "step": 5382, "time_per_iteration": 2.602062463760376 }, { "auxiliary_loss_clip": 0.01172402, "auxiliary_loss_mlp": 0.01027586, "balance_loss_clip": 1.05091047, "balance_loss_mlp": 1.020594, "epoch": 0.6472674803102266, "flos": 20521838978400.0, "grad_norm": 1.978740444970244, "language_loss": 0.71915317, "learning_rate": 1.169105558377609e-06, "loss": 0.74115312, "num_input_tokens_seen": 115857420, "step": 5383, "time_per_iteration": 2.4638514518737793 }, { "auxiliary_loss_clip": 0.01116958, "auxiliary_loss_mlp": 0.00761833, "balance_loss_clip": 1.04952836, "balance_loss_mlp": 1.00063097, "epoch": 0.6473877232008658, "flos": 24715450993440.0, "grad_norm": 1.664272909802723, "language_loss": 0.7821973, "learning_rate": 1.1683970553429587e-06, "loss": 0.80098522, "num_input_tokens_seen": 115878875, "step": 5384, "time_per_iteration": 2.6224820613861084 }, { "auxiliary_loss_clip": 0.01136247, "auxiliary_loss_mlp": 0.01026686, "balance_loss_clip": 1.04851592, "balance_loss_mlp": 1.01934552, "epoch": 0.6475079660915048, "flos": 15885555257280.0, "grad_norm": 1.870054786834644, "language_loss": 0.82559109, "learning_rate": 1.1676886784618128e-06, "loss": 0.84722042, "num_input_tokens_seen": 115895540, "step": 5385, "time_per_iteration": 2.5177903175354004 }, { "auxiliary_loss_clip": 0.01158929, "auxiliary_loss_mlp": 0.01026588, "balance_loss_clip": 1.04853153, "balance_loss_mlp": 1.0183897, "epoch": 0.6476282089821439, "flos": 17381593889760.0, "grad_norm": 2.0778632363552014, "language_loss": 0.83585936, "learning_rate": 1.1669804278416332e-06, "loss": 0.85771453, "num_input_tokens_seen": 115910265, "step": 5386, "time_per_iteration": 2.4732136726379395 }, { "auxiliary_loss_clip": 0.01150697, "auxiliary_loss_mlp": 0.01027067, "balance_loss_clip": 1.04801583, "balance_loss_mlp": 1.01976836, "epoch": 0.6477484518727831, "flos": 20194082379360.0, "grad_norm": 1.9152285807790157, "language_loss": 0.71228832, "learning_rate": 1.1662723035898602e-06, "loss": 0.73406595, "num_input_tokens_seen": 115930025, "step": 5387, "time_per_iteration": 2.512061595916748 }, { "auxiliary_loss_clip": 0.01158542, "auxiliary_loss_mlp": 0.01024573, "balance_loss_clip": 1.04893565, "balance_loss_mlp": 1.01717663, "epoch": 0.6478686947634221, "flos": 25410430542240.0, "grad_norm": 1.7585206889865814, "language_loss": 0.8169961, "learning_rate": 1.165564305813915e-06, "loss": 0.83882725, "num_input_tokens_seen": 115949025, "step": 5388, "time_per_iteration": 2.5246551036834717 }, { "auxiliary_loss_clip": 0.01158392, "auxiliary_loss_mlp": 0.01031557, "balance_loss_clip": 1.04849088, "balance_loss_mlp": 1.02484536, "epoch": 0.6479889376540612, "flos": 20083585058400.0, "grad_norm": 1.6972801640943191, "language_loss": 0.80939186, "learning_rate": 1.1648564346212019e-06, "loss": 0.83129132, "num_input_tokens_seen": 115968145, "step": 5389, "time_per_iteration": 2.4751057624816895 }, { "auxiliary_loss_clip": 0.01153489, "auxiliary_loss_mlp": 0.0102706, "balance_loss_clip": 1.04741299, "balance_loss_mlp": 1.01972306, "epoch": 0.6481091805447003, "flos": 26758085849280.0, "grad_norm": 1.8050645639551224, "language_loss": 0.76371545, "learning_rate": 1.164148690119104e-06, "loss": 0.78552097, "num_input_tokens_seen": 115989425, "step": 5390, "time_per_iteration": 2.5371458530426025 }, { "auxiliary_loss_clip": 0.01172469, "auxiliary_loss_mlp": 0.01027605, "balance_loss_clip": 1.0505197, "balance_loss_mlp": 1.02041626, "epoch": 0.6482294234353394, "flos": 23952098329440.0, "grad_norm": 1.7088164910894779, "language_loss": 0.73975587, "learning_rate": 1.163441072414985e-06, "loss": 0.76175654, "num_input_tokens_seen": 116009630, "step": 5391, "time_per_iteration": 2.475430727005005 }, { "auxiliary_loss_clip": 0.01160455, "auxiliary_loss_mlp": 0.01036673, "balance_loss_clip": 1.05210662, "balance_loss_mlp": 1.02940106, "epoch": 0.6483496663259785, "flos": 26209837445760.0, "grad_norm": 1.800674451362024, "language_loss": 0.70051587, "learning_rate": 1.16273358161619e-06, "loss": 0.72248715, "num_input_tokens_seen": 116029965, "step": 5392, "time_per_iteration": 2.5422985553741455 }, { "auxiliary_loss_clip": 0.01150847, "auxiliary_loss_mlp": 0.01030389, "balance_loss_clip": 1.04905939, "balance_loss_mlp": 1.02296853, "epoch": 0.6484699092166175, "flos": 20922242810880.0, "grad_norm": 2.11277720475972, "language_loss": 0.83507532, "learning_rate": 1.1620262178300446e-06, "loss": 0.8568877, "num_input_tokens_seen": 116048580, "step": 5393, "time_per_iteration": 3.3024773597717285 }, { "auxiliary_loss_clip": 0.01130811, "auxiliary_loss_mlp": 0.01022499, "balance_loss_clip": 1.04428506, "balance_loss_mlp": 1.01515245, "epoch": 0.6485901521072567, "flos": 33072870335040.0, "grad_norm": 1.810207879176183, "language_loss": 0.75996375, "learning_rate": 1.1613189811638563e-06, "loss": 0.78149688, "num_input_tokens_seen": 116070305, "step": 5394, "time_per_iteration": 2.6796741485595703 }, { "auxiliary_loss_clip": 0.01162231, "auxiliary_loss_mlp": 0.01023977, "balance_loss_clip": 1.05093229, "balance_loss_mlp": 1.01762629, "epoch": 0.6487103949978957, "flos": 22274064485280.0, "grad_norm": 1.6626273306472579, "language_loss": 0.77522343, "learning_rate": 1.1606118717249117e-06, "loss": 0.79708552, "num_input_tokens_seen": 116090405, "step": 5395, "time_per_iteration": 3.378410577774048 }, { "auxiliary_loss_clip": 0.01179902, "auxiliary_loss_mlp": 0.01027077, "balance_loss_clip": 1.05243897, "balance_loss_mlp": 1.01958442, "epoch": 0.6488306378885348, "flos": 22930403773440.0, "grad_norm": 2.1389066101319965, "language_loss": 0.67853141, "learning_rate": 1.1599048896204787e-06, "loss": 0.70060122, "num_input_tokens_seen": 116110285, "step": 5396, "time_per_iteration": 2.4746994972229004 }, { "auxiliary_loss_clip": 0.01135574, "auxiliary_loss_mlp": 0.01027967, "balance_loss_clip": 1.04593325, "balance_loss_mlp": 1.02084422, "epoch": 0.648950880779174, "flos": 20376113508000.0, "grad_norm": 1.7269274301109947, "language_loss": 0.80860817, "learning_rate": 1.1591980349578061e-06, "loss": 0.83024359, "num_input_tokens_seen": 116128955, "step": 5397, "time_per_iteration": 3.2456467151641846 }, { "auxiliary_loss_clip": 0.01039108, "auxiliary_loss_mlp": 0.01001374, "balance_loss_clip": 1.00956631, "balance_loss_mlp": 1.00014007, "epoch": 0.649071123669813, "flos": 59930900307360.0, "grad_norm": 0.7361011345550771, "language_loss": 0.54321909, "learning_rate": 1.158491307844123e-06, "loss": 0.56362391, "num_input_tokens_seen": 116188875, "step": 5398, "time_per_iteration": 3.1522700786590576 }, { "auxiliary_loss_clip": 0.01145427, "auxiliary_loss_mlp": 0.01025837, "balance_loss_clip": 1.04818106, "balance_loss_mlp": 1.01860094, "epoch": 0.6491913665604521, "flos": 20446569806880.0, "grad_norm": 1.5842956450928516, "language_loss": 0.83855033, "learning_rate": 1.1577847083866387e-06, "loss": 0.86026293, "num_input_tokens_seen": 116207910, "step": 5399, "time_per_iteration": 2.5384628772735596 }, { "auxiliary_loss_clip": 0.01133936, "auxiliary_loss_mlp": 0.01026687, "balance_loss_clip": 1.04363501, "balance_loss_mlp": 1.01941824, "epoch": 0.6493116094510912, "flos": 16946823914880.0, "grad_norm": 1.8458779051987075, "language_loss": 0.72029668, "learning_rate": 1.1570782366925453e-06, "loss": 0.74190295, "num_input_tokens_seen": 116226425, "step": 5400, "time_per_iteration": 2.561638593673706 }, { "auxiliary_loss_clip": 0.01145387, "auxiliary_loss_mlp": 0.01025296, "balance_loss_clip": 1.043257, "balance_loss_mlp": 1.01814973, "epoch": 0.6494318523417303, "flos": 18802938022080.0, "grad_norm": 2.189421107498657, "language_loss": 0.75569034, "learning_rate": 1.1563718928690132e-06, "loss": 0.77739722, "num_input_tokens_seen": 116243860, "step": 5401, "time_per_iteration": 2.521806240081787 }, { "auxiliary_loss_clip": 0.01132147, "auxiliary_loss_mlp": 0.01025863, "balance_loss_clip": 1.04710102, "balance_loss_mlp": 1.01846004, "epoch": 0.6495520952323693, "flos": 18982850050080.0, "grad_norm": 1.9104670155381835, "language_loss": 0.71625614, "learning_rate": 1.1556656770231942e-06, "loss": 0.73783618, "num_input_tokens_seen": 116260055, "step": 5402, "time_per_iteration": 3.3389763832092285 }, { "auxiliary_loss_clip": 0.01160068, "auxiliary_loss_mlp": 0.01027254, "balance_loss_clip": 1.04819536, "balance_loss_mlp": 1.02036381, "epoch": 0.6496723381230085, "flos": 22745391537120.0, "grad_norm": 1.7334031989250145, "language_loss": 0.76259416, "learning_rate": 1.1549595892622207e-06, "loss": 0.7844674, "num_input_tokens_seen": 116278825, "step": 5403, "time_per_iteration": 2.5062694549560547 }, { "auxiliary_loss_clip": 0.01027211, "auxiliary_loss_mlp": 0.0100254, "balance_loss_clip": 1.01454961, "balance_loss_mlp": 1.0013653, "epoch": 0.6497925810136476, "flos": 62145294041280.0, "grad_norm": 0.8239273717271165, "language_loss": 0.59056371, "learning_rate": 1.1542536296932047e-06, "loss": 0.61086118, "num_input_tokens_seen": 116342360, "step": 5404, "time_per_iteration": 3.136378288269043 }, { "auxiliary_loss_clip": 0.01138163, "auxiliary_loss_mlp": 0.01029712, "balance_loss_clip": 1.0443213, "balance_loss_mlp": 1.0222913, "epoch": 0.6499128239042866, "flos": 20156735129280.0, "grad_norm": 1.8530408374381866, "language_loss": 0.70029706, "learning_rate": 1.1535477984232414e-06, "loss": 0.7219758, "num_input_tokens_seen": 116362235, "step": 5405, "time_per_iteration": 2.56707501411438 }, { "auxiliary_loss_clip": 0.01116965, "auxiliary_loss_mlp": 0.01024732, "balance_loss_clip": 1.04258323, "balance_loss_mlp": 1.01795769, "epoch": 0.6500330667949258, "flos": 24462424811520.0, "grad_norm": 2.3784435521522154, "language_loss": 0.77159333, "learning_rate": 1.152842095559404e-06, "loss": 0.79301035, "num_input_tokens_seen": 116382895, "step": 5406, "time_per_iteration": 2.6561877727508545 }, { "auxiliary_loss_clip": 0.01149771, "auxiliary_loss_mlp": 0.01026549, "balance_loss_clip": 1.04635191, "balance_loss_mlp": 1.01987672, "epoch": 0.6501533096855648, "flos": 25477402896000.0, "grad_norm": 1.688537022260673, "language_loss": 0.76576734, "learning_rate": 1.1521365212087474e-06, "loss": 0.78753054, "num_input_tokens_seen": 116402880, "step": 5407, "time_per_iteration": 2.5773112773895264 }, { "auxiliary_loss_clip": 0.01161262, "auxiliary_loss_mlp": 0.01026726, "balance_loss_clip": 1.04801273, "balance_loss_mlp": 1.01964748, "epoch": 0.6502735525762039, "flos": 44819250912000.0, "grad_norm": 2.266309143861472, "language_loss": 0.70653069, "learning_rate": 1.1514310754783062e-06, "loss": 0.7284106, "num_input_tokens_seen": 116425830, "step": 5408, "time_per_iteration": 2.7233707904815674 }, { "auxiliary_loss_clip": 0.01148465, "auxiliary_loss_mlp": 0.01026403, "balance_loss_clip": 1.04883099, "balance_loss_mlp": 1.01891088, "epoch": 0.6503937954668431, "flos": 28658551013760.0, "grad_norm": 2.046396389941613, "language_loss": 0.73057359, "learning_rate": 1.1507257584750964e-06, "loss": 0.75232226, "num_input_tokens_seen": 116446010, "step": 5409, "time_per_iteration": 2.6192848682403564 }, { "auxiliary_loss_clip": 0.01176904, "auxiliary_loss_mlp": 0.01022028, "balance_loss_clip": 1.05304456, "balance_loss_mlp": 1.01520634, "epoch": 0.6505140383574821, "flos": 20922566063520.0, "grad_norm": 2.7693092058230304, "language_loss": 0.77547073, "learning_rate": 1.150020570306113e-06, "loss": 0.79746002, "num_input_tokens_seen": 116465150, "step": 5410, "time_per_iteration": 2.464603900909424 }, { "auxiliary_loss_clip": 0.01140369, "auxiliary_loss_mlp": 0.01033264, "balance_loss_clip": 1.04501772, "balance_loss_mlp": 1.02568245, "epoch": 0.6506342812481212, "flos": 20595240468000.0, "grad_norm": 1.8920930974833876, "language_loss": 0.74761295, "learning_rate": 1.1493155110783338e-06, "loss": 0.76934928, "num_input_tokens_seen": 116483675, "step": 5411, "time_per_iteration": 2.637303590774536 }, { "auxiliary_loss_clip": 0.01160208, "auxiliary_loss_mlp": 0.01021218, "balance_loss_clip": 1.04934978, "balance_loss_mlp": 1.01425874, "epoch": 0.6507545241387603, "flos": 30226482623520.0, "grad_norm": 2.0985187201976276, "language_loss": 0.70677906, "learning_rate": 1.1486105808987155e-06, "loss": 0.72859335, "num_input_tokens_seen": 116505165, "step": 5412, "time_per_iteration": 2.567263126373291 }, { "auxiliary_loss_clip": 0.01163282, "auxiliary_loss_mlp": 0.01022105, "balance_loss_clip": 1.0520345, "balance_loss_mlp": 1.01457143, "epoch": 0.6508747670293994, "flos": 17128244455200.0, "grad_norm": 2.0843457892093573, "language_loss": 0.80917561, "learning_rate": 1.1479057798741947e-06, "loss": 0.83102953, "num_input_tokens_seen": 116523220, "step": 5413, "time_per_iteration": 2.4959359169006348 }, { "auxiliary_loss_clip": 0.01059492, "auxiliary_loss_mlp": 0.01004632, "balance_loss_clip": 1.02700078, "balance_loss_mlp": 1.00321949, "epoch": 0.6509950099200384, "flos": 68559836676960.0, "grad_norm": 0.7837911623725516, "language_loss": 0.53336799, "learning_rate": 1.14720110811169e-06, "loss": 0.55400926, "num_input_tokens_seen": 116580450, "step": 5414, "time_per_iteration": 3.1655240058898926 }, { "auxiliary_loss_clip": 0.01165976, "auxiliary_loss_mlp": 0.01025753, "balance_loss_clip": 1.05180788, "balance_loss_mlp": 1.01804936, "epoch": 0.6511152528106776, "flos": 22347465974880.0, "grad_norm": 1.8409997856498468, "language_loss": 0.76638472, "learning_rate": 1.146496565718098e-06, "loss": 0.788302, "num_input_tokens_seen": 116601020, "step": 5415, "time_per_iteration": 2.55680251121521 }, { "auxiliary_loss_clip": 0.01146267, "auxiliary_loss_mlp": 0.01024898, "balance_loss_clip": 1.04915142, "balance_loss_mlp": 1.01695275, "epoch": 0.6512354957013167, "flos": 20522162231040.0, "grad_norm": 3.041868455670668, "language_loss": 0.75684416, "learning_rate": 1.1457921528002996e-06, "loss": 0.77855575, "num_input_tokens_seen": 116619455, "step": 5416, "time_per_iteration": 2.5293734073638916 }, { "auxiliary_loss_clip": 0.01175223, "auxiliary_loss_mlp": 0.00762638, "balance_loss_clip": 1.0502491, "balance_loss_mlp": 1.00061119, "epoch": 0.6513557385919557, "flos": 32337346926720.0, "grad_norm": 2.3383589267313516, "language_loss": 0.72278947, "learning_rate": 1.1450878694651522e-06, "loss": 0.74216801, "num_input_tokens_seen": 116640020, "step": 5417, "time_per_iteration": 2.6228420734405518 }, { "auxiliary_loss_clip": 0.0111702, "auxiliary_loss_mlp": 0.01024087, "balance_loss_clip": 1.04213262, "balance_loss_mlp": 1.01645124, "epoch": 0.6514759814825949, "flos": 12093209081760.0, "grad_norm": 3.653219874232463, "language_loss": 0.63324648, "learning_rate": 1.1443837158194954e-06, "loss": 0.65465748, "num_input_tokens_seen": 116655165, "step": 5418, "time_per_iteration": 3.354722499847412 }, { "auxiliary_loss_clip": 0.01135815, "auxiliary_loss_mlp": 0.01025504, "balance_loss_clip": 1.05077732, "balance_loss_mlp": 1.01808298, "epoch": 0.651596224373234, "flos": 22526911082400.0, "grad_norm": 1.7911752351763917, "language_loss": 0.7446506, "learning_rate": 1.1436796919701484e-06, "loss": 0.76626378, "num_input_tokens_seen": 116673880, "step": 5419, "time_per_iteration": 2.6002702713012695 }, { "auxiliary_loss_clip": 0.01148485, "auxiliary_loss_mlp": 0.01029017, "balance_loss_clip": 1.05064094, "balance_loss_mlp": 1.02079129, "epoch": 0.651716467263873, "flos": 27818959420320.0, "grad_norm": 2.213236666175846, "language_loss": 0.61892283, "learning_rate": 1.1429757980239115e-06, "loss": 0.64069778, "num_input_tokens_seen": 116694305, "step": 5420, "time_per_iteration": 2.600247383117676 }, { "auxiliary_loss_clip": 0.01176674, "auxiliary_loss_mlp": 0.01028398, "balance_loss_clip": 1.050704, "balance_loss_mlp": 1.02014852, "epoch": 0.6518367101545122, "flos": 24316304254560.0, "grad_norm": 5.336351332657909, "language_loss": 0.81606019, "learning_rate": 1.1422720340875636e-06, "loss": 0.83811092, "num_input_tokens_seen": 116713055, "step": 5421, "time_per_iteration": 3.2874269485473633 }, { "auxiliary_loss_clip": 0.01168996, "auxiliary_loss_mlp": 0.0102611, "balance_loss_clip": 1.05078924, "balance_loss_mlp": 1.01899338, "epoch": 0.6519569530451512, "flos": 20011943499840.0, "grad_norm": 2.20826499879494, "language_loss": 0.79056704, "learning_rate": 1.1415684002678671e-06, "loss": 0.81251812, "num_input_tokens_seen": 116731815, "step": 5422, "time_per_iteration": 2.5160863399505615 }, { "auxiliary_loss_clip": 0.01146603, "auxiliary_loss_mlp": 0.01028991, "balance_loss_clip": 1.04499435, "balance_loss_mlp": 1.02075982, "epoch": 0.6520771959357903, "flos": 21576067911840.0, "grad_norm": 2.3891421907128136, "language_loss": 0.77485847, "learning_rate": 1.1408648966715617e-06, "loss": 0.79661441, "num_input_tokens_seen": 116749335, "step": 5423, "time_per_iteration": 3.2757744789123535 }, { "auxiliary_loss_clip": 0.01144626, "auxiliary_loss_mlp": 0.01023925, "balance_loss_clip": 1.04396987, "balance_loss_mlp": 1.01634383, "epoch": 0.6521974388264293, "flos": 22711025394720.0, "grad_norm": 1.6839570266472932, "language_loss": 0.72852576, "learning_rate": 1.1401615234053683e-06, "loss": 0.75021124, "num_input_tokens_seen": 116768155, "step": 5424, "time_per_iteration": 2.5610475540161133 }, { "auxiliary_loss_clip": 0.0115025, "auxiliary_loss_mlp": 0.01022026, "balance_loss_clip": 1.04840744, "balance_loss_mlp": 1.01459908, "epoch": 0.6523176817170685, "flos": 23002943256000.0, "grad_norm": 1.7631657162324095, "language_loss": 0.7590152, "learning_rate": 1.1394582805759885e-06, "loss": 0.78073794, "num_input_tokens_seen": 116787435, "step": 5425, "time_per_iteration": 2.5414023399353027 }, { "auxiliary_loss_clip": 0.01160538, "auxiliary_loss_mlp": 0.01027988, "balance_loss_clip": 1.05059886, "balance_loss_mlp": 1.02082396, "epoch": 0.6524379246077076, "flos": 21688253329920.0, "grad_norm": 2.1438850306878807, "language_loss": 0.75963813, "learning_rate": 1.1387551682901022e-06, "loss": 0.78152341, "num_input_tokens_seen": 116808040, "step": 5426, "time_per_iteration": 2.5246009826660156 }, { "auxiliary_loss_clip": 0.01127865, "auxiliary_loss_mlp": 0.01019668, "balance_loss_clip": 1.04420161, "balance_loss_mlp": 1.01242018, "epoch": 0.6525581674983466, "flos": 19390940112000.0, "grad_norm": 1.8553608005468343, "language_loss": 0.70714355, "learning_rate": 1.138052186654373e-06, "loss": 0.72861886, "num_input_tokens_seen": 116825510, "step": 5427, "time_per_iteration": 2.548898458480835 }, { "auxiliary_loss_clip": 0.01149502, "auxiliary_loss_mlp": 0.01025697, "balance_loss_clip": 1.04893637, "balance_loss_mlp": 1.01819003, "epoch": 0.6526784103889858, "flos": 17165448037440.0, "grad_norm": 1.9733084641291616, "language_loss": 0.88000262, "learning_rate": 1.1373493357754417e-06, "loss": 0.90175462, "num_input_tokens_seen": 116844415, "step": 5428, "time_per_iteration": 2.5343339443206787 }, { "auxiliary_loss_clip": 0.01175402, "auxiliary_loss_mlp": 0.01023372, "balance_loss_clip": 1.04994512, "balance_loss_mlp": 1.01670218, "epoch": 0.6527986532796248, "flos": 18989171435040.0, "grad_norm": 1.767194274697831, "language_loss": 0.77340615, "learning_rate": 1.1366466157599303e-06, "loss": 0.79539382, "num_input_tokens_seen": 116863690, "step": 5429, "time_per_iteration": 3.1989052295684814 }, { "auxiliary_loss_clip": 0.01113749, "auxiliary_loss_mlp": 0.00762412, "balance_loss_clip": 1.04266715, "balance_loss_mlp": 1.00060022, "epoch": 0.6529188961702639, "flos": 14238583195200.0, "grad_norm": 2.9573707414367374, "language_loss": 0.76014793, "learning_rate": 1.1359440267144412e-06, "loss": 0.77890962, "num_input_tokens_seen": 116881145, "step": 5430, "time_per_iteration": 2.5738375186920166 }, { "auxiliary_loss_clip": 0.01163903, "auxiliary_loss_mlp": 0.01025567, "balance_loss_clip": 1.05020165, "balance_loss_mlp": 1.01828349, "epoch": 0.653039139060903, "flos": 36682933963200.0, "grad_norm": 2.15795397304769, "language_loss": 0.74537587, "learning_rate": 1.1352415687455556e-06, "loss": 0.76727057, "num_input_tokens_seen": 116902405, "step": 5431, "time_per_iteration": 2.633624792098999 }, { "auxiliary_loss_clip": 0.01162409, "auxiliary_loss_mlp": 0.01027369, "balance_loss_clip": 1.0511651, "balance_loss_mlp": 1.0195787, "epoch": 0.6531593819515421, "flos": 25376279901600.0, "grad_norm": 2.977745856961898, "language_loss": 0.63554215, "learning_rate": 1.1345392419598362e-06, "loss": 0.65743995, "num_input_tokens_seen": 116921285, "step": 5432, "time_per_iteration": 2.548130512237549 }, { "auxiliary_loss_clip": 0.01154892, "auxiliary_loss_mlp": 0.0102884, "balance_loss_clip": 1.04755807, "balance_loss_mlp": 1.02124596, "epoch": 0.6532796248421812, "flos": 21178537436160.0, "grad_norm": 2.1272910888718064, "language_loss": 0.71922863, "learning_rate": 1.1338370464638263e-06, "loss": 0.74106592, "num_input_tokens_seen": 116940685, "step": 5433, "time_per_iteration": 2.5116348266601562 }, { "auxiliary_loss_clip": 0.01173018, "auxiliary_loss_mlp": 0.01029885, "balance_loss_clip": 1.04815924, "balance_loss_mlp": 1.02247298, "epoch": 0.6533998677328203, "flos": 17675954104320.0, "grad_norm": 2.1442377580490724, "language_loss": 0.6410464, "learning_rate": 1.1331349823640474e-06, "loss": 0.66307545, "num_input_tokens_seen": 116958115, "step": 5434, "time_per_iteration": 2.4355616569519043 }, { "auxiliary_loss_clip": 0.01161949, "auxiliary_loss_mlp": 0.00761598, "balance_loss_clip": 1.04956007, "balance_loss_mlp": 1.00060487, "epoch": 0.6535201106234594, "flos": 28400388706560.0, "grad_norm": 2.0214862899790615, "language_loss": 0.77986622, "learning_rate": 1.132433049767003e-06, "loss": 0.79910159, "num_input_tokens_seen": 116976030, "step": 5435, "time_per_iteration": 2.5668559074401855 }, { "auxiliary_loss_clip": 0.01146299, "auxiliary_loss_mlp": 0.01021072, "balance_loss_clip": 1.04960918, "balance_loss_mlp": 1.01436615, "epoch": 0.6536403535140984, "flos": 23586671227680.0, "grad_norm": 1.6065635009503823, "language_loss": 0.81153947, "learning_rate": 1.1317312487791748e-06, "loss": 0.83321321, "num_input_tokens_seen": 116997680, "step": 5436, "time_per_iteration": 2.5688915252685547 }, { "auxiliary_loss_clip": 0.01156186, "auxiliary_loss_mlp": 0.01026733, "balance_loss_clip": 1.04825187, "balance_loss_mlp": 1.01939571, "epoch": 0.6537605964047376, "flos": 21579480023040.0, "grad_norm": 1.8886197034761798, "language_loss": 0.72914743, "learning_rate": 1.1310295795070253e-06, "loss": 0.75097662, "num_input_tokens_seen": 117017620, "step": 5437, "time_per_iteration": 2.545234203338623 }, { "auxiliary_loss_clip": 0.01118098, "auxiliary_loss_mlp": 0.01025886, "balance_loss_clip": 1.04312801, "balance_loss_mlp": 1.01872098, "epoch": 0.6538808392953767, "flos": 26834001526080.0, "grad_norm": 2.2391983500610877, "language_loss": 0.80807734, "learning_rate": 1.1303280420569982e-06, "loss": 0.82951713, "num_input_tokens_seen": 117039505, "step": 5438, "time_per_iteration": 2.6668264865875244 }, { "auxiliary_loss_clip": 0.01155313, "auxiliary_loss_mlp": 0.01029848, "balance_loss_clip": 1.04748571, "balance_loss_mlp": 1.02253485, "epoch": 0.6540010821860157, "flos": 30738245784000.0, "grad_norm": 1.6650202204434654, "language_loss": 0.77261776, "learning_rate": 1.1296266365355158e-06, "loss": 0.79446936, "num_input_tokens_seen": 117062890, "step": 5439, "time_per_iteration": 2.597443103790283 }, { "auxiliary_loss_clip": 0.01136792, "auxiliary_loss_mlp": 0.01025806, "balance_loss_clip": 1.0476923, "balance_loss_mlp": 1.01840591, "epoch": 0.6541213250766549, "flos": 26907151596960.0, "grad_norm": 6.063120676414693, "language_loss": 0.73769677, "learning_rate": 1.1289253630489806e-06, "loss": 0.75932276, "num_input_tokens_seen": 117083940, "step": 5440, "time_per_iteration": 2.6441853046417236 }, { "auxiliary_loss_clip": 0.0116841, "auxiliary_loss_mlp": 0.01025093, "balance_loss_clip": 1.04971504, "balance_loss_mlp": 1.01733828, "epoch": 0.6542415679672939, "flos": 19172387823360.0, "grad_norm": 2.281013926891011, "language_loss": 0.72556561, "learning_rate": 1.1282242217037753e-06, "loss": 0.74750066, "num_input_tokens_seen": 117101440, "step": 5441, "time_per_iteration": 2.509918451309204 }, { "auxiliary_loss_clip": 0.01110719, "auxiliary_loss_mlp": 0.01025848, "balance_loss_clip": 1.03974843, "balance_loss_mlp": 1.017313, "epoch": 0.654361810857933, "flos": 48173522752320.0, "grad_norm": 2.010848472946986, "language_loss": 0.61388659, "learning_rate": 1.127523212606262e-06, "loss": 0.6352523, "num_input_tokens_seen": 117124265, "step": 5442, "time_per_iteration": 2.8371849060058594 }, { "auxiliary_loss_clip": 0.01159095, "auxiliary_loss_mlp": 0.0102829, "balance_loss_clip": 1.04863608, "balance_loss_mlp": 1.02072048, "epoch": 0.6544820537485722, "flos": 26943169919520.0, "grad_norm": 1.5683989634872637, "language_loss": 0.72843468, "learning_rate": 1.1268223358627835e-06, "loss": 0.75030851, "num_input_tokens_seen": 117146755, "step": 5443, "time_per_iteration": 2.555049180984497 }, { "auxiliary_loss_clip": 0.01176255, "auxiliary_loss_mlp": 0.01026402, "balance_loss_clip": 1.05039704, "balance_loss_mlp": 1.01905298, "epoch": 0.6546022966392112, "flos": 20886332239200.0, "grad_norm": 2.1419659384245344, "language_loss": 0.71713823, "learning_rate": 1.126121591579663e-06, "loss": 0.73916483, "num_input_tokens_seen": 117165960, "step": 5444, "time_per_iteration": 2.503300428390503 }, { "auxiliary_loss_clip": 0.01159568, "auxiliary_loss_mlp": 0.01029239, "balance_loss_clip": 1.05082989, "balance_loss_mlp": 1.02228272, "epoch": 0.6547225395298503, "flos": 24936697054080.0, "grad_norm": 1.6935862149136907, "language_loss": 0.69229996, "learning_rate": 1.1254209798632018e-06, "loss": 0.71418804, "num_input_tokens_seen": 117186980, "step": 5445, "time_per_iteration": 3.315274953842163 }, { "auxiliary_loss_clip": 0.01092926, "auxiliary_loss_mlp": 0.01022788, "balance_loss_clip": 1.04135895, "balance_loss_mlp": 1.01574588, "epoch": 0.6548427824204894, "flos": 22565946429600.0, "grad_norm": 1.5566888034419737, "language_loss": 0.84575039, "learning_rate": 1.124720500819683e-06, "loss": 0.86690754, "num_input_tokens_seen": 117205135, "step": 5446, "time_per_iteration": 2.6623129844665527 }, { "auxiliary_loss_clip": 0.01178235, "auxiliary_loss_mlp": 0.0102636, "balance_loss_clip": 1.05279994, "balance_loss_mlp": 1.01902854, "epoch": 0.6549630253111285, "flos": 18442503377760.0, "grad_norm": 1.8660142232748924, "language_loss": 0.82188308, "learning_rate": 1.1240201545553682e-06, "loss": 0.84392905, "num_input_tokens_seen": 117222935, "step": 5447, "time_per_iteration": 3.2200839519500732 }, { "auxiliary_loss_clip": 0.01132231, "auxiliary_loss_mlp": 0.01029905, "balance_loss_clip": 1.04732549, "balance_loss_mlp": 1.02247286, "epoch": 0.6550832682017675, "flos": 25187316799680.0, "grad_norm": 1.7170095393403324, "language_loss": 0.73056662, "learning_rate": 1.1233199411764987e-06, "loss": 0.75218797, "num_input_tokens_seen": 117242370, "step": 5448, "time_per_iteration": 2.6378049850463867 }, { "auxiliary_loss_clip": 0.01118361, "auxiliary_loss_mlp": 0.01022345, "balance_loss_clip": 1.04238582, "balance_loss_mlp": 1.01547897, "epoch": 0.6552035110924067, "flos": 22748157143040.0, "grad_norm": 1.9160010666623253, "language_loss": 0.68950486, "learning_rate": 1.1226198607892978e-06, "loss": 0.71091199, "num_input_tokens_seen": 117262930, "step": 5449, "time_per_iteration": 3.306000232696533 }, { "auxiliary_loss_clip": 0.0112126, "auxiliary_loss_mlp": 0.01025684, "balance_loss_clip": 1.04601026, "balance_loss_mlp": 1.0185256, "epoch": 0.6553237539830458, "flos": 21799181654400.0, "grad_norm": 1.76551767473952, "language_loss": 0.796215, "learning_rate": 1.1219199134999664e-06, "loss": 0.81768441, "num_input_tokens_seen": 117281430, "step": 5450, "time_per_iteration": 2.620103359222412 }, { "auxiliary_loss_clip": 0.01149157, "auxiliary_loss_mlp": 0.01025736, "balance_loss_clip": 1.05079031, "balance_loss_mlp": 1.01753461, "epoch": 0.6554439968736848, "flos": 20887230163200.0, "grad_norm": 2.2173468461929238, "language_loss": 0.78332162, "learning_rate": 1.1212200994146863e-06, "loss": 0.80507052, "num_input_tokens_seen": 117299185, "step": 5451, "time_per_iteration": 2.5091302394866943 }, { "auxiliary_loss_clip": 0.01128798, "auxiliary_loss_mlp": 0.01026319, "balance_loss_clip": 1.04170096, "balance_loss_mlp": 1.0187434, "epoch": 0.655564239764324, "flos": 16139048359680.0, "grad_norm": 1.8922190404499133, "language_loss": 0.75541317, "learning_rate": 1.120520418639618e-06, "loss": 0.77696431, "num_input_tokens_seen": 117317720, "step": 5452, "time_per_iteration": 2.5790019035339355 }, { "auxiliary_loss_clip": 0.01163327, "auxiliary_loss_mlp": 0.01024019, "balance_loss_clip": 1.05354333, "balance_loss_mlp": 1.01682711, "epoch": 0.655684482654963, "flos": 29570358837120.0, "grad_norm": 2.0186242922799824, "language_loss": 0.83240724, "learning_rate": 1.119820871280903e-06, "loss": 0.85428071, "num_input_tokens_seen": 117338795, "step": 5453, "time_per_iteration": 2.5488784313201904 }, { "auxiliary_loss_clip": 0.01159514, "auxiliary_loss_mlp": 0.01026179, "balance_loss_clip": 1.04896855, "balance_loss_mlp": 1.01893663, "epoch": 0.6558047255456021, "flos": 29789413963200.0, "grad_norm": 2.0668183340667032, "language_loss": 0.73509741, "learning_rate": 1.1191214574446614e-06, "loss": 0.75695431, "num_input_tokens_seen": 117359040, "step": 5454, "time_per_iteration": 3.2776360511779785 }, { "auxiliary_loss_clip": 0.01138371, "auxiliary_loss_mlp": 0.01028435, "balance_loss_clip": 1.04418159, "balance_loss_mlp": 1.02141309, "epoch": 0.6559249684362413, "flos": 29059170348000.0, "grad_norm": 1.4783854048320522, "language_loss": 0.79773784, "learning_rate": 1.118422177236995e-06, "loss": 0.81940585, "num_input_tokens_seen": 117380865, "step": 5455, "time_per_iteration": 2.5954620838165283 }, { "auxiliary_loss_clip": 0.01147671, "auxiliary_loss_mlp": 0.01029841, "balance_loss_clip": 1.04688025, "balance_loss_mlp": 1.02184176, "epoch": 0.6560452113268803, "flos": 20225467414080.0, "grad_norm": 1.877118480559071, "language_loss": 0.85752213, "learning_rate": 1.1177230307639835e-06, "loss": 0.87929726, "num_input_tokens_seen": 117398405, "step": 5456, "time_per_iteration": 2.567880868911743 }, { "auxiliary_loss_clip": 0.01129567, "auxiliary_loss_mlp": 0.01030689, "balance_loss_clip": 1.04580605, "balance_loss_mlp": 1.02366459, "epoch": 0.6561654542175194, "flos": 25045542194880.0, "grad_norm": 1.6903384348947912, "language_loss": 0.78672969, "learning_rate": 1.1170240181316865e-06, "loss": 0.8083322, "num_input_tokens_seen": 117419850, "step": 5457, "time_per_iteration": 2.5898149013519287 }, { "auxiliary_loss_clip": 0.0112643, "auxiliary_loss_mlp": 0.01025373, "balance_loss_clip": 1.04208338, "balance_loss_mlp": 1.01739812, "epoch": 0.6562856971081584, "flos": 22856714948160.0, "grad_norm": 2.076726557911142, "language_loss": 0.79682124, "learning_rate": 1.1163251394461442e-06, "loss": 0.81833923, "num_input_tokens_seen": 117438330, "step": 5458, "time_per_iteration": 2.584071159362793 }, { "auxiliary_loss_clip": 0.01157723, "auxiliary_loss_mlp": 0.01023962, "balance_loss_clip": 1.04835248, "balance_loss_mlp": 1.01666045, "epoch": 0.6564059399987976, "flos": 18872568230880.0, "grad_norm": 1.900665789732618, "language_loss": 0.8240658, "learning_rate": 1.1156263948133746e-06, "loss": 0.84588265, "num_input_tokens_seen": 117454985, "step": 5459, "time_per_iteration": 2.4710545539855957 }, { "auxiliary_loss_clip": 0.01109492, "auxiliary_loss_mlp": 0.00762776, "balance_loss_clip": 1.04388428, "balance_loss_mlp": 1.00063086, "epoch": 0.6565261828894366, "flos": 25484191201440.0, "grad_norm": 1.8614442131244227, "language_loss": 0.77558094, "learning_rate": 1.1149277843393787e-06, "loss": 0.79430366, "num_input_tokens_seen": 117476145, "step": 5460, "time_per_iteration": 2.6350817680358887 }, { "auxiliary_loss_clip": 0.01096123, "auxiliary_loss_mlp": 0.00762547, "balance_loss_clip": 1.03765941, "balance_loss_mlp": 1.00057042, "epoch": 0.6566464257800757, "flos": 19683504478560.0, "grad_norm": 2.388915612757396, "language_loss": 0.6356566, "learning_rate": 1.1142293081301342e-06, "loss": 0.65424335, "num_input_tokens_seen": 117494025, "step": 5461, "time_per_iteration": 2.630455493927002 }, { "auxiliary_loss_clip": 0.01139432, "auxiliary_loss_mlp": 0.01024019, "balance_loss_clip": 1.0445087, "balance_loss_mlp": 1.01726913, "epoch": 0.6567666686707149, "flos": 23514131745120.0, "grad_norm": 1.7047423359406526, "language_loss": 0.67836463, "learning_rate": 1.1135309662915995e-06, "loss": 0.69999921, "num_input_tokens_seen": 117514190, "step": 5462, "time_per_iteration": 2.5425572395324707 }, { "auxiliary_loss_clip": 0.01121867, "auxiliary_loss_mlp": 0.01024024, "balance_loss_clip": 1.04366064, "balance_loss_mlp": 1.01631761, "epoch": 0.6568869115613539, "flos": 32781347560320.0, "grad_norm": 1.9585669119522593, "language_loss": 0.60185581, "learning_rate": 1.112832758929712e-06, "loss": 0.62331474, "num_input_tokens_seen": 117536800, "step": 5463, "time_per_iteration": 2.686333656311035 }, { "auxiliary_loss_clip": 0.01160146, "auxiliary_loss_mlp": 0.01021362, "balance_loss_clip": 1.05082953, "balance_loss_mlp": 1.0140661, "epoch": 0.657007154451993, "flos": 18442431543840.0, "grad_norm": 1.929882112591817, "language_loss": 0.74871594, "learning_rate": 1.11213468615039e-06, "loss": 0.770531, "num_input_tokens_seen": 117556230, "step": 5464, "time_per_iteration": 2.482328414916992 }, { "auxiliary_loss_clip": 0.01098844, "auxiliary_loss_mlp": 0.01027977, "balance_loss_clip": 1.04264021, "balance_loss_mlp": 1.02110481, "epoch": 0.6571273973426321, "flos": 25156721938080.0, "grad_norm": 1.5374893195106174, "language_loss": 0.74985051, "learning_rate": 1.1114367480595292e-06, "loss": 0.7711187, "num_input_tokens_seen": 117577310, "step": 5465, "time_per_iteration": 2.6719906330108643 }, { "auxiliary_loss_clip": 0.01104099, "auxiliary_loss_mlp": 0.01030153, "balance_loss_clip": 1.04490054, "balance_loss_mlp": 1.02236521, "epoch": 0.6572476402332712, "flos": 17529833547360.0, "grad_norm": 2.052695072502341, "language_loss": 0.81527364, "learning_rate": 1.1107389447630086e-06, "loss": 0.83661616, "num_input_tokens_seen": 117596010, "step": 5466, "time_per_iteration": 2.5948538780212402 }, { "auxiliary_loss_clip": 0.01141451, "auxiliary_loss_mlp": 0.00762134, "balance_loss_clip": 1.0448482, "balance_loss_mlp": 1.00056636, "epoch": 0.6573678831239103, "flos": 17014263189120.0, "grad_norm": 2.0688708141503964, "language_loss": 0.78706014, "learning_rate": 1.1100412763666818e-06, "loss": 0.80609608, "num_input_tokens_seen": 117611270, "step": 5467, "time_per_iteration": 2.5222508907318115 }, { "auxiliary_loss_clip": 0.01149738, "auxiliary_loss_mlp": 0.01027174, "balance_loss_clip": 1.04931998, "balance_loss_mlp": 1.01965451, "epoch": 0.6574881260145494, "flos": 23910081874560.0, "grad_norm": 1.6156813634975447, "language_loss": 0.7971741, "learning_rate": 1.1093437429763865e-06, "loss": 0.81894326, "num_input_tokens_seen": 117631535, "step": 5468, "time_per_iteration": 2.556971788406372 }, { "auxiliary_loss_clip": 0.0116033, "auxiliary_loss_mlp": 0.01022856, "balance_loss_clip": 1.05024648, "balance_loss_mlp": 1.01567113, "epoch": 0.6576083689051885, "flos": 11218461172800.0, "grad_norm": 2.225093546086045, "language_loss": 0.73607552, "learning_rate": 1.1086463446979361e-06, "loss": 0.75790739, "num_input_tokens_seen": 117649885, "step": 5469, "time_per_iteration": 2.4799280166625977 }, { "auxiliary_loss_clip": 0.01163414, "auxiliary_loss_mlp": 0.01026773, "balance_loss_clip": 1.05230856, "balance_loss_mlp": 1.01880956, "epoch": 0.6577286117958275, "flos": 22455556859520.0, "grad_norm": 1.7210616913952608, "language_loss": 0.7735647, "learning_rate": 1.1079490816371277e-06, "loss": 0.79546654, "num_input_tokens_seen": 117669650, "step": 5470, "time_per_iteration": 3.2783396244049072 }, { "auxiliary_loss_clip": 0.01159474, "auxiliary_loss_mlp": 0.00762551, "balance_loss_clip": 1.04812622, "balance_loss_mlp": 1.00053966, "epoch": 0.6578488546864667, "flos": 21872188057440.0, "grad_norm": 1.8705514782705097, "language_loss": 0.74220586, "learning_rate": 1.1072519538997352e-06, "loss": 0.76142609, "num_input_tokens_seen": 117688790, "step": 5471, "time_per_iteration": 2.514887809753418 }, { "auxiliary_loss_clip": 0.01145595, "auxiliary_loss_mlp": 0.01026022, "balance_loss_clip": 1.04329395, "balance_loss_mlp": 1.01829076, "epoch": 0.6579690975771058, "flos": 23543756848800.0, "grad_norm": 1.7322500336725029, "language_loss": 0.82496703, "learning_rate": 1.1065549615915095e-06, "loss": 0.8466832, "num_input_tokens_seen": 117708620, "step": 5472, "time_per_iteration": 2.531935453414917 }, { "auxiliary_loss_clip": 0.01160438, "auxiliary_loss_mlp": 0.01025689, "balance_loss_clip": 1.05034423, "balance_loss_mlp": 1.01789868, "epoch": 0.6580893404677448, "flos": 32744000310240.0, "grad_norm": 2.5565601396080124, "language_loss": 0.78141892, "learning_rate": 1.105858104818187e-06, "loss": 0.80328012, "num_input_tokens_seen": 117729775, "step": 5473, "time_per_iteration": 3.3337762355804443 }, { "auxiliary_loss_clip": 0.01165333, "auxiliary_loss_mlp": 0.01028372, "balance_loss_clip": 1.05003893, "balance_loss_mlp": 1.02057576, "epoch": 0.658209583358384, "flos": 15888141278400.0, "grad_norm": 3.31875861945362, "language_loss": 0.7473675, "learning_rate": 1.105161383685478e-06, "loss": 0.76930451, "num_input_tokens_seen": 117746160, "step": 5474, "time_per_iteration": 2.4621546268463135 }, { "auxiliary_loss_clip": 0.01038585, "auxiliary_loss_mlp": 0.01004041, "balance_loss_clip": 1.01154113, "balance_loss_mlp": 1.00271177, "epoch": 0.658329826249023, "flos": 62695912964160.0, "grad_norm": 0.731544453652665, "language_loss": 0.56394136, "learning_rate": 1.1044647982990771e-06, "loss": 0.58436763, "num_input_tokens_seen": 117808045, "step": 5475, "time_per_iteration": 3.88421893119812 }, { "auxiliary_loss_clip": 0.01148441, "auxiliary_loss_mlp": 0.01029243, "balance_loss_clip": 1.04970229, "balance_loss_mlp": 1.02126169, "epoch": 0.6584500691396621, "flos": 31722629006880.0, "grad_norm": 2.2515243435909174, "language_loss": 0.64475465, "learning_rate": 1.1037683487646536e-06, "loss": 0.66653144, "num_input_tokens_seen": 117828330, "step": 5476, "time_per_iteration": 2.605677843093872 }, { "auxiliary_loss_clip": 0.01146922, "auxiliary_loss_mlp": 0.00762327, "balance_loss_clip": 1.05083704, "balance_loss_mlp": 1.00050843, "epoch": 0.6585703120303013, "flos": 18406089968640.0, "grad_norm": 1.8948673121135557, "language_loss": 0.77352273, "learning_rate": 1.1030720351878583e-06, "loss": 0.79261523, "num_input_tokens_seen": 117846450, "step": 5477, "time_per_iteration": 2.4827070236206055 }, { "auxiliary_loss_clip": 0.01052066, "auxiliary_loss_mlp": 0.01000128, "balance_loss_clip": 1.01065278, "balance_loss_mlp": 0.99885827, "epoch": 0.6586905549209403, "flos": 58309889819040.0, "grad_norm": 0.8322715775653602, "language_loss": 0.57688797, "learning_rate": 1.102375857674323e-06, "loss": 0.59740984, "num_input_tokens_seen": 117908365, "step": 5478, "time_per_iteration": 3.08695387840271 }, { "auxiliary_loss_clip": 0.01145222, "auxiliary_loss_mlp": 0.01024003, "balance_loss_clip": 1.0449152, "balance_loss_mlp": 1.01731563, "epoch": 0.6588107978115794, "flos": 22782631036320.0, "grad_norm": 1.6769099726539778, "language_loss": 0.90275353, "learning_rate": 1.1016798163296561e-06, "loss": 0.92444575, "num_input_tokens_seen": 117927565, "step": 5479, "time_per_iteration": 2.5354928970336914 }, { "auxiliary_loss_clip": 0.01159623, "auxiliary_loss_mlp": 0.01031538, "balance_loss_clip": 1.04867911, "balance_loss_mlp": 1.02357507, "epoch": 0.6589310407022185, "flos": 20667528531840.0, "grad_norm": 1.8510492565890624, "language_loss": 0.65974414, "learning_rate": 1.1009839112594471e-06, "loss": 0.68165576, "num_input_tokens_seen": 117945590, "step": 5480, "time_per_iteration": 2.491281509399414 }, { "auxiliary_loss_clip": 0.01163465, "auxiliary_loss_mlp": 0.01028857, "balance_loss_clip": 1.0498457, "balance_loss_mlp": 1.02136743, "epoch": 0.6590512835928576, "flos": 25630599094080.0, "grad_norm": 2.6470786457692475, "language_loss": 0.7194466, "learning_rate": 1.1002881425692638e-06, "loss": 0.74136984, "num_input_tokens_seen": 117966020, "step": 5481, "time_per_iteration": 3.2976417541503906 }, { "auxiliary_loss_clip": 0.01153183, "auxiliary_loss_mlp": 0.01027504, "balance_loss_clip": 1.04574347, "balance_loss_mlp": 1.02013648, "epoch": 0.6591715264834966, "flos": 23726111230080.0, "grad_norm": 1.680685870043248, "language_loss": 0.75138116, "learning_rate": 1.0995925103646532e-06, "loss": 0.77318799, "num_input_tokens_seen": 117984620, "step": 5482, "time_per_iteration": 2.5027222633361816 }, { "auxiliary_loss_clip": 0.01127692, "auxiliary_loss_mlp": 0.01024638, "balance_loss_clip": 1.0474689, "balance_loss_mlp": 1.01732731, "epoch": 0.6592917693741358, "flos": 35773855828800.0, "grad_norm": 1.7244424163258076, "language_loss": 0.6666072, "learning_rate": 1.0988970147511437e-06, "loss": 0.6881305, "num_input_tokens_seen": 118006500, "step": 5483, "time_per_iteration": 2.694948434829712 }, { "auxiliary_loss_clip": 0.01148113, "auxiliary_loss_mlp": 0.01025248, "balance_loss_clip": 1.05043578, "balance_loss_mlp": 1.0176363, "epoch": 0.6594120122647749, "flos": 21396838306080.0, "grad_norm": 1.944391268738488, "language_loss": 0.80794829, "learning_rate": 1.0982016558342405e-06, "loss": 0.82968187, "num_input_tokens_seen": 118025470, "step": 5484, "time_per_iteration": 2.5121045112609863 }, { "auxiliary_loss_clip": 0.01176167, "auxiliary_loss_mlp": 0.01025257, "balance_loss_clip": 1.05131078, "balance_loss_mlp": 1.01794982, "epoch": 0.6595322551554139, "flos": 19351832930880.0, "grad_norm": 1.6980415018455692, "language_loss": 0.71162468, "learning_rate": 1.0975064337194291e-06, "loss": 0.73363894, "num_input_tokens_seen": 118043515, "step": 5485, "time_per_iteration": 2.4720757007598877 }, { "auxiliary_loss_clip": 0.01124888, "auxiliary_loss_mlp": 0.01027899, "balance_loss_clip": 1.04348814, "balance_loss_mlp": 1.02061558, "epoch": 0.6596524980460531, "flos": 16837116767040.0, "grad_norm": 1.5161246513690372, "language_loss": 0.70367384, "learning_rate": 1.0968113485121743e-06, "loss": 0.72520173, "num_input_tokens_seen": 118063105, "step": 5486, "time_per_iteration": 2.567145347595215 }, { "auxiliary_loss_clip": 0.01162377, "auxiliary_loss_mlp": 0.00762823, "balance_loss_clip": 1.04776096, "balance_loss_mlp": 1.0004406, "epoch": 0.6597727409366921, "flos": 21798571066080.0, "grad_norm": 3.1104593769606255, "language_loss": 0.80226856, "learning_rate": 1.0961164003179185e-06, "loss": 0.82152057, "num_input_tokens_seen": 118081615, "step": 5487, "time_per_iteration": 2.5254268646240234 }, { "auxiliary_loss_clip": 0.01130646, "auxiliary_loss_mlp": 0.01031236, "balance_loss_clip": 1.04592943, "balance_loss_mlp": 1.02356505, "epoch": 0.6598929838273312, "flos": 23730708600960.0, "grad_norm": 1.9856211439486018, "language_loss": 0.84506881, "learning_rate": 1.0954215892420884e-06, "loss": 0.86668766, "num_input_tokens_seen": 118102315, "step": 5488, "time_per_iteration": 2.571664571762085 }, { "auxiliary_loss_clip": 0.01137968, "auxiliary_loss_mlp": 0.01027865, "balance_loss_clip": 1.04848433, "balance_loss_mlp": 1.0200274, "epoch": 0.6600132267179702, "flos": 19974524415840.0, "grad_norm": 1.7350768390736, "language_loss": 0.70344466, "learning_rate": 1.094726915390082e-06, "loss": 0.72510302, "num_input_tokens_seen": 118120650, "step": 5489, "time_per_iteration": 2.5663793087005615 }, { "auxiliary_loss_clip": 0.01163827, "auxiliary_loss_mlp": 0.0102677, "balance_loss_clip": 1.05157483, "balance_loss_mlp": 1.01981378, "epoch": 0.6601334696086094, "flos": 22342653102240.0, "grad_norm": 1.8687069322091667, "language_loss": 0.69385511, "learning_rate": 1.0940323788672836e-06, "loss": 0.71576107, "num_input_tokens_seen": 118139825, "step": 5490, "time_per_iteration": 2.4930641651153564 }, { "auxiliary_loss_clip": 0.01159241, "auxiliary_loss_mlp": 0.01023981, "balance_loss_clip": 1.05032444, "balance_loss_mlp": 1.01671493, "epoch": 0.6602537124992485, "flos": 25703102659680.0, "grad_norm": 1.7645056957974121, "language_loss": 0.7376259, "learning_rate": 1.0933379797790522e-06, "loss": 0.75945812, "num_input_tokens_seen": 118159240, "step": 5491, "time_per_iteration": 2.5332112312316895 }, { "auxiliary_loss_clip": 0.01176081, "auxiliary_loss_mlp": 0.01026936, "balance_loss_clip": 1.05208242, "balance_loss_mlp": 1.01965201, "epoch": 0.6603739553898875, "flos": 25848576711360.0, "grad_norm": 2.0753116662591924, "language_loss": 0.71408552, "learning_rate": 1.0926437182307293e-06, "loss": 0.73611563, "num_input_tokens_seen": 118178050, "step": 5492, "time_per_iteration": 2.490546703338623 }, { "auxiliary_loss_clip": 0.01152286, "auxiliary_loss_mlp": 0.0102938, "balance_loss_clip": 1.04647565, "balance_loss_mlp": 1.02203059, "epoch": 0.6604941982805267, "flos": 24570300194400.0, "grad_norm": 1.7086476833729258, "language_loss": 0.77868569, "learning_rate": 1.0919495943276338e-06, "loss": 0.8005023, "num_input_tokens_seen": 118199070, "step": 5493, "time_per_iteration": 2.5944266319274902 }, { "auxiliary_loss_clip": 0.01134248, "auxiliary_loss_mlp": 0.0103077, "balance_loss_clip": 1.04255116, "balance_loss_mlp": 1.02271438, "epoch": 0.6606144411711657, "flos": 13261778450880.0, "grad_norm": 2.4574186418217567, "language_loss": 0.76218903, "learning_rate": 1.0912556081750611e-06, "loss": 0.78383917, "num_input_tokens_seen": 118217000, "step": 5494, "time_per_iteration": 2.5349864959716797 }, { "auxiliary_loss_clip": 0.01142843, "auxiliary_loss_mlp": 0.01024436, "balance_loss_clip": 1.04804969, "balance_loss_mlp": 1.01745653, "epoch": 0.6607346840618048, "flos": 25155285259680.0, "grad_norm": 1.8065911152764749, "language_loss": 0.76678652, "learning_rate": 1.0905617598782909e-06, "loss": 0.78845936, "num_input_tokens_seen": 118237205, "step": 5495, "time_per_iteration": 2.5668532848358154 }, { "auxiliary_loss_clip": 0.01110674, "auxiliary_loss_mlp": 0.01025944, "balance_loss_clip": 1.04306769, "balance_loss_mlp": 1.01892865, "epoch": 0.660854926952444, "flos": 17638032182880.0, "grad_norm": 2.366256995375718, "language_loss": 0.81504083, "learning_rate": 1.0898680495425775e-06, "loss": 0.83640707, "num_input_tokens_seen": 118255495, "step": 5496, "time_per_iteration": 3.3228225708007812 }, { "auxiliary_loss_clip": 0.01148839, "auxiliary_loss_mlp": 0.0102596, "balance_loss_clip": 1.04720986, "balance_loss_mlp": 1.01884949, "epoch": 0.660975169843083, "flos": 16836003341280.0, "grad_norm": 1.654701547396033, "language_loss": 0.80209649, "learning_rate": 1.0891744772731594e-06, "loss": 0.82384443, "num_input_tokens_seen": 118273310, "step": 5497, "time_per_iteration": 2.523881196975708 }, { "auxiliary_loss_clip": 0.01161852, "auxiliary_loss_mlp": 0.0102756, "balance_loss_clip": 1.04836464, "balance_loss_mlp": 1.02110183, "epoch": 0.6610954127337221, "flos": 26870414935200.0, "grad_norm": 1.5394291473701542, "language_loss": 0.65288311, "learning_rate": 1.088481043175248e-06, "loss": 0.67477727, "num_input_tokens_seen": 118293880, "step": 5498, "time_per_iteration": 2.5324153900146484 }, { "auxiliary_loss_clip": 0.01137315, "auxiliary_loss_mlp": 0.01024099, "balance_loss_clip": 1.04583168, "balance_loss_mlp": 1.01704741, "epoch": 0.6612156556243612, "flos": 26465701067520.0, "grad_norm": 1.6776585251997953, "language_loss": 0.75255698, "learning_rate": 1.0877877473540368e-06, "loss": 0.77417117, "num_input_tokens_seen": 118314465, "step": 5499, "time_per_iteration": 3.3487813472747803 }, { "auxiliary_loss_clip": 0.01176281, "auxiliary_loss_mlp": 0.01025685, "balance_loss_clip": 1.0509367, "balance_loss_mlp": 1.01859236, "epoch": 0.6613358985150003, "flos": 19791918615840.0, "grad_norm": 2.6249784052628793, "language_loss": 0.72617567, "learning_rate": 1.0870945899147002e-06, "loss": 0.74819541, "num_input_tokens_seen": 118331110, "step": 5500, "time_per_iteration": 2.4637293815612793 }, { "auxiliary_loss_clip": 0.01160255, "auxiliary_loss_mlp": 0.01025301, "balance_loss_clip": 1.05121076, "balance_loss_mlp": 1.01822627, "epoch": 0.6614561414056394, "flos": 26831631006720.0, "grad_norm": 2.0063135182485006, "language_loss": 0.76178718, "learning_rate": 1.0864015709623879e-06, "loss": 0.78364277, "num_input_tokens_seen": 118351980, "step": 5501, "time_per_iteration": 3.298889398574829 }, { "auxiliary_loss_clip": 0.01164774, "auxiliary_loss_mlp": 0.01024872, "balance_loss_clip": 1.04904151, "balance_loss_mlp": 1.01751935, "epoch": 0.6615763842962785, "flos": 22894600952640.0, "grad_norm": 2.477565551407841, "language_loss": 0.80190206, "learning_rate": 1.0857086906022313e-06, "loss": 0.82379842, "num_input_tokens_seen": 118370315, "step": 5502, "time_per_iteration": 2.530749559402466 }, { "auxiliary_loss_clip": 0.01093547, "auxiliary_loss_mlp": 0.01029123, "balance_loss_clip": 1.04371059, "balance_loss_mlp": 1.02114201, "epoch": 0.6616966271869176, "flos": 24790325078400.0, "grad_norm": 2.2763511662102305, "language_loss": 0.72990942, "learning_rate": 1.0850159489393388e-06, "loss": 0.75113606, "num_input_tokens_seen": 118389575, "step": 5503, "time_per_iteration": 2.6340830326080322 }, { "auxiliary_loss_clip": 0.01123263, "auxiliary_loss_mlp": 0.01025211, "balance_loss_clip": 1.04035795, "balance_loss_mlp": 1.01766813, "epoch": 0.6618168700775566, "flos": 17202112865280.0, "grad_norm": 1.6862048441966166, "language_loss": 0.82103306, "learning_rate": 1.0843233460787992e-06, "loss": 0.84251785, "num_input_tokens_seen": 118406790, "step": 5504, "time_per_iteration": 2.5529308319091797 }, { "auxiliary_loss_clip": 0.0112305, "auxiliary_loss_mlp": 0.01027827, "balance_loss_clip": 1.04729879, "balance_loss_mlp": 1.02040052, "epoch": 0.6619371129681958, "flos": 25447095370080.0, "grad_norm": 1.944973522559446, "language_loss": 0.77772075, "learning_rate": 1.0836308821256805e-06, "loss": 0.7992295, "num_input_tokens_seen": 118427590, "step": 5505, "time_per_iteration": 2.611388921737671 }, { "auxiliary_loss_clip": 0.01161187, "auxiliary_loss_mlp": 0.01028857, "balance_loss_clip": 1.05072737, "balance_loss_mlp": 1.02174354, "epoch": 0.6620573558588349, "flos": 18040447365120.0, "grad_norm": 2.0756552624674702, "language_loss": 0.77487284, "learning_rate": 1.0829385571850282e-06, "loss": 0.79677331, "num_input_tokens_seen": 118444570, "step": 5506, "time_per_iteration": 3.275320529937744 }, { "auxiliary_loss_clip": 0.01179257, "auxiliary_loss_mlp": 0.0103122, "balance_loss_clip": 1.05006206, "balance_loss_mlp": 1.0229888, "epoch": 0.6621775987494739, "flos": 17785589418240.0, "grad_norm": 2.6256813089875934, "language_loss": 0.83470678, "learning_rate": 1.0822463713618679e-06, "loss": 0.85681158, "num_input_tokens_seen": 118461425, "step": 5507, "time_per_iteration": 2.440821647644043 }, { "auxiliary_loss_clip": 0.01135701, "auxiliary_loss_mlp": 0.01026127, "balance_loss_clip": 1.046785, "balance_loss_mlp": 1.01893616, "epoch": 0.6622978416401131, "flos": 17492594048160.0, "grad_norm": 2.0037981594118213, "language_loss": 0.84763128, "learning_rate": 1.0815543247612034e-06, "loss": 0.86924958, "num_input_tokens_seen": 118478495, "step": 5508, "time_per_iteration": 2.5463738441467285 }, { "auxiliary_loss_clip": 0.01142857, "auxiliary_loss_mlp": 0.010218, "balance_loss_clip": 1.04201984, "balance_loss_mlp": 1.01491821, "epoch": 0.6624180845307521, "flos": 21648355975680.0, "grad_norm": 1.708725711591626, "language_loss": 0.82817537, "learning_rate": 1.0808624174880168e-06, "loss": 0.84982193, "num_input_tokens_seen": 118499145, "step": 5509, "time_per_iteration": 2.570100784301758 }, { "auxiliary_loss_clip": 0.011727, "auxiliary_loss_mlp": 0.01027574, "balance_loss_clip": 1.0504595, "balance_loss_mlp": 1.02019787, "epoch": 0.6625383274213912, "flos": 23805905938560.0, "grad_norm": 1.6105332604617544, "language_loss": 0.80092525, "learning_rate": 1.080170649647272e-06, "loss": 0.82292801, "num_input_tokens_seen": 118518950, "step": 5510, "time_per_iteration": 2.4991986751556396 }, { "auxiliary_loss_clip": 0.01174075, "auxiliary_loss_mlp": 0.01030504, "balance_loss_clip": 1.05032575, "balance_loss_mlp": 1.02346754, "epoch": 0.6626585703120303, "flos": 33262946862720.0, "grad_norm": 1.6302781600153644, "language_loss": 0.67014247, "learning_rate": 1.0794790213439068e-06, "loss": 0.69218826, "num_input_tokens_seen": 118545850, "step": 5511, "time_per_iteration": 2.6189699172973633 }, { "auxiliary_loss_clip": 0.01119326, "auxiliary_loss_mlp": 0.01029116, "balance_loss_clip": 1.0445559, "balance_loss_mlp": 1.02081871, "epoch": 0.6627788132026694, "flos": 22085783805600.0, "grad_norm": 1.9480962271603275, "language_loss": 0.78605783, "learning_rate": 1.078787532682843e-06, "loss": 0.80754232, "num_input_tokens_seen": 118563325, "step": 5512, "time_per_iteration": 2.613241195678711 }, { "auxiliary_loss_clip": 0.01157802, "auxiliary_loss_mlp": 0.01029431, "balance_loss_clip": 1.04835153, "balance_loss_mlp": 1.02224839, "epoch": 0.6628990560933085, "flos": 36173613156000.0, "grad_norm": 2.0036264136717463, "language_loss": 0.75858945, "learning_rate": 1.0780961837689773e-06, "loss": 0.78046173, "num_input_tokens_seen": 118582835, "step": 5513, "time_per_iteration": 2.612069845199585 }, { "auxiliary_loss_clip": 0.01140039, "auxiliary_loss_mlp": 0.01027508, "balance_loss_clip": 1.04739141, "balance_loss_mlp": 1.02071881, "epoch": 0.6630192989839476, "flos": 18513570264960.0, "grad_norm": 1.5645203403088903, "language_loss": 0.6988529, "learning_rate": 1.0774049747071883e-06, "loss": 0.7205283, "num_input_tokens_seen": 118600715, "step": 5514, "time_per_iteration": 2.49336838722229 }, { "auxiliary_loss_clip": 0.01118044, "auxiliary_loss_mlp": 0.01025195, "balance_loss_clip": 1.04766858, "balance_loss_mlp": 1.01713371, "epoch": 0.6631395418745867, "flos": 35809514981760.0, "grad_norm": 1.6402743470366112, "language_loss": 0.68244326, "learning_rate": 1.076713905602332e-06, "loss": 0.7038756, "num_input_tokens_seen": 118621290, "step": 5515, "time_per_iteration": 2.7203001976013184 }, { "auxiliary_loss_clip": 0.01165294, "auxiliary_loss_mlp": 0.01030562, "balance_loss_clip": 1.05095887, "balance_loss_mlp": 1.02319777, "epoch": 0.6632597847652257, "flos": 20047746320640.0, "grad_norm": 1.7470266774263672, "language_loss": 0.81361568, "learning_rate": 1.07602297655924e-06, "loss": 0.83557427, "num_input_tokens_seen": 118639610, "step": 5516, "time_per_iteration": 2.482387065887451 }, { "auxiliary_loss_clip": 0.01175372, "auxiliary_loss_mlp": 0.01023642, "balance_loss_clip": 1.05187142, "balance_loss_mlp": 1.01655769, "epoch": 0.6633800276558649, "flos": 21214483924800.0, "grad_norm": 1.807760569789939, "language_loss": 0.81201792, "learning_rate": 1.0753321876827292e-06, "loss": 0.83400804, "num_input_tokens_seen": 118658895, "step": 5517, "time_per_iteration": 2.4751522541046143 }, { "auxiliary_loss_clip": 0.01173364, "auxiliary_loss_mlp": 0.01030189, "balance_loss_clip": 1.04920781, "balance_loss_mlp": 1.02289295, "epoch": 0.663500270546504, "flos": 23987757482400.0, "grad_norm": 1.973782951384103, "language_loss": 0.7428965, "learning_rate": 1.0746415390775893e-06, "loss": 0.76493204, "num_input_tokens_seen": 118677025, "step": 5518, "time_per_iteration": 2.4678103923797607 }, { "auxiliary_loss_clip": 0.0117466, "auxiliary_loss_mlp": 0.01026115, "balance_loss_clip": 1.05192411, "balance_loss_mlp": 1.01890254, "epoch": 0.663620513437143, "flos": 17932392397440.0, "grad_norm": 1.9213508040679899, "language_loss": 0.76547384, "learning_rate": 1.0739510308485939e-06, "loss": 0.78748161, "num_input_tokens_seen": 118694240, "step": 5519, "time_per_iteration": 2.4331002235412598 }, { "auxiliary_loss_clip": 0.0104057, "auxiliary_loss_mlp": 0.01002593, "balance_loss_clip": 1.01095581, "balance_loss_mlp": 1.00125754, "epoch": 0.6637407563277821, "flos": 57840250864320.0, "grad_norm": 0.8224011878337583, "language_loss": 0.62510246, "learning_rate": 1.07326066310049e-06, "loss": 0.6455341, "num_input_tokens_seen": 118758365, "step": 5520, "time_per_iteration": 3.1792168617248535 }, { "auxiliary_loss_clip": 0.01130627, "auxiliary_loss_mlp": 0.01028492, "balance_loss_clip": 1.04629564, "balance_loss_mlp": 1.02070189, "epoch": 0.6638609992184212, "flos": 27306011000160.0, "grad_norm": 5.0783429147339385, "language_loss": 0.79591823, "learning_rate": 1.0725704359380059e-06, "loss": 0.81750941, "num_input_tokens_seen": 118778220, "step": 5521, "time_per_iteration": 2.6356847286224365 }, { "auxiliary_loss_clip": 0.01175203, "auxiliary_loss_mlp": 0.01030079, "balance_loss_clip": 1.05162334, "balance_loss_mlp": 1.02298951, "epoch": 0.6639812421090603, "flos": 18624857759040.0, "grad_norm": 2.290289735174239, "language_loss": 0.72187972, "learning_rate": 1.0718803494658497e-06, "loss": 0.74393255, "num_input_tokens_seen": 118797110, "step": 5522, "time_per_iteration": 2.43585467338562 }, { "auxiliary_loss_clip": 0.01066128, "auxiliary_loss_mlp": 0.01030485, "balance_loss_clip": 1.03847051, "balance_loss_mlp": 1.02338934, "epoch": 0.6641014849996993, "flos": 15924482853600.0, "grad_norm": 2.3432507707467605, "language_loss": 0.83461744, "learning_rate": 1.071190403788707e-06, "loss": 0.85558355, "num_input_tokens_seen": 118812415, "step": 5523, "time_per_iteration": 3.6343207359313965 }, { "auxiliary_loss_clip": 0.01141411, "auxiliary_loss_mlp": 0.01027778, "balance_loss_clip": 1.04946601, "balance_loss_mlp": 1.02032125, "epoch": 0.6642217278903385, "flos": 26505490670880.0, "grad_norm": 2.1289663858381176, "language_loss": 0.75800139, "learning_rate": 1.0705005990112415e-06, "loss": 0.77969325, "num_input_tokens_seen": 118832195, "step": 5524, "time_per_iteration": 2.8236706256866455 }, { "auxiliary_loss_clip": 0.01105417, "auxiliary_loss_mlp": 0.01028357, "balance_loss_clip": 1.04472458, "balance_loss_mlp": 1.02125216, "epoch": 0.6643419707809776, "flos": 15377311958880.0, "grad_norm": 2.8719142014906036, "language_loss": 0.74293041, "learning_rate": 1.0698109352380957e-06, "loss": 0.76426816, "num_input_tokens_seen": 118849795, "step": 5525, "time_per_iteration": 3.715898036956787 }, { "auxiliary_loss_clip": 0.01174408, "auxiliary_loss_mlp": 0.01027052, "balance_loss_clip": 1.05102038, "balance_loss_mlp": 1.01975012, "epoch": 0.6644622136716166, "flos": 25117614756960.0, "grad_norm": 1.852135882466249, "language_loss": 0.77505755, "learning_rate": 1.0691214125738909e-06, "loss": 0.79707217, "num_input_tokens_seen": 118870000, "step": 5526, "time_per_iteration": 3.2163989543914795 }, { "auxiliary_loss_clip": 0.01070347, "auxiliary_loss_mlp": 0.0100224, "balance_loss_clip": 1.01129293, "balance_loss_mlp": 1.0010004, "epoch": 0.6645824565622558, "flos": 66201728822400.0, "grad_norm": 0.7835510324777872, "language_loss": 0.57483923, "learning_rate": 1.0684320311232287e-06, "loss": 0.59556508, "num_input_tokens_seen": 118932905, "step": 5527, "time_per_iteration": 3.127345085144043 }, { "auxiliary_loss_clip": 0.01142674, "auxiliary_loss_mlp": 0.01025761, "balance_loss_clip": 1.04735923, "balance_loss_mlp": 1.0182215, "epoch": 0.6647026994528948, "flos": 25082135188800.0, "grad_norm": 1.776910686465461, "language_loss": 0.8123312, "learning_rate": 1.0677427909906865e-06, "loss": 0.83401555, "num_input_tokens_seen": 118953355, "step": 5528, "time_per_iteration": 2.5845580101013184 }, { "auxiliary_loss_clip": 0.01179254, "auxiliary_loss_mlp": 0.01025335, "balance_loss_clip": 1.05227113, "balance_loss_mlp": 1.01696026, "epoch": 0.6648229423435339, "flos": 18222191158080.0, "grad_norm": 1.7325415085525917, "language_loss": 0.72212118, "learning_rate": 1.0670536922808216e-06, "loss": 0.74416709, "num_input_tokens_seen": 118973480, "step": 5529, "time_per_iteration": 2.4943294525146484 }, { "auxiliary_loss_clip": 0.01147908, "auxiliary_loss_mlp": 0.01020702, "balance_loss_clip": 1.04851937, "balance_loss_mlp": 1.01380587, "epoch": 0.6649431852341731, "flos": 18296885658240.0, "grad_norm": 2.3279942417175055, "language_loss": 0.71834892, "learning_rate": 1.06636473509817e-06, "loss": 0.740035, "num_input_tokens_seen": 118989860, "step": 5530, "time_per_iteration": 2.533167839050293 }, { "auxiliary_loss_clip": 0.01144962, "auxiliary_loss_mlp": 0.00762249, "balance_loss_clip": 1.04742789, "balance_loss_mlp": 1.00046742, "epoch": 0.6650634281248121, "flos": 17019578899200.0, "grad_norm": 2.880007512098417, "language_loss": 0.80946308, "learning_rate": 1.0656759195472447e-06, "loss": 0.82853526, "num_input_tokens_seen": 119007150, "step": 5531, "time_per_iteration": 2.529024839401245 }, { "auxiliary_loss_clip": 0.01047731, "auxiliary_loss_mlp": 0.01003886, "balance_loss_clip": 1.01060629, "balance_loss_mlp": 1.00257516, "epoch": 0.6651836710154512, "flos": 69294821330880.0, "grad_norm": 0.7705843815794495, "language_loss": 0.59788042, "learning_rate": 1.0649872457325414e-06, "loss": 0.61839658, "num_input_tokens_seen": 119068435, "step": 5532, "time_per_iteration": 3.0698580741882324 }, { "auxiliary_loss_clip": 0.01059955, "auxiliary_loss_mlp": 0.0100172, "balance_loss_clip": 1.0093224, "balance_loss_mlp": 1.00039053, "epoch": 0.6653039139060903, "flos": 66883455012960.0, "grad_norm": 0.8437194596179165, "language_loss": 0.55125207, "learning_rate": 1.0642987137585278e-06, "loss": 0.57186884, "num_input_tokens_seen": 119127960, "step": 5533, "time_per_iteration": 3.778890371322632 }, { "auxiliary_loss_clip": 0.01147075, "auxiliary_loss_mlp": 0.0102627, "balance_loss_clip": 1.04909539, "balance_loss_mlp": 1.01902461, "epoch": 0.6654241567967294, "flos": 21470060210880.0, "grad_norm": 1.6475625499599602, "language_loss": 0.8213774, "learning_rate": 1.0636103237296561e-06, "loss": 0.84311086, "num_input_tokens_seen": 119146885, "step": 5534, "time_per_iteration": 2.5265254974365234 }, { "auxiliary_loss_clip": 0.01160407, "auxiliary_loss_mlp": 0.01025414, "balance_loss_clip": 1.05320764, "balance_loss_mlp": 1.01869667, "epoch": 0.6655443996873684, "flos": 25119518355840.0, "grad_norm": 1.8795504200318347, "language_loss": 0.84055245, "learning_rate": 1.062922075750353e-06, "loss": 0.86241066, "num_input_tokens_seen": 119166900, "step": 5535, "time_per_iteration": 2.526592493057251 }, { "auxiliary_loss_clip": 0.01135536, "auxiliary_loss_mlp": 0.01022714, "balance_loss_clip": 1.04725671, "balance_loss_mlp": 1.0158329, "epoch": 0.6656646425780076, "flos": 17457329981760.0, "grad_norm": 1.9511232522529232, "language_loss": 0.72007632, "learning_rate": 1.0622339699250267e-06, "loss": 0.74165881, "num_input_tokens_seen": 119184820, "step": 5536, "time_per_iteration": 2.5174028873443604 }, { "auxiliary_loss_clip": 0.01133193, "auxiliary_loss_mlp": 0.01026753, "balance_loss_clip": 1.046808, "balance_loss_mlp": 1.01996386, "epoch": 0.6657848854686467, "flos": 23434193368800.0, "grad_norm": 1.9491131048613404, "language_loss": 0.79235816, "learning_rate": 1.0615460063580624e-06, "loss": 0.81395757, "num_input_tokens_seen": 119203295, "step": 5537, "time_per_iteration": 2.5749669075012207 }, { "auxiliary_loss_clip": 0.01150478, "auxiliary_loss_mlp": 0.01028791, "balance_loss_clip": 1.04955101, "balance_loss_mlp": 1.02113748, "epoch": 0.6659051283592857, "flos": 11509912113600.0, "grad_norm": 2.083848009192829, "language_loss": 0.72575998, "learning_rate": 1.060858185153821e-06, "loss": 0.74755263, "num_input_tokens_seen": 119221395, "step": 5538, "time_per_iteration": 2.5082061290740967 }, { "auxiliary_loss_clip": 0.01153896, "auxiliary_loss_mlp": 0.010305, "balance_loss_clip": 1.04948509, "balance_loss_mlp": 1.02238154, "epoch": 0.6660253712499249, "flos": 20594558045760.0, "grad_norm": 2.107515990097244, "language_loss": 0.76194984, "learning_rate": 1.0601705064166474e-06, "loss": 0.78379381, "num_input_tokens_seen": 119239790, "step": 5539, "time_per_iteration": 2.521519660949707 }, { "auxiliary_loss_clip": 0.01143149, "auxiliary_loss_mlp": 0.01025833, "balance_loss_clip": 1.04895401, "balance_loss_mlp": 1.01856756, "epoch": 0.666145614140564, "flos": 21251507922240.0, "grad_norm": 2.556144529722695, "language_loss": 0.73603904, "learning_rate": 1.0594829702508596e-06, "loss": 0.75772887, "num_input_tokens_seen": 119257505, "step": 5540, "time_per_iteration": 2.5179803371429443 }, { "auxiliary_loss_clip": 0.01135418, "auxiliary_loss_mlp": 0.01023814, "balance_loss_clip": 1.04471362, "balance_loss_mlp": 1.01625025, "epoch": 0.666265857031203, "flos": 33726192598560.0, "grad_norm": 1.5752203756552723, "language_loss": 0.54940653, "learning_rate": 1.0587955767607592e-06, "loss": 0.57099885, "num_input_tokens_seen": 119279365, "step": 5541, "time_per_iteration": 2.661531686782837 }, { "auxiliary_loss_clip": 0.01176283, "auxiliary_loss_mlp": 0.01027295, "balance_loss_clip": 1.05150151, "balance_loss_mlp": 1.01954031, "epoch": 0.6663860999218422, "flos": 17456647559520.0, "grad_norm": 2.03096812592203, "language_loss": 0.77563685, "learning_rate": 1.0581083260506206e-06, "loss": 0.79767257, "num_input_tokens_seen": 119296150, "step": 5542, "time_per_iteration": 2.4336729049682617 }, { "auxiliary_loss_clip": 0.01143837, "auxiliary_loss_mlp": 0.01022363, "balance_loss_clip": 1.04655504, "balance_loss_mlp": 1.01522577, "epoch": 0.6665063428124812, "flos": 17676744277440.0, "grad_norm": 1.982639171934644, "language_loss": 0.76514781, "learning_rate": 1.0574212182246993e-06, "loss": 0.78680986, "num_input_tokens_seen": 119314845, "step": 5543, "time_per_iteration": 2.54259991645813 }, { "auxiliary_loss_clip": 0.01150477, "auxiliary_loss_mlp": 0.01027881, "balance_loss_clip": 1.04630041, "balance_loss_mlp": 1.01993525, "epoch": 0.6666265857031203, "flos": 27673269866880.0, "grad_norm": 2.226382752077144, "language_loss": 0.75740492, "learning_rate": 1.0567342533872303e-06, "loss": 0.77918851, "num_input_tokens_seen": 119334875, "step": 5544, "time_per_iteration": 2.5794148445129395 }, { "auxiliary_loss_clip": 0.01148452, "auxiliary_loss_mlp": 0.01029395, "balance_loss_clip": 1.04943693, "balance_loss_mlp": 1.02205515, "epoch": 0.6667468285937594, "flos": 25046835205440.0, "grad_norm": 1.5427026632533065, "language_loss": 0.81061643, "learning_rate": 1.0560474316424255e-06, "loss": 0.8323949, "num_input_tokens_seen": 119354635, "step": 5545, "time_per_iteration": 2.5819222927093506 }, { "auxiliary_loss_clip": 0.01147317, "auxiliary_loss_mlp": 0.01024765, "balance_loss_clip": 1.04567266, "balance_loss_mlp": 1.0166527, "epoch": 0.6668670714843985, "flos": 22780476018720.0, "grad_norm": 2.808252486228811, "language_loss": 0.73775458, "learning_rate": 1.0553607530944746e-06, "loss": 0.75947535, "num_input_tokens_seen": 119372690, "step": 5546, "time_per_iteration": 2.5841970443725586 }, { "auxiliary_loss_clip": 0.01132975, "auxiliary_loss_mlp": 0.01023595, "balance_loss_clip": 1.04465806, "balance_loss_mlp": 1.01634717, "epoch": 0.6669873143750376, "flos": 22163890416960.0, "grad_norm": 6.251103799556546, "language_loss": 0.89409602, "learning_rate": 1.0546742178475463e-06, "loss": 0.91566169, "num_input_tokens_seen": 119391685, "step": 5547, "time_per_iteration": 2.562457323074341 }, { "auxiliary_loss_clip": 0.01121363, "auxiliary_loss_mlp": 0.01023234, "balance_loss_clip": 1.04794741, "balance_loss_mlp": 1.01655817, "epoch": 0.6671075572656767, "flos": 20514835171200.0, "grad_norm": 2.170417261148485, "language_loss": 0.86498141, "learning_rate": 1.0539878260057868e-06, "loss": 0.88642728, "num_input_tokens_seen": 119410725, "step": 5548, "time_per_iteration": 3.4303503036499023 }, { "auxiliary_loss_clip": 0.01163643, "auxiliary_loss_mlp": 0.01026381, "balance_loss_clip": 1.05074239, "balance_loss_mlp": 1.01896, "epoch": 0.6672278001563158, "flos": 17931207137760.0, "grad_norm": 2.5481863149636528, "language_loss": 0.68516582, "learning_rate": 1.0533015776733226e-06, "loss": 0.70706606, "num_input_tokens_seen": 119426875, "step": 5549, "time_per_iteration": 2.484386444091797 }, { "auxiliary_loss_clip": 0.01141798, "auxiliary_loss_mlp": 0.01032098, "balance_loss_clip": 1.04700696, "balance_loss_mlp": 1.02487969, "epoch": 0.6673480430469548, "flos": 22342150264800.0, "grad_norm": 2.265692662097166, "language_loss": 0.78779131, "learning_rate": 1.0526154729542566e-06, "loss": 0.80953026, "num_input_tokens_seen": 119446935, "step": 5550, "time_per_iteration": 2.55936598777771 }, { "auxiliary_loss_clip": 0.01133528, "auxiliary_loss_mlp": 0.01029604, "balance_loss_clip": 1.04762006, "balance_loss_mlp": 1.0223856, "epoch": 0.6674682859375939, "flos": 20703834190080.0, "grad_norm": 2.3301834875267735, "language_loss": 0.79940301, "learning_rate": 1.0519295119526699e-06, "loss": 0.82103431, "num_input_tokens_seen": 119463240, "step": 5551, "time_per_iteration": 3.325714111328125 }, { "auxiliary_loss_clip": 0.01151486, "auxiliary_loss_mlp": 0.01022627, "balance_loss_clip": 1.04932106, "balance_loss_mlp": 1.01534963, "epoch": 0.667588528828233, "flos": 26206676753280.0, "grad_norm": 1.5928704622434013, "language_loss": 0.82974875, "learning_rate": 1.0512436947726227e-06, "loss": 0.8514899, "num_input_tokens_seen": 119484655, "step": 5552, "time_per_iteration": 2.569164276123047 }, { "auxiliary_loss_clip": 0.01133884, "auxiliary_loss_mlp": 0.01027616, "balance_loss_clip": 1.04596114, "balance_loss_mlp": 1.02007055, "epoch": 0.6677087717188721, "flos": 23071029035520.0, "grad_norm": 2.8798934401861196, "language_loss": 0.65273595, "learning_rate": 1.0505580215181517e-06, "loss": 0.67435098, "num_input_tokens_seen": 119502895, "step": 5553, "time_per_iteration": 3.2742412090301514 }, { "auxiliary_loss_clip": 0.01034678, "auxiliary_loss_mlp": 0.0100353, "balance_loss_clip": 1.0145272, "balance_loss_mlp": 1.00234354, "epoch": 0.6678290146095112, "flos": 70941326472480.0, "grad_norm": 0.7795650154711927, "language_loss": 0.5665167, "learning_rate": 1.0498724922932753e-06, "loss": 0.58689874, "num_input_tokens_seen": 119561010, "step": 5554, "time_per_iteration": 3.0715599060058594 }, { "auxiliary_loss_clip": 0.0118148, "auxiliary_loss_mlp": 0.01030992, "balance_loss_clip": 1.05450559, "balance_loss_mlp": 1.0234344, "epoch": 0.6679492575001503, "flos": 18661091583360.0, "grad_norm": 2.565045777074016, "language_loss": 0.86466664, "learning_rate": 1.0491871072019851e-06, "loss": 0.88679135, "num_input_tokens_seen": 119578900, "step": 5555, "time_per_iteration": 2.4643046855926514 }, { "auxiliary_loss_clip": 0.01134636, "auxiliary_loss_mlp": 0.01024989, "balance_loss_clip": 1.04322183, "balance_loss_mlp": 1.01819086, "epoch": 0.6680695003907894, "flos": 29711989774080.0, "grad_norm": 1.6560522423577704, "language_loss": 0.63726133, "learning_rate": 1.0485018663482555e-06, "loss": 0.65885758, "num_input_tokens_seen": 119598920, "step": 5556, "time_per_iteration": 2.6111345291137695 }, { "auxiliary_loss_clip": 0.01156486, "auxiliary_loss_mlp": 0.0102838, "balance_loss_clip": 1.04873395, "balance_loss_mlp": 1.02050662, "epoch": 0.6681897432814284, "flos": 28218968166240.0, "grad_norm": 2.6740809133448717, "language_loss": 0.70092177, "learning_rate": 1.0478167698360354e-06, "loss": 0.72277045, "num_input_tokens_seen": 119618220, "step": 5557, "time_per_iteration": 2.573827028274536 }, { "auxiliary_loss_clip": 0.01154647, "auxiliary_loss_mlp": 0.01026374, "balance_loss_clip": 1.0479542, "balance_loss_mlp": 1.01931691, "epoch": 0.6683099861720676, "flos": 25046547869760.0, "grad_norm": 2.2292826857258827, "language_loss": 0.70123816, "learning_rate": 1.0471318177692556e-06, "loss": 0.72304833, "num_input_tokens_seen": 119638520, "step": 5558, "time_per_iteration": 2.514624834060669 }, { "auxiliary_loss_clip": 0.01121463, "auxiliary_loss_mlp": 0.01024104, "balance_loss_clip": 1.0445236, "balance_loss_mlp": 1.01698422, "epoch": 0.6684302290627067, "flos": 22996981040640.0, "grad_norm": 2.463722933706117, "language_loss": 0.75772667, "learning_rate": 1.046447010251821e-06, "loss": 0.77918231, "num_input_tokens_seen": 119655850, "step": 5559, "time_per_iteration": 3.3708248138427734 }, { "auxiliary_loss_clip": 0.01146478, "auxiliary_loss_mlp": 0.01025121, "balance_loss_clip": 1.05032206, "balance_loss_mlp": 1.01825464, "epoch": 0.6685504719533457, "flos": 26573827869120.0, "grad_norm": 1.7566966557415238, "language_loss": 0.75567842, "learning_rate": 1.0457623473876157e-06, "loss": 0.77739441, "num_input_tokens_seen": 119675355, "step": 5560, "time_per_iteration": 2.574012279510498 }, { "auxiliary_loss_clip": 0.01172833, "auxiliary_loss_mlp": 0.01026182, "balance_loss_clip": 1.04955339, "balance_loss_mlp": 1.01895475, "epoch": 0.6686707148439849, "flos": 28986092111040.0, "grad_norm": 1.6774019150972346, "language_loss": 0.71105587, "learning_rate": 1.0450778292805046e-06, "loss": 0.73304605, "num_input_tokens_seen": 119695340, "step": 5561, "time_per_iteration": 2.513742208480835 }, { "auxiliary_loss_clip": 0.01163055, "auxiliary_loss_mlp": 0.01023435, "balance_loss_clip": 1.04843056, "balance_loss_mlp": 1.01621652, "epoch": 0.6687909577346239, "flos": 23623156470720.0, "grad_norm": 1.5318908184871427, "language_loss": 0.78718543, "learning_rate": 1.0443934560343267e-06, "loss": 0.80905032, "num_input_tokens_seen": 119716750, "step": 5562, "time_per_iteration": 2.5358896255493164 }, { "auxiliary_loss_clip": 0.01120285, "auxiliary_loss_mlp": 0.0103044, "balance_loss_clip": 1.04472196, "balance_loss_mlp": 1.02312088, "epoch": 0.668911200625263, "flos": 23148596892480.0, "grad_norm": 1.9473439436874729, "language_loss": 0.78261721, "learning_rate": 1.0437092277529034e-06, "loss": 0.80412447, "num_input_tokens_seen": 119736005, "step": 5563, "time_per_iteration": 2.558422803878784 }, { "auxiliary_loss_clip": 0.0114386, "auxiliary_loss_mlp": 0.01023855, "balance_loss_clip": 1.04779422, "balance_loss_mlp": 1.016855, "epoch": 0.6690314435159022, "flos": 18551923189920.0, "grad_norm": 1.871170957324192, "language_loss": 0.73325372, "learning_rate": 1.0430251445400292e-06, "loss": 0.75493085, "num_input_tokens_seen": 119754050, "step": 5564, "time_per_iteration": 2.520850896835327 }, { "auxiliary_loss_clip": 0.01077107, "auxiliary_loss_mlp": 0.0102763, "balance_loss_clip": 1.04329181, "balance_loss_mlp": 1.02032268, "epoch": 0.6691516864065412, "flos": 31759545253440.0, "grad_norm": 2.216598900513627, "language_loss": 0.62687653, "learning_rate": 1.0423412064994787e-06, "loss": 0.64792389, "num_input_tokens_seen": 119774820, "step": 5565, "time_per_iteration": 3.04144549369812 }, { "auxiliary_loss_clip": 0.01132801, "auxiliary_loss_mlp": 0.01027035, "balance_loss_clip": 1.04472923, "balance_loss_mlp": 1.02006125, "epoch": 0.6692719292971803, "flos": 34933869148800.0, "grad_norm": 1.9044402822737316, "language_loss": 0.73813742, "learning_rate": 1.0416574137350064e-06, "loss": 0.75973576, "num_input_tokens_seen": 119795525, "step": 5566, "time_per_iteration": 2.984703540802002 }, { "auxiliary_loss_clip": 0.01153243, "auxiliary_loss_mlp": 0.01024862, "balance_loss_clip": 1.0478096, "balance_loss_mlp": 1.01724434, "epoch": 0.6693921721878194, "flos": 20449191744960.0, "grad_norm": 2.075981802279214, "language_loss": 0.80783504, "learning_rate": 1.0409737663503428e-06, "loss": 0.82961613, "num_input_tokens_seen": 119813905, "step": 5567, "time_per_iteration": 2.508213996887207 }, { "auxiliary_loss_clip": 0.0115718, "auxiliary_loss_mlp": 0.01030365, "balance_loss_clip": 1.04591751, "balance_loss_mlp": 1.02307558, "epoch": 0.6695124150784585, "flos": 16614541778880.0, "grad_norm": 1.771373693353358, "language_loss": 0.82562995, "learning_rate": 1.040290264449196e-06, "loss": 0.84750545, "num_input_tokens_seen": 119832010, "step": 5568, "time_per_iteration": 2.519500255584717 }, { "auxiliary_loss_clip": 0.01155105, "auxiliary_loss_mlp": 0.01025174, "balance_loss_clip": 1.04913831, "balance_loss_mlp": 1.01799214, "epoch": 0.6696326579690975, "flos": 26652149982240.0, "grad_norm": 2.037410776764049, "language_loss": 0.63997531, "learning_rate": 1.0396069081352532e-06, "loss": 0.66177809, "num_input_tokens_seen": 119851165, "step": 5569, "time_per_iteration": 2.544234037399292 }, { "auxiliary_loss_clip": 0.0106932, "auxiliary_loss_mlp": 0.01000477, "balance_loss_clip": 1.01040316, "balance_loss_mlp": 0.99915963, "epoch": 0.6697529008597367, "flos": 66964614565920.0, "grad_norm": 0.7824301581782906, "language_loss": 0.56071413, "learning_rate": 1.0389236975121782e-06, "loss": 0.58141208, "num_input_tokens_seen": 119906015, "step": 5570, "time_per_iteration": 2.9892544746398926 }, { "auxiliary_loss_clip": 0.01177928, "auxiliary_loss_mlp": 0.01028656, "balance_loss_clip": 1.05229092, "balance_loss_mlp": 1.02144957, "epoch": 0.6698731437503758, "flos": 20886942827520.0, "grad_norm": 2.2582477292953893, "language_loss": 0.7124089, "learning_rate": 1.0382406326836147e-06, "loss": 0.73447478, "num_input_tokens_seen": 119925160, "step": 5571, "time_per_iteration": 2.457343578338623 }, { "auxiliary_loss_clip": 0.01166539, "auxiliary_loss_mlp": 0.01026872, "balance_loss_clip": 1.05017292, "balance_loss_mlp": 1.01891434, "epoch": 0.6699933866410148, "flos": 20409473975520.0, "grad_norm": 2.2913359520886183, "language_loss": 0.76043159, "learning_rate": 1.0375577137531828e-06, "loss": 0.78236568, "num_input_tokens_seen": 119943720, "step": 5572, "time_per_iteration": 2.514065980911255 }, { "auxiliary_loss_clip": 0.01150543, "auxiliary_loss_mlp": 0.01026344, "balance_loss_clip": 1.04896092, "balance_loss_mlp": 1.0196327, "epoch": 0.670113629531654, "flos": 29023080191520.0, "grad_norm": 1.5550238801129046, "language_loss": 0.71983939, "learning_rate": 1.0368749408244802e-06, "loss": 0.74160826, "num_input_tokens_seen": 119966640, "step": 5573, "time_per_iteration": 2.582146644592285 }, { "auxiliary_loss_clip": 0.01154534, "auxiliary_loss_mlp": 0.01031526, "balance_loss_clip": 1.04836476, "balance_loss_mlp": 1.02429307, "epoch": 0.670233872422293, "flos": 19791703114080.0, "grad_norm": 1.7148261086710355, "language_loss": 0.78895032, "learning_rate": 1.0361923140010836e-06, "loss": 0.81081092, "num_input_tokens_seen": 119985125, "step": 5574, "time_per_iteration": 2.483153820037842 }, { "auxiliary_loss_clip": 0.01165555, "auxiliary_loss_mlp": 0.01028093, "balance_loss_clip": 1.04913878, "balance_loss_mlp": 1.02100039, "epoch": 0.6703541153129321, "flos": 24243692938080.0, "grad_norm": 2.168284771298858, "language_loss": 0.63300729, "learning_rate": 1.0355098333865455e-06, "loss": 0.65494382, "num_input_tokens_seen": 120004355, "step": 5575, "time_per_iteration": 3.288625955581665 }, { "auxiliary_loss_clip": 0.01160827, "auxiliary_loss_mlp": 0.0103179, "balance_loss_clip": 1.05340981, "balance_loss_mlp": 1.02505422, "epoch": 0.6704743582035713, "flos": 26688527474400.0, "grad_norm": 1.5754054078305273, "language_loss": 0.69311488, "learning_rate": 1.0348274990844006e-06, "loss": 0.71504104, "num_input_tokens_seen": 120027115, "step": 5576, "time_per_iteration": 2.5417447090148926 }, { "auxiliary_loss_clip": 0.01160633, "auxiliary_loss_mlp": 0.01030221, "balance_loss_clip": 1.05044127, "balance_loss_mlp": 1.02327657, "epoch": 0.6705946010942103, "flos": 23514383163840.0, "grad_norm": 2.326649939396135, "language_loss": 0.72798383, "learning_rate": 1.034145311198155e-06, "loss": 0.74989235, "num_input_tokens_seen": 120047130, "step": 5577, "time_per_iteration": 3.2979605197906494 }, { "auxiliary_loss_clip": 0.01170439, "auxiliary_loss_mlp": 0.01026248, "balance_loss_clip": 1.0484364, "balance_loss_mlp": 1.0193429, "epoch": 0.6707148439848494, "flos": 24061015304160.0, "grad_norm": 1.629149799006186, "language_loss": 0.63811517, "learning_rate": 1.0334632698312989e-06, "loss": 0.66008198, "num_input_tokens_seen": 120067925, "step": 5578, "time_per_iteration": 3.213024616241455 }, { "auxiliary_loss_clip": 0.01139009, "auxiliary_loss_mlp": 0.01030549, "balance_loss_clip": 1.04638767, "balance_loss_mlp": 1.02333081, "epoch": 0.6708350868754885, "flos": 22528671013440.0, "grad_norm": 2.4824523587314173, "language_loss": 0.75352871, "learning_rate": 1.032781375087295e-06, "loss": 0.77522427, "num_input_tokens_seen": 120087825, "step": 5579, "time_per_iteration": 2.5309224128723145 }, { "auxiliary_loss_clip": 0.01150004, "auxiliary_loss_mlp": 0.01028519, "balance_loss_clip": 1.05064046, "balance_loss_mlp": 1.02154279, "epoch": 0.6709553297661276, "flos": 25227752908320.0, "grad_norm": 1.4107057758514063, "language_loss": 0.67636418, "learning_rate": 1.0320996270695891e-06, "loss": 0.69814938, "num_input_tokens_seen": 120108895, "step": 5580, "time_per_iteration": 2.559951066970825 }, { "auxiliary_loss_clip": 0.01130451, "auxiliary_loss_mlp": 0.01026077, "balance_loss_clip": 1.0431428, "balance_loss_mlp": 1.01893091, "epoch": 0.6710755726567667, "flos": 20448760741440.0, "grad_norm": 1.8602484086056208, "language_loss": 0.73759329, "learning_rate": 1.0314180258815998e-06, "loss": 0.75915849, "num_input_tokens_seen": 120127535, "step": 5581, "time_per_iteration": 2.5698273181915283 }, { "auxiliary_loss_clip": 0.01120274, "auxiliary_loss_mlp": 0.01025768, "balance_loss_clip": 1.0420692, "balance_loss_mlp": 1.01883864, "epoch": 0.6711958155474057, "flos": 25995415607520.0, "grad_norm": 1.9194444329800022, "language_loss": 0.74153751, "learning_rate": 1.0307365716267247e-06, "loss": 0.76299798, "num_input_tokens_seen": 120147980, "step": 5582, "time_per_iteration": 2.5973427295684814 }, { "auxiliary_loss_clip": 0.01160707, "auxiliary_loss_mlp": 0.01028788, "balance_loss_clip": 1.04958284, "balance_loss_mlp": 1.02122426, "epoch": 0.6713160584380449, "flos": 19937715920160.0, "grad_norm": 2.0970416339730575, "language_loss": 0.77903879, "learning_rate": 1.0300552644083423e-06, "loss": 0.80093366, "num_input_tokens_seen": 120166905, "step": 5583, "time_per_iteration": 2.4888973236083984 }, { "auxiliary_loss_clip": 0.01138802, "auxiliary_loss_mlp": 0.01029327, "balance_loss_clip": 1.04892731, "balance_loss_mlp": 1.0217129, "epoch": 0.6714363013286839, "flos": 18223376417760.0, "grad_norm": 2.588769741305662, "language_loss": 0.72194922, "learning_rate": 1.0293741043298036e-06, "loss": 0.74363053, "num_input_tokens_seen": 120185255, "step": 5584, "time_per_iteration": 2.522615432739258 }, { "auxiliary_loss_clip": 0.01140909, "auxiliary_loss_mlp": 0.01025815, "balance_loss_clip": 1.05214369, "balance_loss_mlp": 1.01846528, "epoch": 0.671556544219323, "flos": 25812378804000.0, "grad_norm": 2.3603315286104953, "language_loss": 0.71162295, "learning_rate": 1.0286930914944436e-06, "loss": 0.7332902, "num_input_tokens_seen": 120205070, "step": 5585, "time_per_iteration": 3.3013715744018555 }, { "auxiliary_loss_clip": 0.0117368, "auxiliary_loss_mlp": 0.01028516, "balance_loss_clip": 1.04830933, "balance_loss_mlp": 1.02076161, "epoch": 0.6716767871099621, "flos": 15850434858720.0, "grad_norm": 2.3687795981445796, "language_loss": 0.77564055, "learning_rate": 1.0280122260055684e-06, "loss": 0.7976625, "num_input_tokens_seen": 120220780, "step": 5586, "time_per_iteration": 2.439286231994629 }, { "auxiliary_loss_clip": 0.01178642, "auxiliary_loss_mlp": 0.01030616, "balance_loss_clip": 1.05255115, "balance_loss_mlp": 1.02332664, "epoch": 0.6717970300006012, "flos": 19756115795040.0, "grad_norm": 2.031276036939015, "language_loss": 0.81928289, "learning_rate": 1.0273315079664652e-06, "loss": 0.84137541, "num_input_tokens_seen": 120238735, "step": 5587, "time_per_iteration": 2.4595582485198975 }, { "auxiliary_loss_clip": 0.01164056, "auxiliary_loss_mlp": 0.01025502, "balance_loss_clip": 1.04984426, "balance_loss_mlp": 1.01860905, "epoch": 0.6719172728912403, "flos": 25485053208480.0, "grad_norm": 2.1660371047345786, "language_loss": 0.74102211, "learning_rate": 1.0266509374803992e-06, "loss": 0.76291764, "num_input_tokens_seen": 120259895, "step": 5588, "time_per_iteration": 2.5380733013153076 }, { "auxiliary_loss_clip": 0.01176174, "auxiliary_loss_mlp": 0.00762636, "balance_loss_clip": 1.05179071, "balance_loss_mlp": 1.00049067, "epoch": 0.6720375157818794, "flos": 15880347298080.0, "grad_norm": 2.4486252187399646, "language_loss": 0.84375942, "learning_rate": 1.0259705146506123e-06, "loss": 0.8631475, "num_input_tokens_seen": 120274790, "step": 5589, "time_per_iteration": 2.4343605041503906 }, { "auxiliary_loss_clip": 0.01164065, "auxiliary_loss_mlp": 0.01023898, "balance_loss_clip": 1.04997921, "balance_loss_mlp": 1.01642704, "epoch": 0.6721577586725185, "flos": 32010847421280.0, "grad_norm": 1.9313342018212336, "language_loss": 0.77869701, "learning_rate": 1.025290239580324e-06, "loss": 0.80057669, "num_input_tokens_seen": 120295460, "step": 5590, "time_per_iteration": 2.5698776245117188 }, { "auxiliary_loss_clip": 0.01116931, "auxiliary_loss_mlp": 0.01028226, "balance_loss_clip": 1.04305005, "balance_loss_mlp": 1.02127647, "epoch": 0.6722780015631575, "flos": 20737877079840.0, "grad_norm": 3.6158439650322967, "language_loss": 0.75590396, "learning_rate": 1.0246101123727313e-06, "loss": 0.77735555, "num_input_tokens_seen": 120314440, "step": 5591, "time_per_iteration": 2.5926339626312256 }, { "auxiliary_loss_clip": 0.01160484, "auxiliary_loss_mlp": 0.01023849, "balance_loss_clip": 1.04814494, "balance_loss_mlp": 1.01725996, "epoch": 0.6723982444537967, "flos": 16909620332640.0, "grad_norm": 1.9042232775844434, "language_loss": 0.78686368, "learning_rate": 1.0239301331310085e-06, "loss": 0.808707, "num_input_tokens_seen": 120332060, "step": 5592, "time_per_iteration": 2.460618019104004 }, { "auxiliary_loss_clip": 0.01156248, "auxiliary_loss_mlp": 0.01033633, "balance_loss_clip": 1.04748392, "balance_loss_mlp": 1.02678728, "epoch": 0.6725184873444358, "flos": 20667815867520.0, "grad_norm": 2.0576243824164098, "language_loss": 0.88499177, "learning_rate": 1.0232503019583088e-06, "loss": 0.90689051, "num_input_tokens_seen": 120351670, "step": 5593, "time_per_iteration": 2.491070032119751 }, { "auxiliary_loss_clip": 0.0115614, "auxiliary_loss_mlp": 0.01026518, "balance_loss_clip": 1.0482384, "balance_loss_mlp": 1.01937139, "epoch": 0.6726387302350748, "flos": 23727619742400.0, "grad_norm": 2.051059871214126, "language_loss": 0.69739413, "learning_rate": 1.0225706189577619e-06, "loss": 0.71922076, "num_input_tokens_seen": 120370195, "step": 5594, "time_per_iteration": 2.4959752559661865 }, { "auxiliary_loss_clip": 0.01162716, "auxiliary_loss_mlp": 0.01027097, "balance_loss_clip": 1.05032825, "balance_loss_mlp": 1.01986361, "epoch": 0.672758973125714, "flos": 15188277023040.0, "grad_norm": 3.773377538525824, "language_loss": 0.74818611, "learning_rate": 1.021891084232475e-06, "loss": 0.77008426, "num_input_tokens_seen": 120388130, "step": 5595, "time_per_iteration": 2.4869308471679688 }, { "auxiliary_loss_clip": 0.01158517, "auxiliary_loss_mlp": 0.01029719, "balance_loss_clip": 1.04772496, "balance_loss_mlp": 1.0220803, "epoch": 0.672879216016353, "flos": 18077255860800.0, "grad_norm": 2.017766529702513, "language_loss": 0.79999077, "learning_rate": 1.0212116978855325e-06, "loss": 0.82187307, "num_input_tokens_seen": 120406145, "step": 5596, "time_per_iteration": 2.459171772003174 }, { "auxiliary_loss_clip": 0.01133152, "auxiliary_loss_mlp": 0.01026937, "balance_loss_clip": 1.04715288, "balance_loss_mlp": 1.02010298, "epoch": 0.6729994589069921, "flos": 23476353491520.0, "grad_norm": 1.6835871040103798, "language_loss": 0.78852963, "learning_rate": 1.020532460019997e-06, "loss": 0.81013048, "num_input_tokens_seen": 120425395, "step": 5597, "time_per_iteration": 2.5665435791015625 }, { "auxiliary_loss_clip": 0.0109661, "auxiliary_loss_mlp": 0.01027488, "balance_loss_clip": 1.04554462, "balance_loss_mlp": 1.02035642, "epoch": 0.6731197017976313, "flos": 26322022863840.0, "grad_norm": 1.7211065782847224, "language_loss": 0.71033823, "learning_rate": 1.0198533707389096e-06, "loss": 0.73157924, "num_input_tokens_seen": 120446270, "step": 5598, "time_per_iteration": 2.785954475402832 }, { "auxiliary_loss_clip": 0.01158103, "auxiliary_loss_mlp": 0.00762263, "balance_loss_clip": 1.04922247, "balance_loss_mlp": 1.0004319, "epoch": 0.6732399446882703, "flos": 21616432186560.0, "grad_norm": 2.5987752438087437, "language_loss": 0.72828561, "learning_rate": 1.0191744301452853e-06, "loss": 0.74748927, "num_input_tokens_seen": 120465570, "step": 5599, "time_per_iteration": 2.678325891494751 }, { "auxiliary_loss_clip": 0.01172313, "auxiliary_loss_mlp": 0.01024328, "balance_loss_clip": 1.04921055, "balance_loss_mlp": 1.01710355, "epoch": 0.6733601875789094, "flos": 25880177247840.0, "grad_norm": 1.590292084684059, "language_loss": 0.70291191, "learning_rate": 1.0184956383421208e-06, "loss": 0.72487831, "num_input_tokens_seen": 120484220, "step": 5600, "time_per_iteration": 3.286966562271118 }, { "auxiliary_loss_clip": 0.01164196, "auxiliary_loss_mlp": 0.01029482, "balance_loss_clip": 1.04976046, "balance_loss_mlp": 1.02255273, "epoch": 0.6734804304695485, "flos": 22929577683360.0, "grad_norm": 2.5692755575502106, "language_loss": 0.65391225, "learning_rate": 1.017816995432387e-06, "loss": 0.67584902, "num_input_tokens_seen": 120503320, "step": 5601, "time_per_iteration": 2.5193588733673096 }, { "auxiliary_loss_clip": 0.01146379, "auxiliary_loss_mlp": 0.01026237, "balance_loss_clip": 1.04759657, "balance_loss_mlp": 1.01929307, "epoch": 0.6736006733601876, "flos": 18697971912960.0, "grad_norm": 2.0936685346129527, "language_loss": 0.74173498, "learning_rate": 1.0171385015190353e-06, "loss": 0.76346123, "num_input_tokens_seen": 120523180, "step": 5602, "time_per_iteration": 3.3355491161346436 }, { "auxiliary_loss_clip": 0.01142497, "auxiliary_loss_mlp": 0.00761776, "balance_loss_clip": 1.05115175, "balance_loss_mlp": 1.00044656, "epoch": 0.6737209162508266, "flos": 19427748607680.0, "grad_norm": 2.155990557445812, "language_loss": 0.7290619, "learning_rate": 1.0164601567049908e-06, "loss": 0.74810463, "num_input_tokens_seen": 120541710, "step": 5603, "time_per_iteration": 2.5148117542266846 }, { "auxiliary_loss_clip": 0.01146468, "auxiliary_loss_mlp": 0.01029994, "balance_loss_clip": 1.04877746, "balance_loss_mlp": 1.02319646, "epoch": 0.6738411591414658, "flos": 20158064056800.0, "grad_norm": 1.874898368254317, "language_loss": 0.80446243, "learning_rate": 1.015781961093158e-06, "loss": 0.82622707, "num_input_tokens_seen": 120561030, "step": 5604, "time_per_iteration": 3.4963042736053467 }, { "auxiliary_loss_clip": 0.01148731, "auxiliary_loss_mlp": 0.01025748, "balance_loss_clip": 1.04518366, "balance_loss_mlp": 1.01897645, "epoch": 0.6739614020321049, "flos": 21653851270560.0, "grad_norm": 1.5118299179732677, "language_loss": 0.77476227, "learning_rate": 1.0151039147864197e-06, "loss": 0.79650706, "num_input_tokens_seen": 120581005, "step": 5605, "time_per_iteration": 2.6117284297943115 }, { "auxiliary_loss_clip": 0.01090872, "auxiliary_loss_mlp": 0.01031655, "balance_loss_clip": 1.04989696, "balance_loss_mlp": 1.02401376, "epoch": 0.6740816449227439, "flos": 19171705401120.0, "grad_norm": 1.9684847872238929, "language_loss": 0.65683687, "learning_rate": 1.0144260178876336e-06, "loss": 0.67806208, "num_input_tokens_seen": 120600350, "step": 5606, "time_per_iteration": 2.7049002647399902 }, { "auxiliary_loss_clip": 0.01150282, "auxiliary_loss_mlp": 0.01029083, "balance_loss_clip": 1.04672515, "balance_loss_mlp": 1.02226448, "epoch": 0.6742018878133831, "flos": 21097018713600.0, "grad_norm": 2.065599045592197, "language_loss": 0.66954064, "learning_rate": 1.0137482704996388e-06, "loss": 0.69133431, "num_input_tokens_seen": 120614700, "step": 5607, "time_per_iteration": 2.533931016921997 }, { "auxiliary_loss_clip": 0.01137737, "auxiliary_loss_mlp": 0.01026162, "balance_loss_clip": 1.04793072, "balance_loss_mlp": 1.01888418, "epoch": 0.6743221307040221, "flos": 23549970482880.0, "grad_norm": 1.9297234816178672, "language_loss": 0.78704143, "learning_rate": 1.0130706727252461e-06, "loss": 0.80868042, "num_input_tokens_seen": 120631755, "step": 5608, "time_per_iteration": 2.5532565116882324 }, { "auxiliary_loss_clip": 0.01138881, "auxiliary_loss_mlp": 0.01027116, "balance_loss_clip": 1.04873323, "balance_loss_mlp": 1.01976418, "epoch": 0.6744423735946612, "flos": 16249545680640.0, "grad_norm": 2.2548415311933487, "language_loss": 0.68052006, "learning_rate": 1.0123932246672468e-06, "loss": 0.70218003, "num_input_tokens_seen": 120645900, "step": 5609, "time_per_iteration": 2.5373411178588867 }, { "auxiliary_loss_clip": 0.01027694, "auxiliary_loss_mlp": 0.00752743, "balance_loss_clip": 1.01104927, "balance_loss_mlp": 1.00016809, "epoch": 0.6745626164853004, "flos": 57843267888960.0, "grad_norm": 0.7510334668656982, "language_loss": 0.55878496, "learning_rate": 1.0117159264284114e-06, "loss": 0.57658935, "num_input_tokens_seen": 120709070, "step": 5610, "time_per_iteration": 3.133739471435547 }, { "auxiliary_loss_clip": 0.01149289, "auxiliary_loss_mlp": 0.01024959, "balance_loss_clip": 1.04826903, "balance_loss_mlp": 1.01781809, "epoch": 0.6746828593759394, "flos": 20485030482720.0, "grad_norm": 1.6316509352131954, "language_loss": 0.76931417, "learning_rate": 1.0110387781114837e-06, "loss": 0.79105663, "num_input_tokens_seen": 120727685, "step": 5611, "time_per_iteration": 3.463400363922119 }, { "auxiliary_loss_clip": 0.01173794, "auxiliary_loss_mlp": 0.0102736, "balance_loss_clip": 1.04982412, "balance_loss_mlp": 1.02011168, "epoch": 0.6748031022665785, "flos": 19208226561120.0, "grad_norm": 1.977900818165345, "language_loss": 0.77368546, "learning_rate": 1.0103617798191872e-06, "loss": 0.79569697, "num_input_tokens_seen": 120747160, "step": 5612, "time_per_iteration": 2.4683594703674316 }, { "auxiliary_loss_clip": 0.01145573, "auxiliary_loss_mlp": 0.01025654, "balance_loss_clip": 1.04948878, "balance_loss_mlp": 1.0184207, "epoch": 0.6749233451572175, "flos": 15195029411520.0, "grad_norm": 2.4897724619195167, "language_loss": 0.83073497, "learning_rate": 1.0096849316542217e-06, "loss": 0.85244727, "num_input_tokens_seen": 120763710, "step": 5613, "time_per_iteration": 2.491856336593628 }, { "auxiliary_loss_clip": 0.0107519, "auxiliary_loss_mlp": 0.01027066, "balance_loss_clip": 1.03800666, "balance_loss_mlp": 1.0193119, "epoch": 0.6750435880478567, "flos": 26499492538560.0, "grad_norm": 2.310653323605157, "language_loss": 0.74506193, "learning_rate": 1.0090082337192643e-06, "loss": 0.76608455, "num_input_tokens_seen": 120783355, "step": 5614, "time_per_iteration": 2.6649892330169678 }, { "auxiliary_loss_clip": 0.01098207, "auxiliary_loss_mlp": 0.01024171, "balance_loss_clip": 1.03909588, "balance_loss_mlp": 1.0172416, "epoch": 0.6751638309384957, "flos": 23404316846400.0, "grad_norm": 2.9462621876011106, "language_loss": 0.78511006, "learning_rate": 1.0083316861169705e-06, "loss": 0.80633378, "num_input_tokens_seen": 120802090, "step": 5615, "time_per_iteration": 2.6329028606414795 }, { "auxiliary_loss_clip": 0.01136937, "auxiliary_loss_mlp": 0.01025164, "balance_loss_clip": 1.04432261, "balance_loss_mlp": 1.01745069, "epoch": 0.6752840738291348, "flos": 23441412677760.0, "grad_norm": 2.373055927728748, "language_loss": 0.71539873, "learning_rate": 1.0076552889499713e-06, "loss": 0.73701978, "num_input_tokens_seen": 120822855, "step": 5616, "time_per_iteration": 2.6247642040252686 }, { "auxiliary_loss_clip": 0.01161735, "auxiliary_loss_mlp": 0.01025638, "balance_loss_clip": 1.0524056, "balance_loss_mlp": 1.01837862, "epoch": 0.675404316719774, "flos": 30335830601760.0, "grad_norm": 2.5953299272007047, "language_loss": 0.7361573, "learning_rate": 1.006979042320876e-06, "loss": 0.75803101, "num_input_tokens_seen": 120843070, "step": 5617, "time_per_iteration": 2.5678625106811523 }, { "auxiliary_loss_clip": 0.01140372, "auxiliary_loss_mlp": 0.0102517, "balance_loss_clip": 1.04413795, "balance_loss_mlp": 1.01830649, "epoch": 0.675524559610413, "flos": 23622617716320.0, "grad_norm": 2.04684585747578, "language_loss": 0.63278723, "learning_rate": 1.0063029463322702e-06, "loss": 0.65444267, "num_input_tokens_seen": 120863345, "step": 5618, "time_per_iteration": 2.557814598083496 }, { "auxiliary_loss_clip": 0.01112749, "auxiliary_loss_mlp": 0.00762725, "balance_loss_clip": 1.04310107, "balance_loss_mlp": 1.00044894, "epoch": 0.6756448025010521, "flos": 21248634565440.0, "grad_norm": 1.9138177746741598, "language_loss": 0.75117981, "learning_rate": 1.0056270010867164e-06, "loss": 0.76993454, "num_input_tokens_seen": 120880915, "step": 5619, "time_per_iteration": 2.5825326442718506 }, { "auxiliary_loss_clip": 0.01146923, "auxiliary_loss_mlp": 0.01027379, "balance_loss_clip": 1.04430914, "balance_loss_mlp": 1.01960075, "epoch": 0.6757650453916912, "flos": 21646524210720.0, "grad_norm": 2.482219979296895, "language_loss": 0.78402752, "learning_rate": 1.004951206686758e-06, "loss": 0.80577052, "num_input_tokens_seen": 120899190, "step": 5620, "time_per_iteration": 2.546454668045044 }, { "auxiliary_loss_clip": 0.01154509, "auxiliary_loss_mlp": 0.01026285, "balance_loss_clip": 1.04792261, "balance_loss_mlp": 1.01846743, "epoch": 0.6758852882823303, "flos": 21795661792320.0, "grad_norm": 2.4059554588078176, "language_loss": 0.71596825, "learning_rate": 1.0042755632349087e-06, "loss": 0.73777616, "num_input_tokens_seen": 120916080, "step": 5621, "time_per_iteration": 2.484510898590088 }, { "auxiliary_loss_clip": 0.01129185, "auxiliary_loss_mlp": 0.01028809, "balance_loss_clip": 1.04455018, "balance_loss_mlp": 1.02116764, "epoch": 0.6760055311729694, "flos": 27088787639040.0, "grad_norm": 1.9043660078384415, "language_loss": 0.63088542, "learning_rate": 1.0036000708336653e-06, "loss": 0.65246534, "num_input_tokens_seen": 120935210, "step": 5622, "time_per_iteration": 2.600846290588379 }, { "auxiliary_loss_clip": 0.01147691, "auxiliary_loss_mlp": 0.0102565, "balance_loss_clip": 1.0469451, "balance_loss_mlp": 1.01879287, "epoch": 0.6761257740636085, "flos": 17999795754720.0, "grad_norm": 2.4172035613927285, "language_loss": 0.79702812, "learning_rate": 1.0029247295854984e-06, "loss": 0.81876159, "num_input_tokens_seen": 120951830, "step": 5623, "time_per_iteration": 2.487319231033325 }, { "auxiliary_loss_clip": 0.01138221, "auxiliary_loss_mlp": 0.01024702, "balance_loss_clip": 1.04785562, "balance_loss_mlp": 1.01795185, "epoch": 0.6762460169542476, "flos": 15121915257600.0, "grad_norm": 1.7701995966730741, "language_loss": 0.7182464, "learning_rate": 1.0022495395928588e-06, "loss": 0.73987561, "num_input_tokens_seen": 120970310, "step": 5624, "time_per_iteration": 2.5726253986358643 }, { "auxiliary_loss_clip": 0.01070123, "auxiliary_loss_mlp": 0.01000944, "balance_loss_clip": 1.01127982, "balance_loss_mlp": 0.99957317, "epoch": 0.6763662598448866, "flos": 67886981975520.0, "grad_norm": 0.7879253855367601, "language_loss": 0.6239838, "learning_rate": 1.0015745009581697e-06, "loss": 0.64469451, "num_input_tokens_seen": 121031915, "step": 5625, "time_per_iteration": 3.106102705001831 }, { "auxiliary_loss_clip": 0.01160867, "auxiliary_loss_mlp": 0.01031761, "balance_loss_clip": 1.05231833, "balance_loss_mlp": 1.02437639, "epoch": 0.6764865027355258, "flos": 20631833461920.0, "grad_norm": 1.8116523952063517, "language_loss": 0.67096734, "learning_rate": 1.0008996137838343e-06, "loss": 0.69289362, "num_input_tokens_seen": 121050890, "step": 5626, "time_per_iteration": 3.2373363971710205 }, { "auxiliary_loss_clip": 0.01180745, "auxiliary_loss_mlp": 0.01027899, "balance_loss_clip": 1.05236626, "balance_loss_mlp": 1.02028203, "epoch": 0.6766067456261649, "flos": 21215812852320.0, "grad_norm": 1.9055283308001458, "language_loss": 0.79662549, "learning_rate": 1.000224878172234e-06, "loss": 0.81871194, "num_input_tokens_seen": 121070015, "step": 5627, "time_per_iteration": 2.4725852012634277 }, { "auxiliary_loss_clip": 0.01160645, "auxiliary_loss_mlp": 0.01026138, "balance_loss_clip": 1.05039787, "balance_loss_mlp": 1.01866984, "epoch": 0.6767269885168039, "flos": 19938254674560.0, "grad_norm": 2.097691074741096, "language_loss": 0.72335732, "learning_rate": 9.99550294225724e-07, "loss": 0.74522519, "num_input_tokens_seen": 121089170, "step": 5628, "time_per_iteration": 3.250396251678467 }, { "auxiliary_loss_clip": 0.01118827, "auxiliary_loss_mlp": 0.01030067, "balance_loss_clip": 1.04128218, "balance_loss_mlp": 1.02228236, "epoch": 0.6768472314074431, "flos": 20814080092320.0, "grad_norm": 2.028587525599962, "language_loss": 0.72462112, "learning_rate": 9.988758620466402e-07, "loss": 0.74611008, "num_input_tokens_seen": 121108040, "step": 5629, "time_per_iteration": 2.600249767303467 }, { "auxiliary_loss_clip": 0.011095, "auxiliary_loss_mlp": 0.01020893, "balance_loss_clip": 1.04472923, "balance_loss_mlp": 1.01425886, "epoch": 0.6769674742980821, "flos": 23186015976480.0, "grad_norm": 1.7370165693888528, "language_loss": 0.76130176, "learning_rate": 9.982015817372917e-07, "loss": 0.78260565, "num_input_tokens_seen": 121128480, "step": 5630, "time_per_iteration": 3.374800205230713 }, { "auxiliary_loss_clip": 0.01114294, "auxiliary_loss_mlp": 0.01024737, "balance_loss_clip": 1.04282939, "balance_loss_mlp": 1.01700354, "epoch": 0.6770877171887212, "flos": 24242938681920.0, "grad_norm": 1.8504323138902452, "language_loss": 0.8218199, "learning_rate": 9.975274533999657e-07, "loss": 0.84321022, "num_input_tokens_seen": 121148010, "step": 5631, "time_per_iteration": 2.6268150806427 }, { "auxiliary_loss_clip": 0.0117618, "auxiliary_loss_mlp": 0.01028432, "balance_loss_clip": 1.05009472, "balance_loss_mlp": 1.02078152, "epoch": 0.6772079600793603, "flos": 18141570359520.0, "grad_norm": 4.1555256773490905, "language_loss": 0.84083468, "learning_rate": 9.96853477136929e-07, "loss": 0.86288083, "num_input_tokens_seen": 121162755, "step": 5632, "time_per_iteration": 2.4497463703155518 }, { "auxiliary_loss_clip": 0.01123756, "auxiliary_loss_mlp": 0.01031877, "balance_loss_clip": 1.04316783, "balance_loss_mlp": 1.02487683, "epoch": 0.6773282029699994, "flos": 22452072914400.0, "grad_norm": 1.8999024465755896, "language_loss": 0.7507751, "learning_rate": 9.96179653050422e-07, "loss": 0.77233142, "num_input_tokens_seen": 121182915, "step": 5633, "time_per_iteration": 2.5823683738708496 }, { "auxiliary_loss_clip": 0.01122373, "auxiliary_loss_mlp": 0.01029607, "balance_loss_clip": 1.04325366, "balance_loss_mlp": 1.02163219, "epoch": 0.6774484458606385, "flos": 18693733711680.0, "grad_norm": 2.7498141371839844, "language_loss": 0.7392512, "learning_rate": 9.955059812426635e-07, "loss": 0.76077104, "num_input_tokens_seen": 121200445, "step": 5634, "time_per_iteration": 2.5302069187164307 }, { "auxiliary_loss_clip": 0.01177342, "auxiliary_loss_mlp": 0.01030414, "balance_loss_clip": 1.05348253, "balance_loss_mlp": 1.02276707, "epoch": 0.6775686887512776, "flos": 25994050763040.0, "grad_norm": 1.9981669611849757, "language_loss": 0.82920694, "learning_rate": 9.948324618158493e-07, "loss": 0.8512845, "num_input_tokens_seen": 121220785, "step": 5635, "time_per_iteration": 2.5185914039611816 }, { "auxiliary_loss_clip": 0.01162256, "auxiliary_loss_mlp": 0.01024425, "balance_loss_clip": 1.04777253, "balance_loss_mlp": 1.01625919, "epoch": 0.6776889316419167, "flos": 13587990620640.0, "grad_norm": 2.168130882623341, "language_loss": 0.7688061, "learning_rate": 9.941590948721502e-07, "loss": 0.7906729, "num_input_tokens_seen": 121237985, "step": 5636, "time_per_iteration": 2.4853575229644775 }, { "auxiliary_loss_clip": 0.01142298, "auxiliary_loss_mlp": 0.01020909, "balance_loss_clip": 1.0498364, "balance_loss_mlp": 1.01369691, "epoch": 0.6778091745325557, "flos": 27601125470880.0, "grad_norm": 1.8659930658765047, "language_loss": 0.76434696, "learning_rate": 9.934858805137188e-07, "loss": 0.78597897, "num_input_tokens_seen": 121258635, "step": 5637, "time_per_iteration": 3.345146894454956 }, { "auxiliary_loss_clip": 0.01156035, "auxiliary_loss_mlp": 0.01024146, "balance_loss_clip": 1.04973674, "balance_loss_mlp": 1.01758075, "epoch": 0.6779294174231949, "flos": 18734062069440.0, "grad_norm": 1.6516784052443523, "language_loss": 0.8049264, "learning_rate": 9.92812818842677e-07, "loss": 0.82672817, "num_input_tokens_seen": 121277810, "step": 5638, "time_per_iteration": 2.4806230068206787 }, { "auxiliary_loss_clip": 0.01157438, "auxiliary_loss_mlp": 0.01025308, "balance_loss_clip": 1.04946649, "balance_loss_mlp": 1.01785445, "epoch": 0.678049660313834, "flos": 45873803098080.0, "grad_norm": 1.783891158894089, "language_loss": 0.63952708, "learning_rate": 9.921399099611306e-07, "loss": 0.66135454, "num_input_tokens_seen": 121298975, "step": 5639, "time_per_iteration": 2.704127788543701 }, { "auxiliary_loss_clip": 0.01147112, "auxiliary_loss_mlp": 0.01026544, "balance_loss_clip": 1.0470506, "balance_loss_mlp": 1.0193851, "epoch": 0.678169903204473, "flos": 19974560332800.0, "grad_norm": 1.5960558285465984, "language_loss": 0.68544531, "learning_rate": 9.914671539711588e-07, "loss": 0.70718181, "num_input_tokens_seen": 121318495, "step": 5640, "time_per_iteration": 2.5177528858184814 }, { "auxiliary_loss_clip": 0.01080421, "auxiliary_loss_mlp": 0.00762602, "balance_loss_clip": 1.04446173, "balance_loss_mlp": 1.00049484, "epoch": 0.6782901460951122, "flos": 21395617129440.0, "grad_norm": 1.8041571964842404, "language_loss": 0.77759457, "learning_rate": 9.90794550974817e-07, "loss": 0.7960248, "num_input_tokens_seen": 121338890, "step": 5641, "time_per_iteration": 2.8313052654266357 }, { "auxiliary_loss_clip": 0.01128703, "auxiliary_loss_mlp": 0.01029449, "balance_loss_clip": 1.04595518, "balance_loss_mlp": 1.02233195, "epoch": 0.6784103889857512, "flos": 21434005971360.0, "grad_norm": 2.223056836559485, "language_loss": 0.81452113, "learning_rate": 9.901221010741407e-07, "loss": 0.83610272, "num_input_tokens_seen": 121358210, "step": 5642, "time_per_iteration": 2.6867053508758545 }, { "auxiliary_loss_clip": 0.01162395, "auxiliary_loss_mlp": 0.01030809, "balance_loss_clip": 1.04763937, "balance_loss_mlp": 1.02351975, "epoch": 0.6785306318763903, "flos": 32671927748160.0, "grad_norm": 2.0748369034497545, "language_loss": 0.74891055, "learning_rate": 9.894498043711375e-07, "loss": 0.77084261, "num_input_tokens_seen": 121379955, "step": 5643, "time_per_iteration": 2.604799747467041 }, { "auxiliary_loss_clip": 0.01141783, "auxiliary_loss_mlp": 0.01021687, "balance_loss_clip": 1.04408956, "balance_loss_mlp": 1.01500559, "epoch": 0.6786508747670293, "flos": 25632143523360.0, "grad_norm": 1.9741224217932285, "language_loss": 0.69408786, "learning_rate": 9.887776609677962e-07, "loss": 0.71572256, "num_input_tokens_seen": 121401325, "step": 5644, "time_per_iteration": 2.623929738998413 }, { "auxiliary_loss_clip": 0.01117892, "auxiliary_loss_mlp": 0.01027606, "balance_loss_clip": 1.04029655, "balance_loss_mlp": 1.02045357, "epoch": 0.6787711176576685, "flos": 19171884985920.0, "grad_norm": 1.6045543097755772, "language_loss": 0.72095567, "learning_rate": 9.88105670966079e-07, "loss": 0.7424106, "num_input_tokens_seen": 121419785, "step": 5645, "time_per_iteration": 2.5656940937042236 }, { "auxiliary_loss_clip": 0.01102629, "auxiliary_loss_mlp": 0.01023042, "balance_loss_clip": 1.04128158, "balance_loss_mlp": 1.01562762, "epoch": 0.6788913605483076, "flos": 13985161926720.0, "grad_norm": 1.8227786731069031, "language_loss": 0.78516483, "learning_rate": 9.874338344679283e-07, "loss": 0.80642152, "num_input_tokens_seen": 121435630, "step": 5646, "time_per_iteration": 2.572036027908325 }, { "auxiliary_loss_clip": 0.01171859, "auxiliary_loss_mlp": 0.01031093, "balance_loss_clip": 1.05096781, "balance_loss_mlp": 1.02422905, "epoch": 0.6790116034389466, "flos": 22017590275200.0, "grad_norm": 1.8697260466838892, "language_loss": 0.74112183, "learning_rate": 9.86762151575259e-07, "loss": 0.76315135, "num_input_tokens_seen": 121455625, "step": 5647, "time_per_iteration": 2.46181583404541 }, { "auxiliary_loss_clip": 0.01117989, "auxiliary_loss_mlp": 0.00761402, "balance_loss_clip": 1.04860306, "balance_loss_mlp": 1.00045836, "epoch": 0.6791318463295858, "flos": 20922458312640.0, "grad_norm": 1.6775061321944589, "language_loss": 0.80152845, "learning_rate": 9.860906223899651e-07, "loss": 0.82032239, "num_input_tokens_seen": 121475020, "step": 5648, "time_per_iteration": 2.603717565536499 }, { "auxiliary_loss_clip": 0.01150767, "auxiliary_loss_mlp": 0.01025369, "balance_loss_clip": 1.04630494, "balance_loss_mlp": 1.0185678, "epoch": 0.6792520892202248, "flos": 28512753709440.0, "grad_norm": 1.4979377119769857, "language_loss": 0.75268382, "learning_rate": 9.854192470139184e-07, "loss": 0.77444524, "num_input_tokens_seen": 121496500, "step": 5649, "time_per_iteration": 2.5720112323760986 }, { "auxiliary_loss_clip": 0.01146071, "auxiliary_loss_mlp": 0.01035657, "balance_loss_clip": 1.05050588, "balance_loss_mlp": 1.02868295, "epoch": 0.6793723321108639, "flos": 20011907582880.0, "grad_norm": 2.3042013052872714, "language_loss": 0.71507233, "learning_rate": 9.847480255489645e-07, "loss": 0.7368896, "num_input_tokens_seen": 121515525, "step": 5650, "time_per_iteration": 2.5380859375 }, { "auxiliary_loss_clip": 0.0114973, "auxiliary_loss_mlp": 0.01028387, "balance_loss_clip": 1.04699516, "balance_loss_mlp": 1.0212431, "epoch": 0.6794925750015031, "flos": 26649492127200.0, "grad_norm": 1.6339645454017282, "language_loss": 0.68822658, "learning_rate": 9.840769580969295e-07, "loss": 0.71000779, "num_input_tokens_seen": 121535965, "step": 5651, "time_per_iteration": 2.5602099895477295 }, { "auxiliary_loss_clip": 0.01150444, "auxiliary_loss_mlp": 0.0102625, "balance_loss_clip": 1.04640007, "balance_loss_mlp": 1.01905632, "epoch": 0.6796128178921421, "flos": 21580377947040.0, "grad_norm": 2.004263081559183, "language_loss": 0.79856265, "learning_rate": 9.834060447596114e-07, "loss": 0.82032967, "num_input_tokens_seen": 121555235, "step": 5652, "time_per_iteration": 3.2680249214172363 }, { "auxiliary_loss_clip": 0.01159579, "auxiliary_loss_mlp": 0.01026176, "balance_loss_clip": 1.04657435, "balance_loss_mlp": 1.01893425, "epoch": 0.6797330607827812, "flos": 22492006185600.0, "grad_norm": 1.8295258588979983, "language_loss": 0.77945268, "learning_rate": 9.827352856387868e-07, "loss": 0.8013103, "num_input_tokens_seen": 121574945, "step": 5653, "time_per_iteration": 2.524104595184326 }, { "auxiliary_loss_clip": 0.01022015, "auxiliary_loss_mlp": 0.01001898, "balance_loss_clip": 1.00994241, "balance_loss_mlp": 1.00061035, "epoch": 0.6798533036734203, "flos": 66306659014560.0, "grad_norm": 0.7814351791130637, "language_loss": 0.64259779, "learning_rate": 9.820646808362118e-07, "loss": 0.66283697, "num_input_tokens_seen": 121641200, "step": 5654, "time_per_iteration": 3.251014232635498 }, { "auxiliary_loss_clip": 0.01142993, "auxiliary_loss_mlp": 0.01022339, "balance_loss_clip": 1.04771662, "balance_loss_mlp": 1.01526141, "epoch": 0.6799735465640594, "flos": 16180166890560.0, "grad_norm": 2.149523756520981, "language_loss": 0.72788846, "learning_rate": 9.813942304536154e-07, "loss": 0.7495417, "num_input_tokens_seen": 121659170, "step": 5655, "time_per_iteration": 3.2937018871307373 }, { "auxiliary_loss_clip": 0.01147671, "auxiliary_loss_mlp": 0.01021616, "balance_loss_clip": 1.04757369, "balance_loss_mlp": 1.01427269, "epoch": 0.6800937894546984, "flos": 22125752993760.0, "grad_norm": 1.8803446200476817, "language_loss": 0.63839722, "learning_rate": 9.807239345927043e-07, "loss": 0.66009009, "num_input_tokens_seen": 121679180, "step": 5656, "time_per_iteration": 3.530117988586426 }, { "auxiliary_loss_clip": 0.01144099, "auxiliary_loss_mlp": 0.01023834, "balance_loss_clip": 1.0447253, "balance_loss_mlp": 1.01609111, "epoch": 0.6802140323453376, "flos": 31612957776000.0, "grad_norm": 2.2094328507556793, "language_loss": 0.7171756, "learning_rate": 9.80053793355162e-07, "loss": 0.73885489, "num_input_tokens_seen": 121697875, "step": 5657, "time_per_iteration": 2.594400644302368 }, { "auxiliary_loss_clip": 0.01114448, "auxiliary_loss_mlp": 0.01023424, "balance_loss_clip": 1.04518938, "balance_loss_mlp": 1.01622355, "epoch": 0.6803342752359767, "flos": 17712942184800.0, "grad_norm": 1.9231353692604454, "language_loss": 0.74782985, "learning_rate": 9.793838068426472e-07, "loss": 0.76920861, "num_input_tokens_seen": 121715570, "step": 5658, "time_per_iteration": 2.5646231174468994 }, { "auxiliary_loss_clip": 0.01174073, "auxiliary_loss_mlp": 0.01020967, "balance_loss_clip": 1.05100417, "balance_loss_mlp": 1.01390362, "epoch": 0.6804545181266157, "flos": 11326803476160.0, "grad_norm": 2.470491673173159, "language_loss": 0.61132646, "learning_rate": 9.78713975156799e-07, "loss": 0.63327682, "num_input_tokens_seen": 121731435, "step": 5659, "time_per_iteration": 2.4510974884033203 }, { "auxiliary_loss_clip": 0.01135994, "auxiliary_loss_mlp": 0.01028921, "balance_loss_clip": 1.04978037, "balance_loss_mlp": 1.02183044, "epoch": 0.6805747610172549, "flos": 29350980458400.0, "grad_norm": 1.8890871188236487, "language_loss": 0.71850228, "learning_rate": 9.780442983992273e-07, "loss": 0.74015146, "num_input_tokens_seen": 121749950, "step": 5660, "time_per_iteration": 2.6065757274627686 }, { "auxiliary_loss_clip": 0.01139512, "auxiliary_loss_mlp": 0.01026496, "balance_loss_clip": 1.04709435, "balance_loss_mlp": 1.01931417, "epoch": 0.680695003907894, "flos": 37631873534880.0, "grad_norm": 1.740425646673145, "language_loss": 0.71554786, "learning_rate": 9.773747766715238e-07, "loss": 0.73720789, "num_input_tokens_seen": 121770770, "step": 5661, "time_per_iteration": 2.6742913722991943 }, { "auxiliary_loss_clip": 0.01146412, "auxiliary_loss_mlp": 0.01025317, "balance_loss_clip": 1.04558277, "balance_loss_mlp": 1.01860237, "epoch": 0.680815246798533, "flos": 22127369256960.0, "grad_norm": 1.5537346349786199, "language_loss": 0.80112749, "learning_rate": 9.767054100752536e-07, "loss": 0.82284486, "num_input_tokens_seen": 121790720, "step": 5662, "time_per_iteration": 2.5273334980010986 }, { "auxiliary_loss_clip": 0.01129673, "auxiliary_loss_mlp": 0.01027135, "balance_loss_clip": 1.04719973, "balance_loss_mlp": 1.01972282, "epoch": 0.6809354896891722, "flos": 17201825529600.0, "grad_norm": 1.8063900451917496, "language_loss": 0.82067579, "learning_rate": 9.760361987119584e-07, "loss": 0.84224391, "num_input_tokens_seen": 121808455, "step": 5663, "time_per_iteration": 3.290332317352295 }, { "auxiliary_loss_clip": 0.0114524, "auxiliary_loss_mlp": 0.01028486, "balance_loss_clip": 1.04869831, "balance_loss_mlp": 1.02096987, "epoch": 0.6810557325798112, "flos": 12458169263040.0, "grad_norm": 1.938448732920917, "language_loss": 0.67618591, "learning_rate": 9.753671426831592e-07, "loss": 0.69792318, "num_input_tokens_seen": 121824470, "step": 5664, "time_per_iteration": 2.498682975769043 }, { "auxiliary_loss_clip": 0.01149918, "auxiliary_loss_mlp": 0.0102744, "balance_loss_clip": 1.04394174, "balance_loss_mlp": 1.01990557, "epoch": 0.6811759754704503, "flos": 22156168270560.0, "grad_norm": 1.838796665026462, "language_loss": 0.79562712, "learning_rate": 9.746982420903483e-07, "loss": 0.81740069, "num_input_tokens_seen": 121842665, "step": 5665, "time_per_iteration": 2.4852988719940186 }, { "auxiliary_loss_clip": 0.0115517, "auxiliary_loss_mlp": 0.01024062, "balance_loss_clip": 1.04943895, "balance_loss_mlp": 1.01696944, "epoch": 0.6812962183610894, "flos": 17525379844320.0, "grad_norm": 1.566785106052902, "language_loss": 0.74462992, "learning_rate": 9.740294970349993e-07, "loss": 0.76642221, "num_input_tokens_seen": 121859080, "step": 5666, "time_per_iteration": 2.4787354469299316 }, { "auxiliary_loss_clip": 0.0104772, "auxiliary_loss_mlp": 0.01002485, "balance_loss_clip": 1.00962329, "balance_loss_mlp": 1.00116146, "epoch": 0.6814164612517285, "flos": 60274489897440.0, "grad_norm": 0.886537363349886, "language_loss": 0.60963213, "learning_rate": 9.733609076185594e-07, "loss": 0.63013422, "num_input_tokens_seen": 121915485, "step": 5667, "time_per_iteration": 3.0201609134674072 }, { "auxiliary_loss_clip": 0.01160283, "auxiliary_loss_mlp": 0.01022154, "balance_loss_clip": 1.05031228, "balance_loss_mlp": 1.01448274, "epoch": 0.6815367041423676, "flos": 19317754124160.0, "grad_norm": 2.2247138452692776, "language_loss": 0.84189606, "learning_rate": 9.72692473942455e-07, "loss": 0.86372042, "num_input_tokens_seen": 121932710, "step": 5668, "time_per_iteration": 2.4903056621551514 }, { "auxiliary_loss_clip": 0.01123611, "auxiliary_loss_mlp": 0.01031796, "balance_loss_clip": 1.04787529, "balance_loss_mlp": 1.02440763, "epoch": 0.6816569470330067, "flos": 22161699482400.0, "grad_norm": 1.5517441270785095, "language_loss": 0.77638191, "learning_rate": 9.720241961080849e-07, "loss": 0.79793596, "num_input_tokens_seen": 121952025, "step": 5669, "time_per_iteration": 2.5904059410095215 }, { "auxiliary_loss_clip": 0.01172977, "auxiliary_loss_mlp": 0.01019876, "balance_loss_clip": 1.04960394, "balance_loss_mlp": 1.01341128, "epoch": 0.6817771899236458, "flos": 41463506476320.0, "grad_norm": 2.0352115006602265, "language_loss": 0.73048806, "learning_rate": 9.713560742168259e-07, "loss": 0.75241661, "num_input_tokens_seen": 121974650, "step": 5670, "time_per_iteration": 2.6682536602020264 }, { "auxiliary_loss_clip": 0.01126923, "auxiliary_loss_mlp": 0.01026302, "balance_loss_clip": 1.04498208, "balance_loss_mlp": 1.01936412, "epoch": 0.6818974328142848, "flos": 21106141621440.0, "grad_norm": 2.0964498336922817, "language_loss": 0.7144599, "learning_rate": 9.706881083700333e-07, "loss": 0.73599207, "num_input_tokens_seen": 121994335, "step": 5671, "time_per_iteration": 2.5485591888427734 }, { "auxiliary_loss_clip": 0.01105204, "auxiliary_loss_mlp": 0.01025121, "balance_loss_clip": 1.04747653, "balance_loss_mlp": 1.01762319, "epoch": 0.682017675704924, "flos": 20441900602080.0, "grad_norm": 1.9493979882895134, "language_loss": 0.82095659, "learning_rate": 9.700202986690357e-07, "loss": 0.84225982, "num_input_tokens_seen": 122012635, "step": 5672, "time_per_iteration": 2.6030995845794678 }, { "auxiliary_loss_clip": 0.01158611, "auxiliary_loss_mlp": 0.0076248, "balance_loss_clip": 1.04925036, "balance_loss_mlp": 1.0004369, "epoch": 0.682137918595563, "flos": 20044441960320.0, "grad_norm": 2.003711672092472, "language_loss": 0.66359639, "learning_rate": 9.693526452151413e-07, "loss": 0.68280727, "num_input_tokens_seen": 122031685, "step": 5673, "time_per_iteration": 2.4818544387817383 }, { "auxiliary_loss_clip": 0.01138616, "auxiliary_loss_mlp": 0.01028299, "balance_loss_clip": 1.04623842, "balance_loss_mlp": 1.02077389, "epoch": 0.6822581614862021, "flos": 31684563417600.0, "grad_norm": 1.7327297052631438, "language_loss": 0.75291026, "learning_rate": 9.686851481096305e-07, "loss": 0.77457935, "num_input_tokens_seen": 122052995, "step": 5674, "time_per_iteration": 2.6633689403533936 }, { "auxiliary_loss_clip": 0.01098606, "auxiliary_loss_mlp": 0.01027243, "balance_loss_clip": 1.04540026, "balance_loss_mlp": 1.01995301, "epoch": 0.6823784043768413, "flos": 23477574668160.0, "grad_norm": 1.7421617683255224, "language_loss": 0.71836138, "learning_rate": 9.68017807453762e-07, "loss": 0.73961985, "num_input_tokens_seen": 122071740, "step": 5675, "time_per_iteration": 2.635746717453003 }, { "auxiliary_loss_clip": 0.01145453, "auxiliary_loss_mlp": 0.00761645, "balance_loss_clip": 1.04747605, "balance_loss_mlp": 1.00050569, "epoch": 0.6824986472674803, "flos": 14137136948160.0, "grad_norm": 1.643439899099331, "language_loss": 0.72981352, "learning_rate": 9.673506233487721e-07, "loss": 0.7488845, "num_input_tokens_seen": 122089705, "step": 5676, "time_per_iteration": 2.5410451889038086 }, { "auxiliary_loss_clip": 0.01143972, "auxiliary_loss_mlp": 0.00761559, "balance_loss_clip": 1.04573393, "balance_loss_mlp": 1.00039625, "epoch": 0.6826188901581194, "flos": 21505001024640.0, "grad_norm": 1.6607486289274935, "language_loss": 0.86019254, "learning_rate": 9.666835958958717e-07, "loss": 0.8792479, "num_input_tokens_seen": 122109025, "step": 5677, "time_per_iteration": 2.5291173458099365 }, { "auxiliary_loss_clip": 0.01172748, "auxiliary_loss_mlp": 0.01023533, "balance_loss_clip": 1.05095816, "balance_loss_mlp": 1.01643729, "epoch": 0.6827391330487584, "flos": 20810129226720.0, "grad_norm": 2.0095870549492325, "language_loss": 0.8047989, "learning_rate": 9.660167251962484e-07, "loss": 0.82676172, "num_input_tokens_seen": 122127385, "step": 5678, "time_per_iteration": 3.2398669719696045 }, { "auxiliary_loss_clip": 0.01135786, "auxiliary_loss_mlp": 0.0102493, "balance_loss_clip": 1.04591918, "balance_loss_mlp": 1.01821232, "epoch": 0.6828593759393976, "flos": 21688792084320.0, "grad_norm": 1.600769007907262, "language_loss": 0.779989, "learning_rate": 9.653500113510654e-07, "loss": 0.80159611, "num_input_tokens_seen": 122146500, "step": 5679, "time_per_iteration": 2.597761631011963 }, { "auxiliary_loss_clip": 0.01137669, "auxiliary_loss_mlp": 0.01020314, "balance_loss_clip": 1.04421973, "balance_loss_mlp": 1.01319408, "epoch": 0.6829796188300367, "flos": 25337711474880.0, "grad_norm": 2.116421377171469, "language_loss": 0.66902184, "learning_rate": 9.646834544614627e-07, "loss": 0.69060165, "num_input_tokens_seen": 122167000, "step": 5680, "time_per_iteration": 3.3288283348083496 }, { "auxiliary_loss_clip": 0.01138133, "auxiliary_loss_mlp": 0.01028574, "balance_loss_clip": 1.04872632, "balance_loss_mlp": 1.02144551, "epoch": 0.6830998617206757, "flos": 20704803948000.0, "grad_norm": 1.9688767569747234, "language_loss": 0.76320505, "learning_rate": 9.64017054628558e-07, "loss": 0.78487217, "num_input_tokens_seen": 122185825, "step": 5681, "time_per_iteration": 3.2168726921081543 }, { "auxiliary_loss_clip": 0.01120436, "auxiliary_loss_mlp": 0.01026883, "balance_loss_clip": 1.0447216, "balance_loss_mlp": 1.01948285, "epoch": 0.6832201046113149, "flos": 21726642171840.0, "grad_norm": 1.5740176515761086, "language_loss": 0.78820109, "learning_rate": 9.63350811953441e-07, "loss": 0.80967432, "num_input_tokens_seen": 122206200, "step": 5682, "time_per_iteration": 2.599893808364868 }, { "auxiliary_loss_clip": 0.01132261, "auxiliary_loss_mlp": 0.01022468, "balance_loss_clip": 1.04565573, "balance_loss_mlp": 1.01528537, "epoch": 0.6833403475019539, "flos": 19536557831520.0, "grad_norm": 1.9809408007007199, "language_loss": 0.7051996, "learning_rate": 9.626847265371826e-07, "loss": 0.72674686, "num_input_tokens_seen": 122225520, "step": 5683, "time_per_iteration": 2.55680513381958 }, { "auxiliary_loss_clip": 0.01135276, "auxiliary_loss_mlp": 0.01028448, "balance_loss_clip": 1.04392672, "balance_loss_mlp": 1.02135849, "epoch": 0.683460590392593, "flos": 19352156183520.0, "grad_norm": 1.9401346370360595, "language_loss": 0.78447771, "learning_rate": 9.620187984808262e-07, "loss": 0.80611491, "num_input_tokens_seen": 122244320, "step": 5684, "time_per_iteration": 2.5257039070129395 }, { "auxiliary_loss_clip": 0.01145564, "auxiliary_loss_mlp": 0.00761676, "balance_loss_clip": 1.04957008, "balance_loss_mlp": 1.00042796, "epoch": 0.6835808332832322, "flos": 23288503815360.0, "grad_norm": 2.124729896940784, "language_loss": 0.86155534, "learning_rate": 9.613530278853919e-07, "loss": 0.88062769, "num_input_tokens_seen": 122264295, "step": 5685, "time_per_iteration": 2.537489891052246 }, { "auxiliary_loss_clip": 0.0115909, "auxiliary_loss_mlp": 0.0102485, "balance_loss_clip": 1.04999495, "balance_loss_mlp": 1.01773906, "epoch": 0.6837010761738712, "flos": 21653420267040.0, "grad_norm": 1.8480032519417384, "language_loss": 0.74034142, "learning_rate": 9.60687414851879e-07, "loss": 0.76218081, "num_input_tokens_seen": 122285300, "step": 5686, "time_per_iteration": 2.5013058185577393 }, { "auxiliary_loss_clip": 0.0114478, "auxiliary_loss_mlp": 0.01033165, "balance_loss_clip": 1.0488776, "balance_loss_mlp": 1.02552664, "epoch": 0.6838213190645103, "flos": 17566390624320.0, "grad_norm": 2.019967582915539, "language_loss": 0.76887172, "learning_rate": 9.600219594812575e-07, "loss": 0.7906512, "num_input_tokens_seen": 122303240, "step": 5687, "time_per_iteration": 2.4902567863464355 }, { "auxiliary_loss_clip": 0.01169928, "auxiliary_loss_mlp": 0.01025268, "balance_loss_clip": 1.04932129, "balance_loss_mlp": 1.01817226, "epoch": 0.6839415619551494, "flos": 23112542652960.0, "grad_norm": 1.6576491148475374, "language_loss": 0.72731531, "learning_rate": 9.593566618744786e-07, "loss": 0.74926734, "num_input_tokens_seen": 122323390, "step": 5688, "time_per_iteration": 2.4696452617645264 }, { "auxiliary_loss_clip": 0.01173849, "auxiliary_loss_mlp": 0.01026741, "balance_loss_clip": 1.05053008, "balance_loss_mlp": 1.01949644, "epoch": 0.6840618048457885, "flos": 22127872094400.0, "grad_norm": 1.7726549996559644, "language_loss": 0.73867285, "learning_rate": 9.58691522132466e-07, "loss": 0.76067871, "num_input_tokens_seen": 122342200, "step": 5689, "time_per_iteration": 3.176426887512207 }, { "auxiliary_loss_clip": 0.01151545, "auxiliary_loss_mlp": 0.01027801, "balance_loss_clip": 1.05083835, "balance_loss_mlp": 1.0206666, "epoch": 0.6841820477364275, "flos": 22015902178080.0, "grad_norm": 2.1610171757850214, "language_loss": 0.84814501, "learning_rate": 9.58026540356123e-07, "loss": 0.86993849, "num_input_tokens_seen": 122360465, "step": 5690, "time_per_iteration": 2.5540225505828857 }, { "auxiliary_loss_clip": 0.01160647, "auxiliary_loss_mlp": 0.01028735, "balance_loss_clip": 1.04799151, "balance_loss_mlp": 1.02096522, "epoch": 0.6843022906270667, "flos": 24900535063680.0, "grad_norm": 1.6930999881196174, "language_loss": 0.86747062, "learning_rate": 9.573617166463246e-07, "loss": 0.88936448, "num_input_tokens_seen": 122381680, "step": 5691, "time_per_iteration": 2.55532169342041 }, { "auxiliary_loss_clip": 0.01147563, "auxiliary_loss_mlp": 0.01022568, "balance_loss_clip": 1.04644227, "balance_loss_mlp": 1.01511729, "epoch": 0.6844225335177058, "flos": 19969927044960.0, "grad_norm": 1.766725659589681, "language_loss": 0.60218394, "learning_rate": 9.56697051103924e-07, "loss": 0.62388527, "num_input_tokens_seen": 122399120, "step": 5692, "time_per_iteration": 2.544691801071167 }, { "auxiliary_loss_clip": 0.01141996, "auxiliary_loss_mlp": 0.01023603, "balance_loss_clip": 1.0455215, "balance_loss_mlp": 1.01678443, "epoch": 0.6845427764083448, "flos": 25883337940320.0, "grad_norm": 1.9267497742691335, "language_loss": 0.81182826, "learning_rate": 9.560325438297522e-07, "loss": 0.83348423, "num_input_tokens_seen": 122417430, "step": 5693, "time_per_iteration": 2.546790838241577 }, { "auxiliary_loss_clip": 0.01147432, "auxiliary_loss_mlp": 0.01029986, "balance_loss_clip": 1.05128133, "balance_loss_mlp": 1.02304554, "epoch": 0.684663019298984, "flos": 18880146709440.0, "grad_norm": 2.1113030705382663, "language_loss": 0.868527, "learning_rate": 9.553681949246127e-07, "loss": 0.89030123, "num_input_tokens_seen": 122435055, "step": 5694, "time_per_iteration": 2.5407521724700928 }, { "auxiliary_loss_clip": 0.0113953, "auxiliary_loss_mlp": 0.0102726, "balance_loss_clip": 1.0490551, "balance_loss_mlp": 1.0191772, "epoch": 0.684783262189623, "flos": 54193731521760.0, "grad_norm": 1.8368531076641172, "language_loss": 0.75514871, "learning_rate": 9.547040044892886e-07, "loss": 0.77681661, "num_input_tokens_seen": 122462570, "step": 5695, "time_per_iteration": 2.838430643081665 }, { "auxiliary_loss_clip": 0.01058795, "auxiliary_loss_mlp": 0.00999842, "balance_loss_clip": 1.00876665, "balance_loss_mlp": 0.99851876, "epoch": 0.6849035050802621, "flos": 63970274532480.0, "grad_norm": 0.8606225948324648, "language_loss": 0.60163599, "learning_rate": 9.540399726245354e-07, "loss": 0.62222236, "num_input_tokens_seen": 122519275, "step": 5696, "time_per_iteration": 2.9704620838165283 }, { "auxiliary_loss_clip": 0.01139911, "auxiliary_loss_mlp": 0.01024934, "balance_loss_clip": 1.04429543, "balance_loss_mlp": 1.01750767, "epoch": 0.6850237479709013, "flos": 25224125295360.0, "grad_norm": 1.9519801351342352, "language_loss": 0.6915862, "learning_rate": 9.533760994310859e-07, "loss": 0.71323466, "num_input_tokens_seen": 122539675, "step": 5697, "time_per_iteration": 2.562941789627075 }, { "auxiliary_loss_clip": 0.01177131, "auxiliary_loss_mlp": 0.01026674, "balance_loss_clip": 1.05198979, "balance_loss_mlp": 1.01970637, "epoch": 0.6851439908615403, "flos": 19354131616320.0, "grad_norm": 2.0958692617860915, "language_loss": 0.75733215, "learning_rate": 9.527123850096508e-07, "loss": 0.77937025, "num_input_tokens_seen": 122558035, "step": 5698, "time_per_iteration": 2.468916893005371 }, { "auxiliary_loss_clip": 0.01157451, "auxiliary_loss_mlp": 0.01022895, "balance_loss_clip": 1.04757392, "balance_loss_mlp": 1.01561737, "epoch": 0.6852642337521794, "flos": 23182136944800.0, "grad_norm": 1.8613530538419814, "language_loss": 0.7179386, "learning_rate": 9.520488294609142e-07, "loss": 0.73974204, "num_input_tokens_seen": 122576815, "step": 5699, "time_per_iteration": 2.534804344177246 }, { "auxiliary_loss_clip": 0.01023591, "auxiliary_loss_mlp": 0.01001786, "balance_loss_clip": 1.00841415, "balance_loss_mlp": 1.00062418, "epoch": 0.6853844766428185, "flos": 62647215954720.0, "grad_norm": 0.7531074555972648, "language_loss": 0.53869689, "learning_rate": 9.513854328855368e-07, "loss": 0.55895066, "num_input_tokens_seen": 122634690, "step": 5700, "time_per_iteration": 3.1475512981414795 }, { "auxiliary_loss_clip": 0.01168924, "auxiliary_loss_mlp": 0.01019308, "balance_loss_clip": 1.04762304, "balance_loss_mlp": 1.01213109, "epoch": 0.6855047195334576, "flos": 23437246310400.0, "grad_norm": 2.024946972050924, "language_loss": 0.81421447, "learning_rate": 9.507221953841558e-07, "loss": 0.83609676, "num_input_tokens_seen": 122652320, "step": 5701, "time_per_iteration": 2.4658007621765137 }, { "auxiliary_loss_clip": 0.01159636, "auxiliary_loss_mlp": 0.01024804, "balance_loss_clip": 1.050336, "balance_loss_mlp": 1.01779437, "epoch": 0.6856249624240967, "flos": 20664834759840.0, "grad_norm": 1.5432356656725206, "language_loss": 0.77655923, "learning_rate": 9.500591170573824e-07, "loss": 0.79840362, "num_input_tokens_seen": 122672340, "step": 5702, "time_per_iteration": 2.5301668643951416 }, { "auxiliary_loss_clip": 0.01113014, "auxiliary_loss_mlp": 0.01023604, "balance_loss_clip": 1.04448175, "balance_loss_mlp": 1.01647854, "epoch": 0.6857452053147358, "flos": 17087305509120.0, "grad_norm": 1.863182074538667, "language_loss": 0.74167323, "learning_rate": 9.493961980058078e-07, "loss": 0.76303941, "num_input_tokens_seen": 122689935, "step": 5703, "time_per_iteration": 3.349274158477783 }, { "auxiliary_loss_clip": 0.0108486, "auxiliary_loss_mlp": 0.01021635, "balance_loss_clip": 1.03731537, "balance_loss_mlp": 1.01475966, "epoch": 0.6858654482053749, "flos": 30847270509600.0, "grad_norm": 1.7669312386625635, "language_loss": 0.67916501, "learning_rate": 9.48733438329993e-07, "loss": 0.70023, "num_input_tokens_seen": 122710200, "step": 5704, "time_per_iteration": 2.695899724960327 }, { "auxiliary_loss_clip": 0.01170268, "auxiliary_loss_mlp": 0.0076169, "balance_loss_clip": 1.05030525, "balance_loss_mlp": 1.00048041, "epoch": 0.6859856910960139, "flos": 28877318804160.0, "grad_norm": 1.6623454325678442, "language_loss": 0.74458373, "learning_rate": 9.480708381304807e-07, "loss": 0.76390338, "num_input_tokens_seen": 122731495, "step": 5705, "time_per_iteration": 2.54496693611145 }, { "auxiliary_loss_clip": 0.01116065, "auxiliary_loss_mlp": 0.01022365, "balance_loss_clip": 1.04870963, "balance_loss_mlp": 1.0153017, "epoch": 0.6861059339866531, "flos": 19354526702880.0, "grad_norm": 3.038738303877478, "language_loss": 0.84023941, "learning_rate": 9.474083975077858e-07, "loss": 0.8616237, "num_input_tokens_seen": 122748620, "step": 5706, "time_per_iteration": 3.3449597358703613 }, { "auxiliary_loss_clip": 0.01151372, "auxiliary_loss_mlp": 0.01025229, "balance_loss_clip": 1.04690075, "balance_loss_mlp": 1.01748025, "epoch": 0.6862261768772921, "flos": 22199980573440.0, "grad_norm": 2.317589459761087, "language_loss": 0.80415016, "learning_rate": 9.467461165623994e-07, "loss": 0.82591617, "num_input_tokens_seen": 122767670, "step": 5707, "time_per_iteration": 3.2523415088653564 }, { "auxiliary_loss_clip": 0.01158865, "auxiliary_loss_mlp": 0.01021746, "balance_loss_clip": 1.04671693, "balance_loss_mlp": 1.0147512, "epoch": 0.6863464197679312, "flos": 26285681288640.0, "grad_norm": 1.936425526127595, "language_loss": 0.79767251, "learning_rate": 9.46083995394791e-07, "loss": 0.81947863, "num_input_tokens_seen": 122785480, "step": 5708, "time_per_iteration": 2.520376443862915 }, { "auxiliary_loss_clip": 0.01156269, "auxiliary_loss_mlp": 0.0076159, "balance_loss_clip": 1.04743123, "balance_loss_mlp": 1.00047064, "epoch": 0.6864666626585703, "flos": 37815233591040.0, "grad_norm": 3.627340470288116, "language_loss": 0.6322602, "learning_rate": 9.454220341054012e-07, "loss": 0.65143883, "num_input_tokens_seen": 122810265, "step": 5709, "time_per_iteration": 2.637619733810425 }, { "auxiliary_loss_clip": 0.01130346, "auxiliary_loss_mlp": 0.01021036, "balance_loss_clip": 1.0464927, "balance_loss_mlp": 1.01416373, "epoch": 0.6865869055492094, "flos": 19391155613760.0, "grad_norm": 1.8603811460735988, "language_loss": 0.80309272, "learning_rate": 9.447602327946512e-07, "loss": 0.8246066, "num_input_tokens_seen": 122828905, "step": 5710, "time_per_iteration": 2.532407283782959 }, { "auxiliary_loss_clip": 0.01140685, "auxiliary_loss_mlp": 0.01028154, "balance_loss_clip": 1.04391146, "balance_loss_mlp": 1.02091515, "epoch": 0.6867071484398485, "flos": 20375969840160.0, "grad_norm": 1.8577047619546259, "language_loss": 0.76685667, "learning_rate": 9.440985915629338e-07, "loss": 0.78854507, "num_input_tokens_seen": 122846235, "step": 5711, "time_per_iteration": 2.540524482727051 }, { "auxiliary_loss_clip": 0.0117107, "auxiliary_loss_mlp": 0.01019688, "balance_loss_clip": 1.05069077, "balance_loss_mlp": 1.01285148, "epoch": 0.6868273913304875, "flos": 15889146953280.0, "grad_norm": 1.951209272700785, "language_loss": 0.7295742, "learning_rate": 9.434371105106223e-07, "loss": 0.75148177, "num_input_tokens_seen": 122863835, "step": 5712, "time_per_iteration": 2.431464195251465 }, { "auxiliary_loss_clip": 0.01126464, "auxiliary_loss_mlp": 0.01028309, "balance_loss_clip": 1.04498267, "balance_loss_mlp": 1.02087665, "epoch": 0.6869476342211267, "flos": 24462496645440.0, "grad_norm": 1.6651729767983452, "language_loss": 0.70427138, "learning_rate": 9.427757897380602e-07, "loss": 0.72581911, "num_input_tokens_seen": 122883235, "step": 5713, "time_per_iteration": 2.602799892425537 }, { "auxiliary_loss_clip": 0.01125192, "auxiliary_loss_mlp": 0.01025294, "balance_loss_clip": 1.04475951, "balance_loss_mlp": 1.01855242, "epoch": 0.6870678771117658, "flos": 18442575211680.0, "grad_norm": 2.061642339166125, "language_loss": 0.8541342, "learning_rate": 9.421146293455695e-07, "loss": 0.87563908, "num_input_tokens_seen": 122898975, "step": 5714, "time_per_iteration": 2.5320467948913574 }, { "auxiliary_loss_clip": 0.01140384, "auxiliary_loss_mlp": 0.01027565, "balance_loss_clip": 1.04459143, "balance_loss_mlp": 1.02049601, "epoch": 0.6871881200024048, "flos": 22200375660000.0, "grad_norm": 1.8838979912125264, "language_loss": 0.68531036, "learning_rate": 9.414536294334489e-07, "loss": 0.70698988, "num_input_tokens_seen": 122918995, "step": 5715, "time_per_iteration": 2.5556023120880127 }, { "auxiliary_loss_clip": 0.01143669, "auxiliary_loss_mlp": 0.01023298, "balance_loss_clip": 1.04262328, "balance_loss_mlp": 1.0165087, "epoch": 0.687308362893044, "flos": 22127728426560.0, "grad_norm": 2.446510654808326, "language_loss": 0.69585836, "learning_rate": 9.407927901019708e-07, "loss": 0.71752799, "num_input_tokens_seen": 122938125, "step": 5716, "time_per_iteration": 3.2513270378112793 }, { "auxiliary_loss_clip": 0.0116183, "auxiliary_loss_mlp": 0.01023638, "balance_loss_clip": 1.04996419, "balance_loss_mlp": 1.01672673, "epoch": 0.687428605783683, "flos": 25040549737440.0, "grad_norm": 3.0343790195576976, "language_loss": 0.76842999, "learning_rate": 9.401321114513854e-07, "loss": 0.79028475, "num_input_tokens_seen": 122957020, "step": 5717, "time_per_iteration": 2.5462656021118164 }, { "auxiliary_loss_clip": 0.01173796, "auxiliary_loss_mlp": 0.0102963, "balance_loss_clip": 1.0504539, "balance_loss_mlp": 1.02213132, "epoch": 0.6875488486743221, "flos": 23770067200800.0, "grad_norm": 1.6523673063954887, "language_loss": 0.7536155, "learning_rate": 9.394715935819155e-07, "loss": 0.77564967, "num_input_tokens_seen": 122977410, "step": 5718, "time_per_iteration": 2.5117955207824707 }, { "auxiliary_loss_clip": 0.01162738, "auxiliary_loss_mlp": 0.01024851, "balance_loss_clip": 1.04900444, "balance_loss_mlp": 1.01754367, "epoch": 0.6876690915649613, "flos": 25516941080640.0, "grad_norm": 2.0396652689819303, "language_loss": 0.62570322, "learning_rate": 9.388112365937608e-07, "loss": 0.64757907, "num_input_tokens_seen": 122996875, "step": 5719, "time_per_iteration": 2.54535174369812 }, { "auxiliary_loss_clip": 0.01130271, "auxiliary_loss_mlp": 0.01025828, "balance_loss_clip": 1.0464499, "balance_loss_mlp": 1.0180074, "epoch": 0.6877893344556003, "flos": 19428000026400.0, "grad_norm": 2.1138811713323356, "language_loss": 0.82861382, "learning_rate": 9.381510405870985e-07, "loss": 0.85017478, "num_input_tokens_seen": 123015890, "step": 5720, "time_per_iteration": 2.55802583694458 }, { "auxiliary_loss_clip": 0.01153566, "auxiliary_loss_mlp": 0.01028243, "balance_loss_clip": 1.04533172, "balance_loss_mlp": 1.02130485, "epoch": 0.6879095773462394, "flos": 18661307085120.0, "grad_norm": 2.086008048504889, "language_loss": 0.77350295, "learning_rate": 9.374910056620791e-07, "loss": 0.79532105, "num_input_tokens_seen": 123034955, "step": 5721, "time_per_iteration": 2.466648817062378 }, { "auxiliary_loss_clip": 0.01163811, "auxiliary_loss_mlp": 0.01033315, "balance_loss_clip": 1.05178332, "balance_loss_mlp": 1.02587628, "epoch": 0.6880298202368785, "flos": 20883135629760.0, "grad_norm": 1.6841489623625938, "language_loss": 0.80847543, "learning_rate": 9.368311319188293e-07, "loss": 0.83044672, "num_input_tokens_seen": 123052770, "step": 5722, "time_per_iteration": 2.5031824111938477 }, { "auxiliary_loss_clip": 0.01129425, "auxiliary_loss_mlp": 0.01032531, "balance_loss_clip": 1.04452813, "balance_loss_mlp": 1.0252831, "epoch": 0.6881500631275176, "flos": 30153296635680.0, "grad_norm": 1.8382166832428988, "language_loss": 0.79383278, "learning_rate": 9.361714194574515e-07, "loss": 0.81545234, "num_input_tokens_seen": 123075105, "step": 5723, "time_per_iteration": 2.6198675632476807 }, { "auxiliary_loss_clip": 0.01067758, "auxiliary_loss_mlp": 0.01001094, "balance_loss_clip": 1.00976551, "balance_loss_mlp": 0.99992603, "epoch": 0.6882703060181566, "flos": 66181549879200.0, "grad_norm": 0.7345366466274087, "language_loss": 0.58253908, "learning_rate": 9.355118683780228e-07, "loss": 0.60322762, "num_input_tokens_seen": 123145175, "step": 5724, "time_per_iteration": 3.1759033203125 }, { "auxiliary_loss_clip": 0.01173331, "auxiliary_loss_mlp": 0.01028268, "balance_loss_clip": 1.05036783, "balance_loss_mlp": 1.02133298, "epoch": 0.6883905489087958, "flos": 18214648596480.0, "grad_norm": 1.9372623998762952, "language_loss": 0.78741217, "learning_rate": 9.348524787805987e-07, "loss": 0.80942822, "num_input_tokens_seen": 123160365, "step": 5725, "time_per_iteration": 2.443392038345337 }, { "auxiliary_loss_clip": 0.01129702, "auxiliary_loss_mlp": 0.01023754, "balance_loss_clip": 1.04100072, "balance_loss_mlp": 1.0165453, "epoch": 0.6885107917994349, "flos": 14056264730880.0, "grad_norm": 2.849077870575039, "language_loss": 0.85238934, "learning_rate": 9.341932507652053e-07, "loss": 0.8739239, "num_input_tokens_seen": 123174855, "step": 5726, "time_per_iteration": 2.653163194656372 }, { "auxiliary_loss_clip": 0.01139963, "auxiliary_loss_mlp": 0.01028926, "balance_loss_clip": 1.04215622, "balance_loss_mlp": 1.02167463, "epoch": 0.6886310346900739, "flos": 28690726221600.0, "grad_norm": 1.7828941293519132, "language_loss": 0.78802329, "learning_rate": 9.335341844318489e-07, "loss": 0.80971217, "num_input_tokens_seen": 123194995, "step": 5727, "time_per_iteration": 2.571467876434326 }, { "auxiliary_loss_clip": 0.01143526, "auxiliary_loss_mlp": 0.0102727, "balance_loss_clip": 1.04721904, "balance_loss_mlp": 1.02023387, "epoch": 0.6887512775807131, "flos": 24535323463680.0, "grad_norm": 1.6895497677819806, "language_loss": 0.73118436, "learning_rate": 9.328752798805091e-07, "loss": 0.75289232, "num_input_tokens_seen": 123213465, "step": 5728, "time_per_iteration": 2.5457656383514404 }, { "auxiliary_loss_clip": 0.01158807, "auxiliary_loss_mlp": 0.01022917, "balance_loss_clip": 1.04931474, "balance_loss_mlp": 1.01602972, "epoch": 0.6888715204713521, "flos": 22414366494720.0, "grad_norm": 2.0992230275870853, "language_loss": 0.75751978, "learning_rate": 9.322165372111399e-07, "loss": 0.77933705, "num_input_tokens_seen": 123231610, "step": 5729, "time_per_iteration": 3.3488805294036865 }, { "auxiliary_loss_clip": 0.01128683, "auxiliary_loss_mlp": 0.01027526, "balance_loss_clip": 1.04807663, "balance_loss_mlp": 1.02078187, "epoch": 0.6889917633619912, "flos": 22054326936960.0, "grad_norm": 1.8817392330888483, "language_loss": 0.75428045, "learning_rate": 9.315579565236747e-07, "loss": 0.77584255, "num_input_tokens_seen": 123250715, "step": 5730, "time_per_iteration": 2.5645971298217773 }, { "auxiliary_loss_clip": 0.01140978, "auxiliary_loss_mlp": 0.01032047, "balance_loss_clip": 1.04883409, "balance_loss_mlp": 1.0247333, "epoch": 0.6891120062526304, "flos": 23949727810080.0, "grad_norm": 1.7430353766705575, "language_loss": 0.73714924, "learning_rate": 9.308995379180162e-07, "loss": 0.75887942, "num_input_tokens_seen": 123270270, "step": 5731, "time_per_iteration": 2.553793430328369 }, { "auxiliary_loss_clip": 0.01058528, "auxiliary_loss_mlp": 0.01001046, "balance_loss_clip": 1.00968099, "balance_loss_mlp": 0.99979991, "epoch": 0.6892322491432694, "flos": 64117364847360.0, "grad_norm": 0.7367882437177322, "language_loss": 0.59540021, "learning_rate": 9.302412814940488e-07, "loss": 0.615996, "num_input_tokens_seen": 123333045, "step": 5732, "time_per_iteration": 3.890312910079956 }, { "auxiliary_loss_clip": 0.01140388, "auxiliary_loss_mlp": 0.01026126, "balance_loss_clip": 1.04424083, "balance_loss_mlp": 1.01896191, "epoch": 0.6893524920339085, "flos": 23002440418560.0, "grad_norm": 2.051764249767517, "language_loss": 0.70874441, "learning_rate": 9.295831873516276e-07, "loss": 0.7304095, "num_input_tokens_seen": 123352320, "step": 5733, "time_per_iteration": 3.288655996322632 }, { "auxiliary_loss_clip": 0.01173958, "auxiliary_loss_mlp": 0.01029246, "balance_loss_clip": 1.05183148, "balance_loss_mlp": 1.02227211, "epoch": 0.6894727349245476, "flos": 21396263634720.0, "grad_norm": 1.6316132076305316, "language_loss": 0.76084632, "learning_rate": 9.289252555905873e-07, "loss": 0.7828784, "num_input_tokens_seen": 123372400, "step": 5734, "time_per_iteration": 2.490537166595459 }, { "auxiliary_loss_clip": 0.01161484, "auxiliary_loss_mlp": 0.01027566, "balance_loss_clip": 1.05173182, "balance_loss_mlp": 1.0204587, "epoch": 0.6895929778151867, "flos": 19865320105440.0, "grad_norm": 6.9329461592698545, "language_loss": 0.76098299, "learning_rate": 9.282674863107334e-07, "loss": 0.78287351, "num_input_tokens_seen": 123390215, "step": 5735, "time_per_iteration": 2.487889528274536 }, { "auxiliary_loss_clip": 0.01156376, "auxiliary_loss_mlp": 0.0102501, "balance_loss_clip": 1.05132329, "balance_loss_mlp": 1.0179739, "epoch": 0.6897132207058257, "flos": 18179169028320.0, "grad_norm": 2.1367136038511068, "language_loss": 0.75619137, "learning_rate": 9.276098796118488e-07, "loss": 0.77800524, "num_input_tokens_seen": 123406700, "step": 5736, "time_per_iteration": 2.4653515815734863 }, { "auxiliary_loss_clip": 0.01143602, "auxiliary_loss_mlp": 0.01024886, "balance_loss_clip": 1.04770172, "balance_loss_mlp": 1.01806426, "epoch": 0.6898334635964649, "flos": 32561645928960.0, "grad_norm": 1.8619017672238025, "language_loss": 0.66079807, "learning_rate": 9.269524355936938e-07, "loss": 0.6824829, "num_input_tokens_seen": 123429880, "step": 5737, "time_per_iteration": 2.6052441596984863 }, { "auxiliary_loss_clip": 0.01135077, "auxiliary_loss_mlp": 0.0102739, "balance_loss_clip": 1.04296756, "balance_loss_mlp": 1.02037406, "epoch": 0.689953706487104, "flos": 22819008528480.0, "grad_norm": 1.6796198871747499, "language_loss": 0.85142696, "learning_rate": 9.262951543560002e-07, "loss": 0.87305164, "num_input_tokens_seen": 123449105, "step": 5738, "time_per_iteration": 2.5445144176483154 }, { "auxiliary_loss_clip": 0.01146644, "auxiliary_loss_mlp": 0.01022523, "balance_loss_clip": 1.05110812, "balance_loss_mlp": 1.01478004, "epoch": 0.690073949377743, "flos": 18515366112960.0, "grad_norm": 2.2996535421220936, "language_loss": 0.85962373, "learning_rate": 9.256380359984795e-07, "loss": 0.88131535, "num_input_tokens_seen": 123466215, "step": 5739, "time_per_iteration": 2.488802671432495 }, { "auxiliary_loss_clip": 0.01122012, "auxiliary_loss_mlp": 0.01027546, "balance_loss_clip": 1.04008079, "balance_loss_mlp": 1.02059066, "epoch": 0.6901941922683821, "flos": 34857199215840.0, "grad_norm": 2.691492086576804, "language_loss": 0.74459207, "learning_rate": 9.249810806208139e-07, "loss": 0.76608765, "num_input_tokens_seen": 123485480, "step": 5740, "time_per_iteration": 2.7165305614471436 }, { "auxiliary_loss_clip": 0.01112455, "auxiliary_loss_mlp": 0.00761696, "balance_loss_clip": 1.03923202, "balance_loss_mlp": 1.00058734, "epoch": 0.6903144351590212, "flos": 16253676131040.0, "grad_norm": 1.7868404851472652, "language_loss": 0.79909527, "learning_rate": 9.243242883226627e-07, "loss": 0.81783682, "num_input_tokens_seen": 123504575, "step": 5741, "time_per_iteration": 3.3230178356170654 }, { "auxiliary_loss_clip": 0.01160993, "auxiliary_loss_mlp": 0.01027616, "balance_loss_clip": 1.04583764, "balance_loss_mlp": 1.02023709, "epoch": 0.6904346780496603, "flos": 28035141189600.0, "grad_norm": 1.895341511797567, "language_loss": 0.69623709, "learning_rate": 9.236676592036628e-07, "loss": 0.71812308, "num_input_tokens_seen": 123524250, "step": 5742, "time_per_iteration": 2.5649373531341553 }, { "auxiliary_loss_clip": 0.01142164, "auxiliary_loss_mlp": 0.01024258, "balance_loss_clip": 1.049263, "balance_loss_mlp": 1.01718879, "epoch": 0.6905549209402994, "flos": 23624269896480.0, "grad_norm": 1.5740727854713954, "language_loss": 0.73216331, "learning_rate": 9.230111933634228e-07, "loss": 0.75382751, "num_input_tokens_seen": 123545845, "step": 5743, "time_per_iteration": 2.550928831100464 }, { "auxiliary_loss_clip": 0.01162599, "auxiliary_loss_mlp": 0.01030714, "balance_loss_clip": 1.05176771, "balance_loss_mlp": 1.02366829, "epoch": 0.6906751638309385, "flos": 23114949089280.0, "grad_norm": 1.6139854644922844, "language_loss": 0.80822814, "learning_rate": 9.223548909015288e-07, "loss": 0.83016121, "num_input_tokens_seen": 123567535, "step": 5744, "time_per_iteration": 2.54473614692688 }, { "auxiliary_loss_clip": 0.01109185, "auxiliary_loss_mlp": 0.01023647, "balance_loss_clip": 1.0432775, "balance_loss_mlp": 1.01682258, "epoch": 0.6907954067215776, "flos": 27305472245760.0, "grad_norm": 1.985276201371435, "language_loss": 0.71864915, "learning_rate": 9.216987519175407e-07, "loss": 0.73997748, "num_input_tokens_seen": 123587710, "step": 5745, "time_per_iteration": 2.6340603828430176 }, { "auxiliary_loss_clip": 0.01152684, "auxiliary_loss_mlp": 0.01024189, "balance_loss_clip": 1.04925656, "balance_loss_mlp": 1.01714659, "epoch": 0.6909156496122166, "flos": 21689402672640.0, "grad_norm": 2.0529418611967905, "language_loss": 0.68494964, "learning_rate": 9.210427765109942e-07, "loss": 0.70671833, "num_input_tokens_seen": 123607385, "step": 5746, "time_per_iteration": 2.497575521469116 }, { "auxiliary_loss_clip": 0.01146846, "auxiliary_loss_mlp": 0.01027658, "balance_loss_clip": 1.04621029, "balance_loss_mlp": 1.02043414, "epoch": 0.6910358925028558, "flos": 22561456809600.0, "grad_norm": 1.8006876195974606, "language_loss": 0.81352389, "learning_rate": 9.20386964781402e-07, "loss": 0.83526897, "num_input_tokens_seen": 123625405, "step": 5747, "time_per_iteration": 2.559518814086914 }, { "auxiliary_loss_clip": 0.01140003, "auxiliary_loss_mlp": 0.01027325, "balance_loss_clip": 1.04549932, "balance_loss_mlp": 1.02029133, "epoch": 0.6911561353934949, "flos": 22054111435200.0, "grad_norm": 2.410338059435435, "language_loss": 0.84359735, "learning_rate": 9.197313168282472e-07, "loss": 0.86527067, "num_input_tokens_seen": 123642850, "step": 5748, "time_per_iteration": 2.5512096881866455 }, { "auxiliary_loss_clip": 0.01151829, "auxiliary_loss_mlp": 0.01028807, "balance_loss_clip": 1.04499865, "balance_loss_mlp": 1.02151728, "epoch": 0.6912763782841339, "flos": 24206561189760.0, "grad_norm": 2.5416582549741356, "language_loss": 0.72147334, "learning_rate": 9.190758327509935e-07, "loss": 0.7432797, "num_input_tokens_seen": 123661595, "step": 5749, "time_per_iteration": 2.525831937789917 }, { "auxiliary_loss_clip": 0.0102448, "auxiliary_loss_mlp": 0.00753123, "balance_loss_clip": 1.00914645, "balance_loss_mlp": 1.00017381, "epoch": 0.6913966211747731, "flos": 52329650237760.0, "grad_norm": 0.9326214498236789, "language_loss": 0.64444041, "learning_rate": 9.184205126490767e-07, "loss": 0.66221642, "num_input_tokens_seen": 123710490, "step": 5750, "time_per_iteration": 2.9812443256378174 }, { "auxiliary_loss_clip": 0.01037723, "auxiliary_loss_mlp": 0.00753039, "balance_loss_clip": 1.01028228, "balance_loss_mlp": 1.00007939, "epoch": 0.6915168640654121, "flos": 66741285321600.0, "grad_norm": 1.0604442685991322, "language_loss": 0.59711838, "learning_rate": 9.177653566219075e-07, "loss": 0.615026, "num_input_tokens_seen": 123765215, "step": 5751, "time_per_iteration": 3.045989751815796 }, { "auxiliary_loss_clip": 0.01132931, "auxiliary_loss_mlp": 0.01033906, "balance_loss_clip": 1.04332376, "balance_loss_mlp": 1.02682495, "epoch": 0.6916371069560512, "flos": 18296526488640.0, "grad_norm": 1.9257616744944512, "language_loss": 0.76087844, "learning_rate": 9.171103647688744e-07, "loss": 0.78254682, "num_input_tokens_seen": 123783955, "step": 5752, "time_per_iteration": 2.5402462482452393 }, { "auxiliary_loss_clip": 0.01080219, "auxiliary_loss_mlp": 0.0102751, "balance_loss_clip": 1.0399797, "balance_loss_mlp": 1.02087927, "epoch": 0.6917573498466904, "flos": 19645798058880.0, "grad_norm": 1.7982432745402372, "language_loss": 0.69495869, "learning_rate": 9.164555371893367e-07, "loss": 0.71603596, "num_input_tokens_seen": 123803885, "step": 5753, "time_per_iteration": 2.6434996128082275 }, { "auxiliary_loss_clip": 0.01158457, "auxiliary_loss_mlp": 0.00762363, "balance_loss_clip": 1.05025816, "balance_loss_mlp": 1.00050831, "epoch": 0.6918775927373294, "flos": 14210322936000.0, "grad_norm": 1.9038191413354733, "language_loss": 0.75249487, "learning_rate": 9.158008739826333e-07, "loss": 0.77170306, "num_input_tokens_seen": 123821485, "step": 5754, "time_per_iteration": 2.48860502243042 }, { "auxiliary_loss_clip": 0.01143737, "auxiliary_loss_mlp": 0.01027843, "balance_loss_clip": 1.048877, "balance_loss_mlp": 1.01978445, "epoch": 0.6919978356279685, "flos": 23985458796960.0, "grad_norm": 1.557547790769456, "language_loss": 0.86676794, "learning_rate": 9.151463752480744e-07, "loss": 0.88848376, "num_input_tokens_seen": 123840215, "step": 5755, "time_per_iteration": 3.312346935272217 }, { "auxiliary_loss_clip": 0.01121693, "auxiliary_loss_mlp": 0.01027887, "balance_loss_clip": 1.0448463, "balance_loss_mlp": 1.02104712, "epoch": 0.6921180785186076, "flos": 23622940968960.0, "grad_norm": 1.5159197699541243, "language_loss": 0.80362105, "learning_rate": 9.144920410849493e-07, "loss": 0.82511681, "num_input_tokens_seen": 123861450, "step": 5756, "time_per_iteration": 2.563119649887085 }, { "auxiliary_loss_clip": 0.01149749, "auxiliary_loss_mlp": 0.0102464, "balance_loss_clip": 1.04769874, "balance_loss_mlp": 1.01752281, "epoch": 0.6922383214092467, "flos": 21142626864480.0, "grad_norm": 1.6762985222135787, "language_loss": 0.80392718, "learning_rate": 9.138378715925176e-07, "loss": 0.82567108, "num_input_tokens_seen": 123880545, "step": 5757, "time_per_iteration": 2.5572116374969482 }, { "auxiliary_loss_clip": 0.0113531, "auxiliary_loss_mlp": 0.01027588, "balance_loss_clip": 1.04420972, "balance_loss_mlp": 1.02079272, "epoch": 0.6923585642998857, "flos": 21470670799200.0, "grad_norm": 2.2540330922419356, "language_loss": 0.8137697, "learning_rate": 9.131838668700167e-07, "loss": 0.83539867, "num_input_tokens_seen": 123900615, "step": 5758, "time_per_iteration": 3.28902530670166 }, { "auxiliary_loss_clip": 0.01126464, "auxiliary_loss_mlp": 0.01024068, "balance_loss_clip": 1.0409621, "balance_loss_mlp": 1.01713252, "epoch": 0.6924788071905249, "flos": 21105207780480.0, "grad_norm": 1.8335628854378005, "language_loss": 0.86455363, "learning_rate": 9.125300270166598e-07, "loss": 0.88605899, "num_input_tokens_seen": 123921220, "step": 5759, "time_per_iteration": 3.3720157146453857 }, { "auxiliary_loss_clip": 0.01131742, "auxiliary_loss_mlp": 0.01021161, "balance_loss_clip": 1.04290557, "balance_loss_mlp": 1.01397228, "epoch": 0.692599050081164, "flos": 26250022135680.0, "grad_norm": 1.71724051070998, "language_loss": 0.855259, "learning_rate": 9.118763521316324e-07, "loss": 0.87678802, "num_input_tokens_seen": 123941795, "step": 5760, "time_per_iteration": 2.604654550552368 }, { "auxiliary_loss_clip": 0.01171875, "auxiliary_loss_mlp": 0.00761968, "balance_loss_clip": 1.0491792, "balance_loss_mlp": 1.00058293, "epoch": 0.692719292971803, "flos": 20885218813440.0, "grad_norm": 1.6175457737895869, "language_loss": 0.76256716, "learning_rate": 9.112228423140987e-07, "loss": 0.78190565, "num_input_tokens_seen": 123960715, "step": 5761, "time_per_iteration": 2.454845428466797 }, { "auxiliary_loss_clip": 0.01148189, "auxiliary_loss_mlp": 0.01026394, "balance_loss_clip": 1.04749894, "balance_loss_mlp": 1.01904225, "epoch": 0.6928395358624422, "flos": 25921942284000.0, "grad_norm": 2.257496602031366, "language_loss": 0.86457992, "learning_rate": 9.105694976631932e-07, "loss": 0.88632578, "num_input_tokens_seen": 123978625, "step": 5762, "time_per_iteration": 2.562068462371826 }, { "auxiliary_loss_clip": 0.01156141, "auxiliary_loss_mlp": 0.01027241, "balance_loss_clip": 1.04845762, "balance_loss_mlp": 1.02007031, "epoch": 0.6929597787530812, "flos": 23586563476800.0, "grad_norm": 2.244866692345435, "language_loss": 0.72696149, "learning_rate": 9.099163182780283e-07, "loss": 0.74879533, "num_input_tokens_seen": 123996780, "step": 5763, "time_per_iteration": 2.497640371322632 }, { "auxiliary_loss_clip": 0.01142545, "auxiliary_loss_mlp": 0.01023169, "balance_loss_clip": 1.04821908, "balance_loss_mlp": 1.01617742, "epoch": 0.6930800216437203, "flos": 18255659376480.0, "grad_norm": 3.0655920066406916, "language_loss": 0.49138331, "learning_rate": 9.092633042576916e-07, "loss": 0.51304048, "num_input_tokens_seen": 124014045, "step": 5764, "time_per_iteration": 2.4997291564941406 }, { "auxiliary_loss_clip": 0.01138951, "auxiliary_loss_mlp": 0.01029337, "balance_loss_clip": 1.04673409, "balance_loss_mlp": 1.02226472, "epoch": 0.6932002645343595, "flos": 29168626077120.0, "grad_norm": 1.9689065297594703, "language_loss": 0.56387138, "learning_rate": 9.086104557012446e-07, "loss": 0.58555424, "num_input_tokens_seen": 124034615, "step": 5765, "time_per_iteration": 2.5748863220214844 }, { "auxiliary_loss_clip": 0.01149551, "auxiliary_loss_mlp": 0.01025775, "balance_loss_clip": 1.04647589, "balance_loss_mlp": 1.01890516, "epoch": 0.6933205074249985, "flos": 23842750351200.0, "grad_norm": 4.092719734237994, "language_loss": 0.65569794, "learning_rate": 9.079577727077239e-07, "loss": 0.67745113, "num_input_tokens_seen": 124053445, "step": 5766, "time_per_iteration": 2.5200681686401367 }, { "auxiliary_loss_clip": 0.01157525, "auxiliary_loss_mlp": 0.01027326, "balance_loss_clip": 1.04849005, "balance_loss_mlp": 1.02017939, "epoch": 0.6934407503156376, "flos": 24166699752480.0, "grad_norm": 2.917435390684062, "language_loss": 0.72054762, "learning_rate": 9.073052553761404e-07, "loss": 0.74239612, "num_input_tokens_seen": 124072810, "step": 5767, "time_per_iteration": 3.255655288696289 }, { "auxiliary_loss_clip": 0.01116725, "auxiliary_loss_mlp": 0.01030166, "balance_loss_clip": 1.04368794, "balance_loss_mlp": 1.02198231, "epoch": 0.6935609932062767, "flos": 20631330624480.0, "grad_norm": 1.7540813917845, "language_loss": 0.7806592, "learning_rate": 9.066529038054805e-07, "loss": 0.80212814, "num_input_tokens_seen": 124092875, "step": 5768, "time_per_iteration": 2.598505735397339 }, { "auxiliary_loss_clip": 0.01142366, "auxiliary_loss_mlp": 0.01026094, "balance_loss_clip": 1.04678798, "balance_loss_mlp": 1.01940632, "epoch": 0.6936812360969158, "flos": 18254186781120.0, "grad_norm": 1.6976788095114252, "language_loss": 0.73937404, "learning_rate": 9.060007180947071e-07, "loss": 0.76105869, "num_input_tokens_seen": 124110930, "step": 5769, "time_per_iteration": 2.5085995197296143 }, { "auxiliary_loss_clip": 0.0110998, "auxiliary_loss_mlp": 0.01030638, "balance_loss_clip": 1.03931534, "balance_loss_mlp": 1.02362013, "epoch": 0.6938014789875548, "flos": 31317340467840.0, "grad_norm": 1.7439805273690694, "language_loss": 0.73433244, "learning_rate": 9.053486983427534e-07, "loss": 0.75573862, "num_input_tokens_seen": 124132180, "step": 5770, "time_per_iteration": 2.684971570968628 }, { "auxiliary_loss_clip": 0.01148284, "auxiliary_loss_mlp": 0.01023089, "balance_loss_clip": 1.04540634, "balance_loss_mlp": 1.01601362, "epoch": 0.6939217218781939, "flos": 17528432785920.0, "grad_norm": 1.7358008261516966, "language_loss": 0.70472765, "learning_rate": 9.046968446485326e-07, "loss": 0.72644138, "num_input_tokens_seen": 124150585, "step": 5771, "time_per_iteration": 2.5057015419006348 }, { "auxiliary_loss_clip": 0.01163098, "auxiliary_loss_mlp": 0.01025939, "balance_loss_clip": 1.05046153, "balance_loss_mlp": 1.01855445, "epoch": 0.6940419647688331, "flos": 18551779522080.0, "grad_norm": 2.5271102237360727, "language_loss": 0.708601, "learning_rate": 9.040451571109295e-07, "loss": 0.7304914, "num_input_tokens_seen": 124166205, "step": 5772, "time_per_iteration": 2.4751060009002686 }, { "auxiliary_loss_clip": 0.01047247, "auxiliary_loss_mlp": 0.01006275, "balance_loss_clip": 1.02551913, "balance_loss_mlp": 1.00504136, "epoch": 0.6941622076594721, "flos": 66926297557920.0, "grad_norm": 0.9098886681687236, "language_loss": 0.60456884, "learning_rate": 9.033936358288042e-07, "loss": 0.62510407, "num_input_tokens_seen": 124219940, "step": 5773, "time_per_iteration": 3.037073850631714 }, { "auxiliary_loss_clip": 0.01170691, "auxiliary_loss_mlp": 0.01025553, "balance_loss_clip": 1.04803443, "balance_loss_mlp": 1.01865101, "epoch": 0.6942824505501112, "flos": 26578066070400.0, "grad_norm": 1.6590008384153794, "language_loss": 0.82487142, "learning_rate": 9.027422809009937e-07, "loss": 0.84683388, "num_input_tokens_seen": 124239885, "step": 5774, "time_per_iteration": 2.506715774536133 }, { "auxiliary_loss_clip": 0.01159112, "auxiliary_loss_mlp": 0.01024529, "balance_loss_clip": 1.04574275, "balance_loss_mlp": 1.01754367, "epoch": 0.6944026934407503, "flos": 21248311312800.0, "grad_norm": 1.5793958327986377, "language_loss": 0.8319943, "learning_rate": 9.020910924263054e-07, "loss": 0.8538307, "num_input_tokens_seen": 124258410, "step": 5775, "time_per_iteration": 2.4957828521728516 }, { "auxiliary_loss_clip": 0.01043131, "auxiliary_loss_mlp": 0.01005745, "balance_loss_clip": 1.02326906, "balance_loss_mlp": 1.0045706, "epoch": 0.6945229363313894, "flos": 70677201949920.0, "grad_norm": 0.8182202505559012, "language_loss": 0.58195567, "learning_rate": 9.014400705035261e-07, "loss": 0.60244441, "num_input_tokens_seen": 124315315, "step": 5776, "time_per_iteration": 3.188506603240967 }, { "auxiliary_loss_clip": 0.01172772, "auxiliary_loss_mlp": 0.01025773, "balance_loss_clip": 1.05183923, "balance_loss_mlp": 1.01874828, "epoch": 0.6946431792220285, "flos": 18952937610720.0, "grad_norm": 1.895961803296539, "language_loss": 0.76033401, "learning_rate": 9.00789215231414e-07, "loss": 0.78231955, "num_input_tokens_seen": 124333710, "step": 5777, "time_per_iteration": 2.4532368183135986 }, { "auxiliary_loss_clip": 0.01126398, "auxiliary_loss_mlp": 0.00762848, "balance_loss_clip": 1.04006231, "balance_loss_mlp": 1.00061131, "epoch": 0.6947634221126676, "flos": 20338838091840.0, "grad_norm": 1.6815526925250561, "language_loss": 0.81631172, "learning_rate": 9.001385267087056e-07, "loss": 0.83520424, "num_input_tokens_seen": 124352855, "step": 5778, "time_per_iteration": 2.5562376976013184 }, { "auxiliary_loss_clip": 0.01162063, "auxiliary_loss_mlp": 0.01024759, "balance_loss_clip": 1.05014944, "balance_loss_mlp": 1.017609, "epoch": 0.6948836650033067, "flos": 21833727381600.0, "grad_norm": 1.562698011045386, "language_loss": 0.70388258, "learning_rate": 8.994880050341072e-07, "loss": 0.72575074, "num_input_tokens_seen": 124372960, "step": 5779, "time_per_iteration": 2.5126123428344727 }, { "auxiliary_loss_clip": 0.01137379, "auxiliary_loss_mlp": 0.01033871, "balance_loss_clip": 1.04589522, "balance_loss_mlp": 1.02656305, "epoch": 0.6950039078939457, "flos": 23657522613120.0, "grad_norm": 1.9172108093204063, "language_loss": 0.77472937, "learning_rate": 8.988376503063026e-07, "loss": 0.79644191, "num_input_tokens_seen": 124394220, "step": 5780, "time_per_iteration": 2.57454776763916 }, { "auxiliary_loss_clip": 0.01123281, "auxiliary_loss_mlp": 0.01022892, "balance_loss_clip": 1.04518092, "balance_loss_mlp": 1.01504779, "epoch": 0.6951241507845849, "flos": 21792465182880.0, "grad_norm": 2.070414734019755, "language_loss": 0.81544805, "learning_rate": 8.981874626239521e-07, "loss": 0.83690977, "num_input_tokens_seen": 124412795, "step": 5781, "time_per_iteration": 3.421635866165161 }, { "auxiliary_loss_clip": 0.01160703, "auxiliary_loss_mlp": 0.0102872, "balance_loss_clip": 1.05162406, "balance_loss_mlp": 1.02151656, "epoch": 0.695244393675224, "flos": 14647571181120.0, "grad_norm": 2.2091236378786214, "language_loss": 0.8835175, "learning_rate": 8.975374420856872e-07, "loss": 0.90541172, "num_input_tokens_seen": 124429690, "step": 5782, "time_per_iteration": 2.4667577743530273 }, { "auxiliary_loss_clip": 0.01117289, "auxiliary_loss_mlp": 0.01020154, "balance_loss_clip": 1.04021788, "balance_loss_mlp": 1.01348984, "epoch": 0.695364636565863, "flos": 16873206923520.0, "grad_norm": 2.2648747812282113, "language_loss": 0.7230581, "learning_rate": 8.968875887901157e-07, "loss": 0.74443245, "num_input_tokens_seen": 124447070, "step": 5783, "time_per_iteration": 2.5395402908325195 }, { "auxiliary_loss_clip": 0.01144184, "auxiliary_loss_mlp": 0.01030725, "balance_loss_clip": 1.04478061, "balance_loss_mlp": 1.02345586, "epoch": 0.6954848794565022, "flos": 19354526702880.0, "grad_norm": 1.9384314619787215, "language_loss": 0.63066924, "learning_rate": 8.9623790283582e-07, "loss": 0.65241838, "num_input_tokens_seen": 124464950, "step": 5784, "time_per_iteration": 3.289111375808716 }, { "auxiliary_loss_clip": 0.01130552, "auxiliary_loss_mlp": 0.01031179, "balance_loss_clip": 1.04671884, "balance_loss_mlp": 1.02387726, "epoch": 0.6956051223471412, "flos": 18990213026880.0, "grad_norm": 2.3559862778739067, "language_loss": 0.76182115, "learning_rate": 8.955883843213561e-07, "loss": 0.7834385, "num_input_tokens_seen": 124483965, "step": 5785, "time_per_iteration": 3.3073084354400635 }, { "auxiliary_loss_clip": 0.0116608, "auxiliary_loss_mlp": 0.01025448, "balance_loss_clip": 1.04934108, "balance_loss_mlp": 1.01836729, "epoch": 0.6957253652377803, "flos": 16107232321440.0, "grad_norm": 1.9116414470007235, "language_loss": 0.87081122, "learning_rate": 8.949390333452569e-07, "loss": 0.89272648, "num_input_tokens_seen": 124501910, "step": 5786, "time_per_iteration": 2.485724449157715 }, { "auxiliary_loss_clip": 0.01172074, "auxiliary_loss_mlp": 0.01024526, "balance_loss_clip": 1.05033672, "balance_loss_mlp": 1.01748669, "epoch": 0.6958456081284194, "flos": 29388650961120.0, "grad_norm": 2.0726089851867586, "language_loss": 0.67996413, "learning_rate": 8.942898500060279e-07, "loss": 0.70193011, "num_input_tokens_seen": 124521625, "step": 5787, "time_per_iteration": 2.5090479850769043 }, { "auxiliary_loss_clip": 0.01121758, "auxiliary_loss_mlp": 0.01023361, "balance_loss_clip": 1.04682517, "balance_loss_mlp": 1.01566029, "epoch": 0.6959658510190585, "flos": 25154854256160.0, "grad_norm": 2.2925875516634546, "language_loss": 0.7222544, "learning_rate": 8.936408344021493e-07, "loss": 0.74370557, "num_input_tokens_seen": 124538540, "step": 5788, "time_per_iteration": 2.6222028732299805 }, { "auxiliary_loss_clip": 0.01150766, "auxiliary_loss_mlp": 0.01029208, "balance_loss_clip": 1.05011344, "balance_loss_mlp": 1.0218116, "epoch": 0.6960860939096976, "flos": 42814394309760.0, "grad_norm": 2.096310489603306, "language_loss": 0.70912135, "learning_rate": 8.929919866320765e-07, "loss": 0.73092109, "num_input_tokens_seen": 124559355, "step": 5789, "time_per_iteration": 2.709017038345337 }, { "auxiliary_loss_clip": 0.01135077, "auxiliary_loss_mlp": 0.00762428, "balance_loss_clip": 1.04358101, "balance_loss_mlp": 1.00055075, "epoch": 0.6962063368003367, "flos": 17566570209120.0, "grad_norm": 1.8206264694641803, "language_loss": 0.81673789, "learning_rate": 8.923433067942385e-07, "loss": 0.83571291, "num_input_tokens_seen": 124577920, "step": 5790, "time_per_iteration": 2.559743881225586 }, { "auxiliary_loss_clip": 0.0113377, "auxiliary_loss_mlp": 0.01028489, "balance_loss_clip": 1.0455327, "balance_loss_mlp": 1.02136636, "epoch": 0.6963265796909758, "flos": 21251651590080.0, "grad_norm": 2.154782731172655, "language_loss": 0.68665153, "learning_rate": 8.916947949870417e-07, "loss": 0.70827413, "num_input_tokens_seen": 124597585, "step": 5791, "time_per_iteration": 2.5533902645111084 }, { "auxiliary_loss_clip": 0.01058175, "auxiliary_loss_mlp": 0.01004062, "balance_loss_clip": 1.00854063, "balance_loss_mlp": 1.00277472, "epoch": 0.6964468225816148, "flos": 68828307151680.0, "grad_norm": 0.7511142407313649, "language_loss": 0.5821048, "learning_rate": 8.910464513088615e-07, "loss": 0.60272717, "num_input_tokens_seen": 124661625, "step": 5792, "time_per_iteration": 3.1899352073669434 }, { "auxiliary_loss_clip": 0.01135921, "auxiliary_loss_mlp": 0.01025458, "balance_loss_clip": 1.04425251, "balance_loss_mlp": 1.01812673, "epoch": 0.696567065472254, "flos": 18950890344000.0, "grad_norm": 1.7775882923153956, "language_loss": 0.78355801, "learning_rate": 8.903982758580542e-07, "loss": 0.80517185, "num_input_tokens_seen": 124680565, "step": 5793, "time_per_iteration": 3.2537174224853516 }, { "auxiliary_loss_clip": 0.01142015, "auxiliary_loss_mlp": 0.01026542, "balance_loss_clip": 1.04732823, "balance_loss_mlp": 1.01965761, "epoch": 0.696687308362893, "flos": 22856679031200.0, "grad_norm": 2.117940380250935, "language_loss": 0.80291557, "learning_rate": 8.897502687329457e-07, "loss": 0.82460111, "num_input_tokens_seen": 124700365, "step": 5794, "time_per_iteration": 2.5429091453552246 }, { "auxiliary_loss_clip": 0.01125003, "auxiliary_loss_mlp": 0.01027183, "balance_loss_clip": 1.04305625, "balance_loss_mlp": 1.01988173, "epoch": 0.6968075512535321, "flos": 24972930878400.0, "grad_norm": 1.8812200617638333, "language_loss": 0.79856515, "learning_rate": 8.891024300318382e-07, "loss": 0.82008702, "num_input_tokens_seen": 124718935, "step": 5795, "time_per_iteration": 2.6478111743927 }, { "auxiliary_loss_clip": 0.0111861, "auxiliary_loss_mlp": 0.01024745, "balance_loss_clip": 1.04070854, "balance_loss_mlp": 1.01788402, "epoch": 0.6969277941441713, "flos": 21030441446400.0, "grad_norm": 1.3880943184603012, "language_loss": 0.75737083, "learning_rate": 8.884547598530103e-07, "loss": 0.77880442, "num_input_tokens_seen": 124739505, "step": 5796, "time_per_iteration": 2.5602736473083496 }, { "auxiliary_loss_clip": 0.01078959, "auxiliary_loss_mlp": 0.01023324, "balance_loss_clip": 1.04071379, "balance_loss_mlp": 1.01644588, "epoch": 0.6970480370348103, "flos": 21579408189120.0, "grad_norm": 1.9513401535123203, "language_loss": 0.75116891, "learning_rate": 8.8780725829471e-07, "loss": 0.77219176, "num_input_tokens_seen": 124757410, "step": 5797, "time_per_iteration": 2.6451644897460938 }, { "auxiliary_loss_clip": 0.011719, "auxiliary_loss_mlp": 0.01023308, "balance_loss_clip": 1.0489198, "balance_loss_mlp": 1.015571, "epoch": 0.6971682799254494, "flos": 22419179367360.0, "grad_norm": 2.1230734131372895, "language_loss": 0.7773447, "learning_rate": 8.87159925455165e-07, "loss": 0.7992968, "num_input_tokens_seen": 124777240, "step": 5798, "time_per_iteration": 2.4522085189819336 }, { "auxiliary_loss_clip": 0.01126085, "auxiliary_loss_mlp": 0.01031076, "balance_loss_clip": 1.04486978, "balance_loss_mlp": 1.02369058, "epoch": 0.6972885228160886, "flos": 20005837616640.0, "grad_norm": 1.9047800930449998, "language_loss": 0.73169971, "learning_rate": 8.865127614325738e-07, "loss": 0.75327134, "num_input_tokens_seen": 124795670, "step": 5799, "time_per_iteration": 2.557332992553711 }, { "auxiliary_loss_clip": 0.01137845, "auxiliary_loss_mlp": 0.01021379, "balance_loss_clip": 1.04466975, "balance_loss_mlp": 1.01374936, "epoch": 0.6974087657067276, "flos": 37853442848160.0, "grad_norm": 2.0498629557270944, "language_loss": 0.66297317, "learning_rate": 8.85865766325113e-07, "loss": 0.68456542, "num_input_tokens_seen": 124819600, "step": 5800, "time_per_iteration": 2.6688125133514404 }, { "auxiliary_loss_clip": 0.01139228, "auxiliary_loss_mlp": 0.01026023, "balance_loss_clip": 1.0455308, "balance_loss_mlp": 1.01828039, "epoch": 0.6975290085973667, "flos": 29489271118080.0, "grad_norm": 2.291593712357621, "language_loss": 0.7183643, "learning_rate": 8.852189402309287e-07, "loss": 0.74001682, "num_input_tokens_seen": 124838785, "step": 5801, "time_per_iteration": 2.622431755065918 }, { "auxiliary_loss_clip": 0.01160134, "auxiliary_loss_mlp": 0.01030239, "balance_loss_clip": 1.05071175, "balance_loss_mlp": 1.02325368, "epoch": 0.6976492514880057, "flos": 12895633009920.0, "grad_norm": 2.255982486557185, "language_loss": 0.74291337, "learning_rate": 8.845722832481441e-07, "loss": 0.76481712, "num_input_tokens_seen": 124854215, "step": 5802, "time_per_iteration": 2.4467484951019287 }, { "auxiliary_loss_clip": 0.01158402, "auxiliary_loss_mlp": 0.01030533, "balance_loss_clip": 1.04832077, "balance_loss_mlp": 1.02319002, "epoch": 0.6977694943786449, "flos": 24352933165440.0, "grad_norm": 1.9943157079905098, "language_loss": 0.77730381, "learning_rate": 8.83925795474858e-07, "loss": 0.7991932, "num_input_tokens_seen": 124874340, "step": 5803, "time_per_iteration": 2.5509166717529297 }, { "auxiliary_loss_clip": 0.01126399, "auxiliary_loss_mlp": 0.01032084, "balance_loss_clip": 1.04671168, "balance_loss_mlp": 1.0239954, "epoch": 0.6978897372692839, "flos": 29898474605760.0, "grad_norm": 2.315186854133191, "language_loss": 0.5946368, "learning_rate": 8.832794770091414e-07, "loss": 0.61622161, "num_input_tokens_seen": 124895175, "step": 5804, "time_per_iteration": 2.6004393100738525 }, { "auxiliary_loss_clip": 0.01148174, "auxiliary_loss_mlp": 0.0103215, "balance_loss_clip": 1.04589605, "balance_loss_mlp": 1.02484512, "epoch": 0.698009980159923, "flos": 21761582985600.0, "grad_norm": 2.7137721808356035, "language_loss": 0.82313073, "learning_rate": 8.826333279490401e-07, "loss": 0.84493399, "num_input_tokens_seen": 124915810, "step": 5805, "time_per_iteration": 2.574049949645996 }, { "auxiliary_loss_clip": 0.01147317, "auxiliary_loss_mlp": 0.01024872, "balance_loss_clip": 1.04859519, "balance_loss_mlp": 1.01809525, "epoch": 0.6981302230505622, "flos": 19857166955520.0, "grad_norm": 2.087956924560068, "language_loss": 0.6803652, "learning_rate": 8.819873483925748e-07, "loss": 0.70208716, "num_input_tokens_seen": 124932930, "step": 5806, "time_per_iteration": 2.5215702056884766 }, { "auxiliary_loss_clip": 0.0113532, "auxiliary_loss_mlp": 0.00761699, "balance_loss_clip": 1.04770458, "balance_loss_mlp": 1.00053072, "epoch": 0.6982504659412012, "flos": 22198651645920.0, "grad_norm": 2.1691722731977086, "language_loss": 0.74689764, "learning_rate": 8.81341538437739e-07, "loss": 0.76586783, "num_input_tokens_seen": 124951220, "step": 5807, "time_per_iteration": 3.363884925842285 }, { "auxiliary_loss_clip": 0.01145579, "auxiliary_loss_mlp": 0.01029612, "balance_loss_clip": 1.04337037, "balance_loss_mlp": 1.02237618, "epoch": 0.6983707088318403, "flos": 35588484422880.0, "grad_norm": 3.449317027214088, "language_loss": 0.68211961, "learning_rate": 8.80695898182503e-07, "loss": 0.70387149, "num_input_tokens_seen": 124972200, "step": 5808, "time_per_iteration": 2.6506636142730713 }, { "auxiliary_loss_clip": 0.01065902, "auxiliary_loss_mlp": 0.01001128, "balance_loss_clip": 1.02608371, "balance_loss_mlp": 0.99995416, "epoch": 0.6984909517224794, "flos": 65440064255520.0, "grad_norm": 0.8667130105701316, "language_loss": 0.65117383, "learning_rate": 8.800504277248093e-07, "loss": 0.67184412, "num_input_tokens_seen": 125036950, "step": 5809, "time_per_iteration": 3.1338226795196533 }, { "auxiliary_loss_clip": 0.01131994, "auxiliary_loss_mlp": 0.00762286, "balance_loss_clip": 1.05248547, "balance_loss_mlp": 1.00055432, "epoch": 0.6986111946131185, "flos": 18546930732480.0, "grad_norm": 1.8595770083567336, "language_loss": 0.74927175, "learning_rate": 8.794051271625753e-07, "loss": 0.76821452, "num_input_tokens_seen": 125054585, "step": 5810, "time_per_iteration": 3.365351915359497 }, { "auxiliary_loss_clip": 0.01143981, "auxiliary_loss_mlp": 0.01025684, "balance_loss_clip": 1.0462935, "balance_loss_mlp": 1.01895499, "epoch": 0.6987314375037575, "flos": 23039177080320.0, "grad_norm": 1.7370142822178662, "language_loss": 0.82945246, "learning_rate": 8.787599965936925e-07, "loss": 0.85114908, "num_input_tokens_seen": 125075515, "step": 5811, "time_per_iteration": 3.247279405593872 }, { "auxiliary_loss_clip": 0.01124735, "auxiliary_loss_mlp": 0.01026187, "balance_loss_clip": 1.04583502, "balance_loss_mlp": 1.0197649, "epoch": 0.6988516803943967, "flos": 38400398241120.0, "grad_norm": 1.9185321440306178, "language_loss": 0.71742499, "learning_rate": 8.781150361160261e-07, "loss": 0.73893428, "num_input_tokens_seen": 125097425, "step": 5812, "time_per_iteration": 2.6987924575805664 }, { "auxiliary_loss_clip": 0.01133493, "auxiliary_loss_mlp": 0.01027557, "balance_loss_clip": 1.04926395, "balance_loss_mlp": 1.02041674, "epoch": 0.6989719232850358, "flos": 24096997709760.0, "grad_norm": 1.6290834403364498, "language_loss": 0.73461628, "learning_rate": 8.774702458274181e-07, "loss": 0.75622678, "num_input_tokens_seen": 125117830, "step": 5813, "time_per_iteration": 2.5561890602111816 }, { "auxiliary_loss_clip": 0.01158657, "auxiliary_loss_mlp": 0.01024749, "balance_loss_clip": 1.0486443, "balance_loss_mlp": 1.0175662, "epoch": 0.6990921661756748, "flos": 14866841808960.0, "grad_norm": 2.2759314517733427, "language_loss": 0.70260429, "learning_rate": 8.768256258256799e-07, "loss": 0.72443837, "num_input_tokens_seen": 125134455, "step": 5814, "time_per_iteration": 2.45926570892334 }, { "auxiliary_loss_clip": 0.01160623, "auxiliary_loss_mlp": 0.01024256, "balance_loss_clip": 1.04843056, "balance_loss_mlp": 1.0166688, "epoch": 0.699212409066314, "flos": 20193723209760.0, "grad_norm": 2.086240264553093, "language_loss": 0.73824084, "learning_rate": 8.76181176208602e-07, "loss": 0.76008964, "num_input_tokens_seen": 125152555, "step": 5815, "time_per_iteration": 2.479509115219116 }, { "auxiliary_loss_clip": 0.01105315, "auxiliary_loss_mlp": 0.01026804, "balance_loss_clip": 1.0406822, "balance_loss_mlp": 1.01905537, "epoch": 0.699332651956953, "flos": 19427892275520.0, "grad_norm": 1.7158616936465083, "language_loss": 0.73566806, "learning_rate": 8.755368970739461e-07, "loss": 0.75698924, "num_input_tokens_seen": 125171915, "step": 5816, "time_per_iteration": 2.5753066539764404 }, { "auxiliary_loss_clip": 0.01135613, "auxiliary_loss_mlp": 0.01026072, "balance_loss_clip": 1.04385948, "balance_loss_mlp": 1.01866031, "epoch": 0.6994528948475921, "flos": 16143717564480.0, "grad_norm": 2.080165564536521, "language_loss": 0.6122638, "learning_rate": 8.748927885194479e-07, "loss": 0.63388073, "num_input_tokens_seen": 125190220, "step": 5817, "time_per_iteration": 2.5266950130462646 }, { "auxiliary_loss_clip": 0.01030438, "auxiliary_loss_mlp": 0.01003976, "balance_loss_clip": 1.01009405, "balance_loss_mlp": 1.00283742, "epoch": 0.6995731377382313, "flos": 64952430903840.0, "grad_norm": 0.7971355697053625, "language_loss": 0.57448661, "learning_rate": 8.742488506428209e-07, "loss": 0.59483075, "num_input_tokens_seen": 125249310, "step": 5818, "time_per_iteration": 3.1048974990844727 }, { "auxiliary_loss_clip": 0.01145868, "auxiliary_loss_mlp": 0.00761952, "balance_loss_clip": 1.04595304, "balance_loss_mlp": 1.00056374, "epoch": 0.6996933806288703, "flos": 24900139977120.0, "grad_norm": 1.7496742541008241, "language_loss": 0.77922153, "learning_rate": 8.736050835417466e-07, "loss": 0.79829967, "num_input_tokens_seen": 125269350, "step": 5819, "time_per_iteration": 3.267319679260254 }, { "auxiliary_loss_clip": 0.01163973, "auxiliary_loss_mlp": 0.01023679, "balance_loss_clip": 1.0492295, "balance_loss_mlp": 1.01656199, "epoch": 0.6998136235195094, "flos": 20777810351040.0, "grad_norm": 1.973311957054365, "language_loss": 0.61461484, "learning_rate": 8.729614873138862e-07, "loss": 0.63649136, "num_input_tokens_seen": 125286985, "step": 5820, "time_per_iteration": 2.4857869148254395 }, { "auxiliary_loss_clip": 0.01124918, "auxiliary_loss_mlp": 0.01021396, "balance_loss_clip": 1.04718935, "balance_loss_mlp": 1.0141958, "epoch": 0.6999338664101485, "flos": 23733474206880.0, "grad_norm": 2.0146661770340075, "language_loss": 0.77516258, "learning_rate": 8.723180620568716e-07, "loss": 0.79662561, "num_input_tokens_seen": 125306240, "step": 5821, "time_per_iteration": 2.6050593852996826 }, { "auxiliary_loss_clip": 0.01143678, "auxiliary_loss_mlp": 0.01027538, "balance_loss_clip": 1.04617929, "balance_loss_mlp": 1.0205164, "epoch": 0.7000541093007876, "flos": 19864601766240.0, "grad_norm": 1.9562730830285662, "language_loss": 0.84989071, "learning_rate": 8.716748078683116e-07, "loss": 0.87160289, "num_input_tokens_seen": 125323015, "step": 5822, "time_per_iteration": 2.5046584606170654 }, { "auxiliary_loss_clip": 0.01079414, "auxiliary_loss_mlp": 0.01029779, "balance_loss_clip": 1.04040468, "balance_loss_mlp": 1.0219233, "epoch": 0.7001743521914267, "flos": 29679060310080.0, "grad_norm": 4.23310768391251, "language_loss": 0.68411541, "learning_rate": 8.710317248457855e-07, "loss": 0.70520735, "num_input_tokens_seen": 125342630, "step": 5823, "time_per_iteration": 2.691835641860962 }, { "auxiliary_loss_clip": 0.01142385, "auxiliary_loss_mlp": 0.01026378, "balance_loss_clip": 1.04960442, "balance_loss_mlp": 1.01915097, "epoch": 0.7002945950820658, "flos": 27489766142880.0, "grad_norm": 1.7582485434584245, "language_loss": 0.7255218, "learning_rate": 8.703888130868482e-07, "loss": 0.74720943, "num_input_tokens_seen": 125364480, "step": 5824, "time_per_iteration": 2.5782105922698975 }, { "auxiliary_loss_clip": 0.01130589, "auxiliary_loss_mlp": 0.01028346, "balance_loss_clip": 1.04623961, "balance_loss_mlp": 1.02143753, "epoch": 0.7004148379727049, "flos": 22158466956000.0, "grad_norm": 2.1233279649154384, "language_loss": 0.81936717, "learning_rate": 8.697460726890307e-07, "loss": 0.84095657, "num_input_tokens_seen": 125381625, "step": 5825, "time_per_iteration": 2.568769931793213 }, { "auxiliary_loss_clip": 0.01127546, "auxiliary_loss_mlp": 0.00762407, "balance_loss_clip": 1.04213238, "balance_loss_mlp": 1.00059199, "epoch": 0.7005350808633439, "flos": 19423761825120.0, "grad_norm": 1.91230375385316, "language_loss": 0.90486228, "learning_rate": 8.691035037498354e-07, "loss": 0.92376179, "num_input_tokens_seen": 125397615, "step": 5826, "time_per_iteration": 2.543375015258789 }, { "auxiliary_loss_clip": 0.01139583, "auxiliary_loss_mlp": 0.01024542, "balance_loss_clip": 1.0431478, "balance_loss_mlp": 1.01748753, "epoch": 0.7006553237539831, "flos": 23476712661120.0, "grad_norm": 1.5733439174169652, "language_loss": 0.72314566, "learning_rate": 8.684611063667391e-07, "loss": 0.74478698, "num_input_tokens_seen": 125418080, "step": 5827, "time_per_iteration": 2.562854051589966 }, { "auxiliary_loss_clip": 0.01157836, "auxiliary_loss_mlp": 0.01025048, "balance_loss_clip": 1.04707503, "balance_loss_mlp": 1.01759195, "epoch": 0.7007755666446221, "flos": 31212877196160.0, "grad_norm": 1.9682395767485759, "language_loss": 0.76489693, "learning_rate": 8.678188806371935e-07, "loss": 0.78672576, "num_input_tokens_seen": 125440115, "step": 5828, "time_per_iteration": 2.6698451042175293 }, { "auxiliary_loss_clip": 0.0115718, "auxiliary_loss_mlp": 0.01029243, "balance_loss_clip": 1.04629493, "balance_loss_mlp": 1.02316046, "epoch": 0.7008958095352612, "flos": 18149903094240.0, "grad_norm": 1.82233045075553, "language_loss": 0.85477459, "learning_rate": 8.671768266586228e-07, "loss": 0.87663889, "num_input_tokens_seen": 125458240, "step": 5829, "time_per_iteration": 2.4779467582702637 }, { "auxiliary_loss_clip": 0.01125, "auxiliary_loss_mlp": 0.01029122, "balance_loss_clip": 1.04416978, "balance_loss_mlp": 1.02211535, "epoch": 0.7010160524259004, "flos": 27452311141920.0, "grad_norm": 1.675211256640848, "language_loss": 0.78282642, "learning_rate": 8.665349445284275e-07, "loss": 0.80436766, "num_input_tokens_seen": 125477980, "step": 5830, "time_per_iteration": 2.587827444076538 }, { "auxiliary_loss_clip": 0.01130546, "auxiliary_loss_mlp": 0.01031658, "balance_loss_clip": 1.04624128, "balance_loss_mlp": 1.02457976, "epoch": 0.7011362953165394, "flos": 23842067928960.0, "grad_norm": 1.4812359123209151, "language_loss": 0.80827838, "learning_rate": 8.658932343439799e-07, "loss": 0.82990038, "num_input_tokens_seen": 125497765, "step": 5831, "time_per_iteration": 2.5711724758148193 }, { "auxiliary_loss_clip": 0.01171693, "auxiliary_loss_mlp": 0.01033545, "balance_loss_clip": 1.04954803, "balance_loss_mlp": 1.02646685, "epoch": 0.7012565382071785, "flos": 24823434127200.0, "grad_norm": 2.060653791744242, "language_loss": 0.77783501, "learning_rate": 8.65251696202627e-07, "loss": 0.79988736, "num_input_tokens_seen": 125514145, "step": 5832, "time_per_iteration": 3.2596073150634766 }, { "auxiliary_loss_clip": 0.0112842, "auxiliary_loss_mlp": 0.01029183, "balance_loss_clip": 1.04650044, "balance_loss_mlp": 1.02098119, "epoch": 0.7013767810978175, "flos": 21397448894400.0, "grad_norm": 3.5429008965154325, "language_loss": 0.87332749, "learning_rate": 8.646103302016896e-07, "loss": 0.89490354, "num_input_tokens_seen": 125533115, "step": 5833, "time_per_iteration": 2.5371780395507812 }, { "auxiliary_loss_clip": 0.01125356, "auxiliary_loss_mlp": 0.01025296, "balance_loss_clip": 1.04331636, "balance_loss_mlp": 1.01747322, "epoch": 0.7014970239884567, "flos": 16687153095360.0, "grad_norm": 2.1044174638154924, "language_loss": 0.88452601, "learning_rate": 8.639691364384614e-07, "loss": 0.90603244, "num_input_tokens_seen": 125550740, "step": 5834, "time_per_iteration": 2.5660510063171387 }, { "auxiliary_loss_clip": 0.01147109, "auxiliary_loss_mlp": 0.01032421, "balance_loss_clip": 1.04768229, "balance_loss_mlp": 1.02512217, "epoch": 0.7016172668790958, "flos": 12568271497440.0, "grad_norm": 1.9761743576802977, "language_loss": 0.72832942, "learning_rate": 8.633281150102136e-07, "loss": 0.75012469, "num_input_tokens_seen": 125567590, "step": 5835, "time_per_iteration": 2.4810895919799805 }, { "auxiliary_loss_clip": 0.01142216, "auxiliary_loss_mlp": 0.01023743, "balance_loss_clip": 1.04714966, "balance_loss_mlp": 1.01682293, "epoch": 0.7017375097697348, "flos": 17452732610880.0, "grad_norm": 2.412762291319239, "language_loss": 0.68456715, "learning_rate": 8.626872660141855e-07, "loss": 0.70622671, "num_input_tokens_seen": 125585500, "step": 5836, "time_per_iteration": 3.954702615737915 }, { "auxiliary_loss_clip": 0.01117827, "auxiliary_loss_mlp": 0.01025665, "balance_loss_clip": 1.04626918, "balance_loss_mlp": 1.01852441, "epoch": 0.701857752660374, "flos": 18513031510560.0, "grad_norm": 1.6820900422932124, "language_loss": 0.74790096, "learning_rate": 8.620465895475957e-07, "loss": 0.76933587, "num_input_tokens_seen": 125603720, "step": 5837, "time_per_iteration": 2.56473708152771 }, { "auxiliary_loss_clip": 0.01109009, "auxiliary_loss_mlp": 0.01028566, "balance_loss_clip": 1.04362464, "balance_loss_mlp": 1.02208662, "epoch": 0.701977995551013, "flos": 24425975485440.0, "grad_norm": 1.477669470170686, "language_loss": 0.75384986, "learning_rate": 8.614060857076333e-07, "loss": 0.77522564, "num_input_tokens_seen": 125624390, "step": 5838, "time_per_iteration": 2.6192688941955566 }, { "auxiliary_loss_clip": 0.01140735, "auxiliary_loss_mlp": 0.01026075, "balance_loss_clip": 1.04608345, "balance_loss_mlp": 1.01826346, "epoch": 0.7020982384416521, "flos": 23002763671200.0, "grad_norm": 2.649268136657988, "language_loss": 0.74781305, "learning_rate": 8.60765754591462e-07, "loss": 0.76948112, "num_input_tokens_seen": 125644085, "step": 5839, "time_per_iteration": 2.520872116088867 }, { "auxiliary_loss_clip": 0.01169797, "auxiliary_loss_mlp": 0.01022195, "balance_loss_clip": 1.04804921, "balance_loss_mlp": 1.01517344, "epoch": 0.7022184813322913, "flos": 20449083994080.0, "grad_norm": 3.864874055320116, "language_loss": 0.72729391, "learning_rate": 8.601255962962211e-07, "loss": 0.74921381, "num_input_tokens_seen": 125663095, "step": 5840, "time_per_iteration": 2.478295087814331 }, { "auxiliary_loss_clip": 0.01167881, "auxiliary_loss_mlp": 0.010297, "balance_loss_clip": 1.04972541, "balance_loss_mlp": 1.02184391, "epoch": 0.7023387242229303, "flos": 19790517854400.0, "grad_norm": 2.4758773611244744, "language_loss": 0.72093654, "learning_rate": 8.594856109190194e-07, "loss": 0.74291241, "num_input_tokens_seen": 125680125, "step": 5841, "time_per_iteration": 2.4599075317382812 }, { "auxiliary_loss_clip": 0.01173241, "auxiliary_loss_mlp": 0.01030279, "balance_loss_clip": 1.04972243, "balance_loss_mlp": 1.02282202, "epoch": 0.7024589671135694, "flos": 33259283332800.0, "grad_norm": 1.627108490981872, "language_loss": 0.69209909, "learning_rate": 8.588457985569446e-07, "loss": 0.71413434, "num_input_tokens_seen": 125703035, "step": 5842, "time_per_iteration": 2.6090219020843506 }, { "auxiliary_loss_clip": 0.01174575, "auxiliary_loss_mlp": 0.01026099, "balance_loss_clip": 1.05087376, "balance_loss_mlp": 1.01816308, "epoch": 0.7025792100042085, "flos": 19098986333760.0, "grad_norm": 2.262142419526225, "language_loss": 0.71940517, "learning_rate": 8.582061593070542e-07, "loss": 0.74141192, "num_input_tokens_seen": 125723765, "step": 5843, "time_per_iteration": 2.4660797119140625 }, { "auxiliary_loss_clip": 0.01172023, "auxiliary_loss_mlp": 0.00762188, "balance_loss_clip": 1.04876709, "balance_loss_mlp": 1.00060332, "epoch": 0.7026994528948476, "flos": 18952614358080.0, "grad_norm": 2.0329591060372194, "language_loss": 0.76897693, "learning_rate": 8.57566693266383e-07, "loss": 0.78831905, "num_input_tokens_seen": 125741455, "step": 5844, "time_per_iteration": 2.4941887855529785 }, { "auxiliary_loss_clip": 0.01148437, "auxiliary_loss_mlp": 0.00762097, "balance_loss_clip": 1.04566753, "balance_loss_mlp": 1.00060809, "epoch": 0.7028196957854866, "flos": 19536665582400.0, "grad_norm": 2.019294209584735, "language_loss": 0.69529724, "learning_rate": 8.569274005319354e-07, "loss": 0.71440268, "num_input_tokens_seen": 125759855, "step": 5845, "time_per_iteration": 3.266803026199341 }, { "auxiliary_loss_clip": 0.01153885, "auxiliary_loss_mlp": 0.01026537, "balance_loss_clip": 1.04721797, "balance_loss_mlp": 1.01928878, "epoch": 0.7029399386761258, "flos": 20845321459200.0, "grad_norm": 1.6014746860500697, "language_loss": 0.79634619, "learning_rate": 8.562882812006913e-07, "loss": 0.8181504, "num_input_tokens_seen": 125777345, "step": 5846, "time_per_iteration": 2.494577646255493 }, { "auxiliary_loss_clip": 0.01170101, "auxiliary_loss_mlp": 0.01027603, "balance_loss_clip": 1.04770803, "balance_loss_mlp": 1.0205102, "epoch": 0.7030601815667649, "flos": 22055009359200.0, "grad_norm": 1.6726138324630033, "language_loss": 0.77369851, "learning_rate": 8.556493353696066e-07, "loss": 0.79567552, "num_input_tokens_seen": 125796345, "step": 5847, "time_per_iteration": 2.4621710777282715 }, { "auxiliary_loss_clip": 0.01162676, "auxiliary_loss_mlp": 0.00762996, "balance_loss_clip": 1.04989541, "balance_loss_mlp": 1.00062025, "epoch": 0.7031804244574039, "flos": 27198746205600.0, "grad_norm": 2.170311184904023, "language_loss": 0.6807034, "learning_rate": 8.550105631356077e-07, "loss": 0.69996011, "num_input_tokens_seen": 125816070, "step": 5848, "time_per_iteration": 2.5685856342315674 }, { "auxiliary_loss_clip": 0.01124744, "auxiliary_loss_mlp": 0.01027418, "balance_loss_clip": 1.0427618, "balance_loss_mlp": 1.01994419, "epoch": 0.7033006673480431, "flos": 22379856684480.0, "grad_norm": 1.968796001111601, "language_loss": 0.77261186, "learning_rate": 8.543719645955961e-07, "loss": 0.79413342, "num_input_tokens_seen": 125834400, "step": 5849, "time_per_iteration": 2.5650577545166016 }, { "auxiliary_loss_clip": 0.01144638, "auxiliary_loss_mlp": 0.01024012, "balance_loss_clip": 1.0458039, "balance_loss_mlp": 1.01645398, "epoch": 0.7034209102386821, "flos": 24715989747840.0, "grad_norm": 1.6553847798571368, "language_loss": 0.74463767, "learning_rate": 8.537335398464467e-07, "loss": 0.76632416, "num_input_tokens_seen": 125854720, "step": 5850, "time_per_iteration": 2.5486130714416504 }, { "auxiliary_loss_clip": 0.01140489, "auxiliary_loss_mlp": 0.01029489, "balance_loss_clip": 1.04227066, "balance_loss_mlp": 1.02200294, "epoch": 0.7035411531293212, "flos": 22556177016480.0, "grad_norm": 2.8817220981867124, "language_loss": 0.85624242, "learning_rate": 8.53095288985007e-07, "loss": 0.8779422, "num_input_tokens_seen": 125868455, "step": 5851, "time_per_iteration": 2.5059916973114014 }, { "auxiliary_loss_clip": 0.0117152, "auxiliary_loss_mlp": 0.01023034, "balance_loss_clip": 1.05088127, "balance_loss_mlp": 1.01651907, "epoch": 0.7036613960199604, "flos": 22674971155200.0, "grad_norm": 1.6291801223554614, "language_loss": 0.82120115, "learning_rate": 8.524572121081009e-07, "loss": 0.84314668, "num_input_tokens_seen": 125888555, "step": 5852, "time_per_iteration": 2.4667365550994873 }, { "auxiliary_loss_clip": 0.01163754, "auxiliary_loss_mlp": 0.01029667, "balance_loss_clip": 1.04874492, "balance_loss_mlp": 1.02207327, "epoch": 0.7037816389105994, "flos": 22492149853440.0, "grad_norm": 2.020384668743264, "language_loss": 0.62837225, "learning_rate": 8.518193093125232e-07, "loss": 0.65030646, "num_input_tokens_seen": 125907610, "step": 5853, "time_per_iteration": 2.5168511867523193 }, { "auxiliary_loss_clip": 0.01150221, "auxiliary_loss_mlp": 0.0102489, "balance_loss_clip": 1.04782844, "balance_loss_mlp": 1.01821768, "epoch": 0.7039018818012385, "flos": 27087494628480.0, "grad_norm": 1.678512072384883, "language_loss": 0.80751741, "learning_rate": 8.511815806950436e-07, "loss": 0.82926846, "num_input_tokens_seen": 125928640, "step": 5854, "time_per_iteration": 2.561891794204712 }, { "auxiliary_loss_clip": 0.01153645, "auxiliary_loss_mlp": 0.01024159, "balance_loss_clip": 1.04517722, "balance_loss_mlp": 1.01709557, "epoch": 0.7040221246918776, "flos": 17749822514400.0, "grad_norm": 1.6904796573190273, "language_loss": 0.77584988, "learning_rate": 8.505440263524044e-07, "loss": 0.79762793, "num_input_tokens_seen": 125947485, "step": 5855, "time_per_iteration": 2.470534086227417 }, { "auxiliary_loss_clip": 0.0116301, "auxiliary_loss_mlp": 0.01028538, "balance_loss_clip": 1.04864621, "balance_loss_mlp": 1.02073026, "epoch": 0.7041423675825167, "flos": 16279853206560.0, "grad_norm": 2.314672504430982, "language_loss": 0.87341481, "learning_rate": 8.49906646381322e-07, "loss": 0.89533031, "num_input_tokens_seen": 125960320, "step": 5856, "time_per_iteration": 2.4330358505249023 }, { "auxiliary_loss_clip": 0.01127978, "auxiliary_loss_mlp": 0.01023709, "balance_loss_clip": 1.04518676, "balance_loss_mlp": 1.01695871, "epoch": 0.7042626104731557, "flos": 25483185526560.0, "grad_norm": 1.9516576497321356, "language_loss": 0.72171128, "learning_rate": 8.492694408784884e-07, "loss": 0.74322814, "num_input_tokens_seen": 125980575, "step": 5857, "time_per_iteration": 2.5885956287384033 }, { "auxiliary_loss_clip": 0.01162784, "auxiliary_loss_mlp": 0.01023176, "balance_loss_clip": 1.04895031, "balance_loss_mlp": 1.01620853, "epoch": 0.7043828533637949, "flos": 17857626063360.0, "grad_norm": 2.4657321384560595, "language_loss": 0.61812139, "learning_rate": 8.486324099405642e-07, "loss": 0.63998097, "num_input_tokens_seen": 125997420, "step": 5858, "time_per_iteration": 2.4611852169036865 }, { "auxiliary_loss_clip": 0.01156426, "auxiliary_loss_mlp": 0.01025712, "balance_loss_clip": 1.04740262, "balance_loss_mlp": 1.01895559, "epoch": 0.704503096254434, "flos": 29494263575520.0, "grad_norm": 1.6971368433528446, "language_loss": 0.7495299, "learning_rate": 8.479955536641887e-07, "loss": 0.77135122, "num_input_tokens_seen": 126018915, "step": 5859, "time_per_iteration": 3.349493980407715 }, { "auxiliary_loss_clip": 0.01132641, "auxiliary_loss_mlp": 0.010289, "balance_loss_clip": 1.04099393, "balance_loss_mlp": 1.02182221, "epoch": 0.704623339145073, "flos": 30920743833120.0, "grad_norm": 1.9136280201820997, "language_loss": 0.66229576, "learning_rate": 8.473588721459716e-07, "loss": 0.68391114, "num_input_tokens_seen": 126038825, "step": 5860, "time_per_iteration": 2.596165180206299 }, { "auxiliary_loss_clip": 0.01161533, "auxiliary_loss_mlp": 0.01030916, "balance_loss_clip": 1.05090535, "balance_loss_mlp": 1.02286363, "epoch": 0.7047435820357122, "flos": 23914751079360.0, "grad_norm": 1.8467827731375914, "language_loss": 0.70175743, "learning_rate": 8.467223654824967e-07, "loss": 0.72368193, "num_input_tokens_seen": 126058280, "step": 5861, "time_per_iteration": 2.524036407470703 }, { "auxiliary_loss_clip": 0.01150529, "auxiliary_loss_mlp": 0.01025894, "balance_loss_clip": 1.04583311, "balance_loss_mlp": 1.0182054, "epoch": 0.7048638249263512, "flos": 46494016312800.0, "grad_norm": 2.750412927037787, "language_loss": 0.6255579, "learning_rate": 8.460860337703233e-07, "loss": 0.64732218, "num_input_tokens_seen": 126078885, "step": 5862, "time_per_iteration": 4.225358247756958 }, { "auxiliary_loss_clip": 0.01117975, "auxiliary_loss_mlp": 0.01024457, "balance_loss_clip": 1.0421958, "balance_loss_mlp": 1.01613629, "epoch": 0.7049840678169903, "flos": 21689223087840.0, "grad_norm": 1.7343343586127609, "language_loss": 0.70505345, "learning_rate": 8.454498771059797e-07, "loss": 0.72647774, "num_input_tokens_seen": 126098260, "step": 5863, "time_per_iteration": 2.5507211685180664 }, { "auxiliary_loss_clip": 0.01107187, "auxiliary_loss_mlp": 0.01023524, "balance_loss_clip": 1.0416261, "balance_loss_mlp": 1.01642776, "epoch": 0.7051043107076294, "flos": 18405084293760.0, "grad_norm": 2.976525774181675, "language_loss": 0.83059984, "learning_rate": 8.448138955859725e-07, "loss": 0.85190696, "num_input_tokens_seen": 126114845, "step": 5864, "time_per_iteration": 2.5584003925323486 }, { "auxiliary_loss_clip": 0.01146731, "auxiliary_loss_mlp": 0.01027288, "balance_loss_clip": 1.04796743, "balance_loss_mlp": 1.02010894, "epoch": 0.7052245535982685, "flos": 19319047134720.0, "grad_norm": 1.866288411407628, "language_loss": 0.89738202, "learning_rate": 8.44178089306778e-07, "loss": 0.91912222, "num_input_tokens_seen": 126132780, "step": 5865, "time_per_iteration": 2.515547275543213 }, { "auxiliary_loss_clip": 0.01171625, "auxiliary_loss_mlp": 0.01024435, "balance_loss_clip": 1.0508436, "balance_loss_mlp": 1.01714802, "epoch": 0.7053447964889076, "flos": 19062141921120.0, "grad_norm": 1.8430751852456388, "language_loss": 0.76798749, "learning_rate": 8.4354245836485e-07, "loss": 0.78994799, "num_input_tokens_seen": 126151225, "step": 5866, "time_per_iteration": 2.469306230545044 }, { "auxiliary_loss_clip": 0.01132647, "auxiliary_loss_mlp": 0.01022577, "balance_loss_clip": 1.04618943, "balance_loss_mlp": 1.01504278, "epoch": 0.7054650393795466, "flos": 27379232904960.0, "grad_norm": 1.664639923105821, "language_loss": 0.7283895, "learning_rate": 8.429070028566108e-07, "loss": 0.74994171, "num_input_tokens_seen": 126172535, "step": 5867, "time_per_iteration": 2.6258721351623535 }, { "auxiliary_loss_clip": 0.01158369, "auxiliary_loss_mlp": 0.01024449, "balance_loss_clip": 1.04914069, "balance_loss_mlp": 1.01707625, "epoch": 0.7055852822701858, "flos": 16102203947040.0, "grad_norm": 3.0751996226972955, "language_loss": 0.74677229, "learning_rate": 8.422717228784586e-07, "loss": 0.76860052, "num_input_tokens_seen": 126189410, "step": 5868, "time_per_iteration": 2.486091375350952 }, { "auxiliary_loss_clip": 0.01118228, "auxiliary_loss_mlp": 0.01033826, "balance_loss_clip": 1.04896617, "balance_loss_mlp": 1.02658129, "epoch": 0.7057055251608249, "flos": 11692302411840.0, "grad_norm": 1.816126086871885, "language_loss": 0.69511455, "learning_rate": 8.416366185267663e-07, "loss": 0.71663511, "num_input_tokens_seen": 126206910, "step": 5869, "time_per_iteration": 2.527883529663086 }, { "auxiliary_loss_clip": 0.01155577, "auxiliary_loss_mlp": 0.01021937, "balance_loss_clip": 1.04526639, "balance_loss_mlp": 1.01512408, "epoch": 0.7058257680514639, "flos": 22711564149120.0, "grad_norm": 1.8672783714002346, "language_loss": 0.77802807, "learning_rate": 8.410016898978778e-07, "loss": 0.7998032, "num_input_tokens_seen": 126224385, "step": 5870, "time_per_iteration": 2.524780035018921 }, { "auxiliary_loss_clip": 0.01112606, "auxiliary_loss_mlp": 0.01024274, "balance_loss_clip": 1.04623413, "balance_loss_mlp": 1.01738381, "epoch": 0.7059460109421031, "flos": 17529546211680.0, "grad_norm": 1.631805042283168, "language_loss": 0.78950447, "learning_rate": 8.403669370881115e-07, "loss": 0.81087327, "num_input_tokens_seen": 126243120, "step": 5871, "time_per_iteration": 3.334017753601074 }, { "auxiliary_loss_clip": 0.01171412, "auxiliary_loss_mlp": 0.01025354, "balance_loss_clip": 1.04931283, "balance_loss_mlp": 1.0183655, "epoch": 0.7060662538327421, "flos": 23544690689760.0, "grad_norm": 1.6846580361789927, "language_loss": 0.78267133, "learning_rate": 8.397323601937587e-07, "loss": 0.80463898, "num_input_tokens_seen": 126263020, "step": 5872, "time_per_iteration": 2.4875941276550293 }, { "auxiliary_loss_clip": 0.01123961, "auxiliary_loss_mlp": 0.01023423, "balance_loss_clip": 1.04317355, "balance_loss_mlp": 1.01654434, "epoch": 0.7061864967233812, "flos": 30260741015040.0, "grad_norm": 13.643378480592677, "language_loss": 0.77177161, "learning_rate": 8.390979593110838e-07, "loss": 0.79324543, "num_input_tokens_seen": 126285150, "step": 5873, "time_per_iteration": 2.6675357818603516 }, { "auxiliary_loss_clip": 0.0115035, "auxiliary_loss_mlp": 0.01022813, "balance_loss_clip": 1.04865956, "balance_loss_mlp": 1.01547313, "epoch": 0.7063067396140204, "flos": 20701463670720.0, "grad_norm": 1.923821659886185, "language_loss": 0.81506622, "learning_rate": 8.384637345363262e-07, "loss": 0.83679783, "num_input_tokens_seen": 126304340, "step": 5874, "time_per_iteration": 2.5298261642456055 }, { "auxiliary_loss_clip": 0.01136146, "auxiliary_loss_mlp": 0.01025256, "balance_loss_clip": 1.04187906, "balance_loss_mlp": 1.01798415, "epoch": 0.7064269825046594, "flos": 32266172288640.0, "grad_norm": 1.7475866817905656, "language_loss": 0.76558673, "learning_rate": 8.378296859656964e-07, "loss": 0.78720081, "num_input_tokens_seen": 126325495, "step": 5875, "time_per_iteration": 2.631837844848633 }, { "auxiliary_loss_clip": 0.0114572, "auxiliary_loss_mlp": 0.01026947, "balance_loss_clip": 1.04659796, "balance_loss_mlp": 1.02049446, "epoch": 0.7065472253952985, "flos": 30227129128800.0, "grad_norm": 2.254979382915701, "language_loss": 0.6844213, "learning_rate": 8.371958136953792e-07, "loss": 0.70614797, "num_input_tokens_seen": 126345525, "step": 5876, "time_per_iteration": 2.591827392578125 }, { "auxiliary_loss_clip": 0.01134331, "auxiliary_loss_mlp": 0.01026633, "balance_loss_clip": 1.04415762, "balance_loss_mlp": 1.01853824, "epoch": 0.7066674682859376, "flos": 16216723967520.0, "grad_norm": 2.832805243192832, "language_loss": 0.66180336, "learning_rate": 8.365621178215326e-07, "loss": 0.68341297, "num_input_tokens_seen": 126361995, "step": 5877, "time_per_iteration": 2.548044443130493 }, { "auxiliary_loss_clip": 0.01152132, "auxiliary_loss_mlp": 0.01022623, "balance_loss_clip": 1.04591203, "balance_loss_mlp": 1.01617122, "epoch": 0.7067877111765767, "flos": 14830464316800.0, "grad_norm": 2.215686465732597, "language_loss": 0.75150228, "learning_rate": 8.359285984402871e-07, "loss": 0.77324975, "num_input_tokens_seen": 126379260, "step": 5878, "time_per_iteration": 2.481804370880127 }, { "auxiliary_loss_clip": 0.01138513, "auxiliary_loss_mlp": 0.01026203, "balance_loss_clip": 1.04650474, "balance_loss_mlp": 1.01953053, "epoch": 0.7069079540672157, "flos": 25440199313760.0, "grad_norm": 1.967570463767749, "language_loss": 0.73702228, "learning_rate": 8.352952556477489e-07, "loss": 0.75866938, "num_input_tokens_seen": 126397170, "step": 5879, "time_per_iteration": 2.5622496604919434 }, { "auxiliary_loss_clip": 0.01156362, "auxiliary_loss_mlp": 0.0102549, "balance_loss_clip": 1.04814422, "balance_loss_mlp": 1.01832557, "epoch": 0.7070281969578549, "flos": 24607755195360.0, "grad_norm": 2.0869907542251913, "language_loss": 0.76909626, "learning_rate": 8.34662089539993e-07, "loss": 0.79091477, "num_input_tokens_seen": 126416680, "step": 5880, "time_per_iteration": 2.524146556854248 }, { "auxiliary_loss_clip": 0.0116856, "auxiliary_loss_mlp": 0.01028528, "balance_loss_clip": 1.04878187, "balance_loss_mlp": 1.02157497, "epoch": 0.707148439848494, "flos": 26724473963040.0, "grad_norm": 2.201826081137103, "language_loss": 0.79311395, "learning_rate": 8.340291002130722e-07, "loss": 0.81508482, "num_input_tokens_seen": 126435870, "step": 5881, "time_per_iteration": 2.5193381309509277 }, { "auxiliary_loss_clip": 0.01173833, "auxiliary_loss_mlp": 0.01029375, "balance_loss_clip": 1.05062854, "balance_loss_mlp": 1.02167463, "epoch": 0.707268682739133, "flos": 15085753267200.0, "grad_norm": 2.2529137795990266, "language_loss": 0.79537392, "learning_rate": 8.3339628776301e-07, "loss": 0.817406, "num_input_tokens_seen": 126454010, "step": 5882, "time_per_iteration": 2.4837400913238525 }, { "auxiliary_loss_clip": 0.01168813, "auxiliary_loss_mlp": 0.01023676, "balance_loss_clip": 1.04710186, "balance_loss_mlp": 1.01653862, "epoch": 0.7073889256297722, "flos": 34313153096640.0, "grad_norm": 1.875687838360935, "language_loss": 0.57107133, "learning_rate": 8.327636522858033e-07, "loss": 0.59299624, "num_input_tokens_seen": 126473615, "step": 5883, "time_per_iteration": 2.5742344856262207 }, { "auxiliary_loss_clip": 0.01114517, "auxiliary_loss_mlp": 0.01021001, "balance_loss_clip": 1.04680383, "balance_loss_mlp": 1.01433444, "epoch": 0.7075091685204112, "flos": 20083944228000.0, "grad_norm": 2.394506356861124, "language_loss": 0.77084976, "learning_rate": 8.321311938774225e-07, "loss": 0.79220492, "num_input_tokens_seen": 126492705, "step": 5884, "time_per_iteration": 3.379209518432617 }, { "auxiliary_loss_clip": 0.01173968, "auxiliary_loss_mlp": 0.01027878, "balance_loss_clip": 1.05029917, "balance_loss_mlp": 1.02071047, "epoch": 0.7076294114110503, "flos": 20777127928800.0, "grad_norm": 1.9178646847215994, "language_loss": 0.79047096, "learning_rate": 8.314989126338104e-07, "loss": 0.81248939, "num_input_tokens_seen": 126512715, "step": 5885, "time_per_iteration": 2.5209615230560303 }, { "auxiliary_loss_clip": 0.01158312, "auxiliary_loss_mlp": 0.01027408, "balance_loss_clip": 1.04625666, "balance_loss_mlp": 1.02007616, "epoch": 0.7077496543016895, "flos": 17967692380800.0, "grad_norm": 1.8535589026255301, "language_loss": 0.84028423, "learning_rate": 8.308668086508847e-07, "loss": 0.86214149, "num_input_tokens_seen": 126530795, "step": 5886, "time_per_iteration": 2.4682536125183105 }, { "auxiliary_loss_clip": 0.01132788, "auxiliary_loss_mlp": 0.01028333, "balance_loss_clip": 1.04264557, "balance_loss_mlp": 1.02097178, "epoch": 0.7078698971923285, "flos": 45478104387360.0, "grad_norm": 1.7009535016050905, "language_loss": 0.73913705, "learning_rate": 8.302348820245342e-07, "loss": 0.76074827, "num_input_tokens_seen": 126553360, "step": 5887, "time_per_iteration": 2.796903610229492 }, { "auxiliary_loss_clip": 0.01127958, "auxiliary_loss_mlp": 0.01025443, "balance_loss_clip": 1.04489112, "balance_loss_mlp": 1.01748633, "epoch": 0.7079901400829676, "flos": 26943708673920.0, "grad_norm": 2.190277131293482, "language_loss": 0.70277733, "learning_rate": 8.296031328506232e-07, "loss": 0.72431135, "num_input_tokens_seen": 126573110, "step": 5888, "time_per_iteration": 4.11287784576416 }, { "auxiliary_loss_clip": 0.01145199, "auxiliary_loss_mlp": 0.01024738, "balance_loss_clip": 1.04692268, "balance_loss_mlp": 1.01740384, "epoch": 0.7081103829736067, "flos": 24423209879520.0, "grad_norm": 1.838743622917989, "language_loss": 0.75644958, "learning_rate": 8.289715612249857e-07, "loss": 0.77814889, "num_input_tokens_seen": 126593725, "step": 5889, "time_per_iteration": 2.5660560131073 }, { "auxiliary_loss_clip": 0.0114057, "auxiliary_loss_mlp": 0.0102781, "balance_loss_clip": 1.04613507, "balance_loss_mlp": 1.02055359, "epoch": 0.7082306258642458, "flos": 18543303119520.0, "grad_norm": 2.390097416566255, "language_loss": 0.77389991, "learning_rate": 8.283401672434305e-07, "loss": 0.79558372, "num_input_tokens_seen": 126608950, "step": 5890, "time_per_iteration": 2.483792304992676 }, { "auxiliary_loss_clip": 0.011417, "auxiliary_loss_mlp": 0.01023854, "balance_loss_clip": 1.0484457, "balance_loss_mlp": 1.01707125, "epoch": 0.7083508687548848, "flos": 23477538751200.0, "grad_norm": 1.88751962618035, "language_loss": 0.70394272, "learning_rate": 8.277089510017412e-07, "loss": 0.72559828, "num_input_tokens_seen": 126629755, "step": 5891, "time_per_iteration": 2.576660633087158 }, { "auxiliary_loss_clip": 0.01141176, "auxiliary_loss_mlp": 0.01022151, "balance_loss_clip": 1.04746413, "balance_loss_mlp": 1.01571965, "epoch": 0.708471111645524, "flos": 22419466703040.0, "grad_norm": 1.6638927028546713, "language_loss": 0.82111913, "learning_rate": 8.270779125956719e-07, "loss": 0.8427524, "num_input_tokens_seen": 126650135, "step": 5892, "time_per_iteration": 2.519760847091675 }, { "auxiliary_loss_clip": 0.01109527, "auxiliary_loss_mlp": 0.0102156, "balance_loss_clip": 1.04413676, "balance_loss_mlp": 1.01428235, "epoch": 0.7085913545361631, "flos": 20922889316160.0, "grad_norm": 2.5649267762526238, "language_loss": 0.79741037, "learning_rate": 8.264470521209505e-07, "loss": 0.81872129, "num_input_tokens_seen": 126668500, "step": 5893, "time_per_iteration": 2.580355167388916 }, { "auxiliary_loss_clip": 0.01148021, "auxiliary_loss_mlp": 0.01027492, "balance_loss_clip": 1.04384661, "balance_loss_mlp": 1.02018452, "epoch": 0.7087115974268021, "flos": 15012387694560.0, "grad_norm": 2.0832927316464707, "language_loss": 0.76587558, "learning_rate": 8.258163696732785e-07, "loss": 0.78763074, "num_input_tokens_seen": 126686090, "step": 5894, "time_per_iteration": 2.474241256713867 }, { "auxiliary_loss_clip": 0.0115272, "auxiliary_loss_mlp": 0.01024862, "balance_loss_clip": 1.04708815, "balance_loss_mlp": 1.01781702, "epoch": 0.7088318403174413, "flos": 21539043914400.0, "grad_norm": 4.823215225036558, "language_loss": 0.76721746, "learning_rate": 8.251858653483288e-07, "loss": 0.7889933, "num_input_tokens_seen": 126704255, "step": 5895, "time_per_iteration": 2.4923481941223145 }, { "auxiliary_loss_clip": 0.01162118, "auxiliary_loss_mlp": 0.01029924, "balance_loss_clip": 1.051162, "balance_loss_mlp": 1.02255058, "epoch": 0.7089520832080803, "flos": 15516787878240.0, "grad_norm": 1.965197912296155, "language_loss": 0.85577691, "learning_rate": 8.245555392417501e-07, "loss": 0.87769735, "num_input_tokens_seen": 126718910, "step": 5896, "time_per_iteration": 2.484191417694092 }, { "auxiliary_loss_clip": 0.01099649, "auxiliary_loss_mlp": 0.01023897, "balance_loss_clip": 1.03969169, "balance_loss_mlp": 1.01678276, "epoch": 0.7090723260987194, "flos": 20412670584960.0, "grad_norm": 1.809380759534401, "language_loss": 0.78621215, "learning_rate": 8.239253914491613e-07, "loss": 0.80744761, "num_input_tokens_seen": 126737235, "step": 5897, "time_per_iteration": 3.3105204105377197 }, { "auxiliary_loss_clip": 0.01125635, "auxiliary_loss_mlp": 0.0102484, "balance_loss_clip": 1.04622006, "balance_loss_mlp": 1.01778293, "epoch": 0.7091925689893585, "flos": 25668341430720.0, "grad_norm": 1.8264968549169145, "language_loss": 0.74913418, "learning_rate": 8.232954220661556e-07, "loss": 0.77063894, "num_input_tokens_seen": 126759970, "step": 5898, "time_per_iteration": 2.6506729125976562 }, { "auxiliary_loss_clip": 0.01173313, "auxiliary_loss_mlp": 0.01029218, "balance_loss_clip": 1.05253828, "balance_loss_mlp": 1.02210712, "epoch": 0.7093128118799976, "flos": 24206633023680.0, "grad_norm": 2.361229890059596, "language_loss": 0.70557016, "learning_rate": 8.226656311882989e-07, "loss": 0.72759545, "num_input_tokens_seen": 126779280, "step": 5899, "time_per_iteration": 2.4709980487823486 }, { "auxiliary_loss_clip": 0.01156406, "auxiliary_loss_mlp": 0.0102277, "balance_loss_clip": 1.05026841, "balance_loss_mlp": 1.01636553, "epoch": 0.7094330547706367, "flos": 16646788820640.0, "grad_norm": 19.632483044144, "language_loss": 0.76789451, "learning_rate": 8.22036018911129e-07, "loss": 0.7896862, "num_input_tokens_seen": 126797310, "step": 5900, "time_per_iteration": 2.493767261505127 }, { "auxiliary_loss_clip": 0.011748, "auxiliary_loss_mlp": 0.01026841, "balance_loss_clip": 1.04868829, "balance_loss_mlp": 1.01950324, "epoch": 0.7095532976612757, "flos": 16283373068640.0, "grad_norm": 2.1460488368513064, "language_loss": 0.80746514, "learning_rate": 8.214065853301599e-07, "loss": 0.82948154, "num_input_tokens_seen": 126812840, "step": 5901, "time_per_iteration": 2.4188077449798584 }, { "auxiliary_loss_clip": 0.01058092, "auxiliary_loss_mlp": 0.01000624, "balance_loss_clip": 1.00855184, "balance_loss_mlp": 0.99929506, "epoch": 0.7096735405519149, "flos": 70722091761600.0, "grad_norm": 0.8133142007199828, "language_loss": 0.58279574, "learning_rate": 8.207773305408734e-07, "loss": 0.60338295, "num_input_tokens_seen": 126880060, "step": 5902, "time_per_iteration": 3.2510411739349365 }, { "auxiliary_loss_clip": 0.01123384, "auxiliary_loss_mlp": 0.0102622, "balance_loss_clip": 1.04363739, "balance_loss_mlp": 1.01886463, "epoch": 0.709793783442554, "flos": 23621504290560.0, "grad_norm": 2.41844820086047, "language_loss": 0.79705727, "learning_rate": 8.201482546387288e-07, "loss": 0.81855333, "num_input_tokens_seen": 126899535, "step": 5903, "time_per_iteration": 2.6078503131866455 }, { "auxiliary_loss_clip": 0.01158408, "auxiliary_loss_mlp": 0.01024355, "balance_loss_clip": 1.04935133, "balance_loss_mlp": 1.01753592, "epoch": 0.709914026333193, "flos": 25993476091680.0, "grad_norm": 1.6338009688709785, "language_loss": 0.91756582, "learning_rate": 8.195193577191553e-07, "loss": 0.93939346, "num_input_tokens_seen": 126921365, "step": 5904, "time_per_iteration": 2.546799898147583 }, { "auxiliary_loss_clip": 0.01146728, "auxiliary_loss_mlp": 0.00762066, "balance_loss_clip": 1.04600942, "balance_loss_mlp": 1.00060093, "epoch": 0.7100342692238322, "flos": 24861535633440.0, "grad_norm": 1.6875094615173185, "language_loss": 0.84588075, "learning_rate": 8.188906398775579e-07, "loss": 0.86496872, "num_input_tokens_seen": 126941910, "step": 5905, "time_per_iteration": 2.589569330215454 }, { "auxiliary_loss_clip": 0.01172839, "auxiliary_loss_mlp": 0.00762479, "balance_loss_clip": 1.04916215, "balance_loss_mlp": 1.00070453, "epoch": 0.7101545121144712, "flos": 24932207434080.0, "grad_norm": 1.725470761355727, "language_loss": 0.68430936, "learning_rate": 8.18262101209311e-07, "loss": 0.70366251, "num_input_tokens_seen": 126961120, "step": 5906, "time_per_iteration": 2.5126471519470215 }, { "auxiliary_loss_clip": 0.01158904, "auxiliary_loss_mlp": 0.01025038, "balance_loss_clip": 1.04894471, "balance_loss_mlp": 1.0180409, "epoch": 0.7102747550051103, "flos": 23768846024160.0, "grad_norm": 1.907476799893709, "language_loss": 0.69824636, "learning_rate": 8.176337418097626e-07, "loss": 0.72008574, "num_input_tokens_seen": 126981590, "step": 5907, "time_per_iteration": 2.49767804145813 }, { "auxiliary_loss_clip": 0.01157185, "auxiliary_loss_mlp": 0.00762049, "balance_loss_clip": 1.04916668, "balance_loss_mlp": 1.00068426, "epoch": 0.7103949978957494, "flos": 15303910469280.0, "grad_norm": 2.0356332777363035, "language_loss": 0.79412419, "learning_rate": 8.170055617742364e-07, "loss": 0.81331652, "num_input_tokens_seen": 126998870, "step": 5908, "time_per_iteration": 2.499025583267212 }, { "auxiliary_loss_clip": 0.01136411, "auxiliary_loss_mlp": 0.01030999, "balance_loss_clip": 1.04422021, "balance_loss_mlp": 1.02356935, "epoch": 0.7105152407863885, "flos": 22638809164800.0, "grad_norm": 2.1125718259147557, "language_loss": 0.71066046, "learning_rate": 8.163775611980252e-07, "loss": 0.73233449, "num_input_tokens_seen": 127017980, "step": 5909, "time_per_iteration": 2.5298094749450684 }, { "auxiliary_loss_clip": 0.01145237, "auxiliary_loss_mlp": 0.0102368, "balance_loss_clip": 1.04805398, "balance_loss_mlp": 1.01655698, "epoch": 0.7106354836770276, "flos": 17238598108320.0, "grad_norm": 1.6306727300572454, "language_loss": 0.79045987, "learning_rate": 8.157497401763982e-07, "loss": 0.81214905, "num_input_tokens_seen": 127035645, "step": 5910, "time_per_iteration": 3.25136661529541 }, { "auxiliary_loss_clip": 0.01155403, "auxiliary_loss_mlp": 0.01032046, "balance_loss_clip": 1.04656887, "balance_loss_mlp": 1.0249474, "epoch": 0.7107557265676667, "flos": 20193651375840.0, "grad_norm": 1.722792461724068, "language_loss": 0.7785033, "learning_rate": 8.151220988045935e-07, "loss": 0.80037779, "num_input_tokens_seen": 127054900, "step": 5911, "time_per_iteration": 2.5025222301483154 }, { "auxiliary_loss_clip": 0.01155508, "auxiliary_loss_mlp": 0.0102447, "balance_loss_clip": 1.04712152, "balance_loss_mlp": 1.018152, "epoch": 0.7108759694583058, "flos": 21507084208320.0, "grad_norm": 1.6617235631333347, "language_loss": 0.82607901, "learning_rate": 8.144946371778234e-07, "loss": 0.84787875, "num_input_tokens_seen": 127075010, "step": 5912, "time_per_iteration": 2.48732590675354 }, { "auxiliary_loss_clip": 0.01144479, "auxiliary_loss_mlp": 0.00762924, "balance_loss_clip": 1.04876423, "balance_loss_mlp": 1.00065351, "epoch": 0.7109962123489448, "flos": 24061913228160.0, "grad_norm": 14.329507956290554, "language_loss": 0.78367537, "learning_rate": 8.138673553912751e-07, "loss": 0.8027494, "num_input_tokens_seen": 127095570, "step": 5913, "time_per_iteration": 2.547243118286133 }, { "auxiliary_loss_clip": 0.01111977, "auxiliary_loss_mlp": 0.01029685, "balance_loss_clip": 1.04325521, "balance_loss_mlp": 1.02288389, "epoch": 0.711116455239584, "flos": 30480478563360.0, "grad_norm": 2.0713632855576094, "language_loss": 0.56679189, "learning_rate": 8.132402535401059e-07, "loss": 0.5882085, "num_input_tokens_seen": 127116825, "step": 5914, "time_per_iteration": 4.170598268508911 }, { "auxiliary_loss_clip": 0.0115473, "auxiliary_loss_mlp": 0.01023854, "balance_loss_clip": 1.04845333, "balance_loss_mlp": 1.01698184, "epoch": 0.711236698130223, "flos": 25045614028800.0, "grad_norm": 1.7636100495834575, "language_loss": 0.74340641, "learning_rate": 8.126133317194465e-07, "loss": 0.76519227, "num_input_tokens_seen": 127137015, "step": 5915, "time_per_iteration": 2.5167198181152344 }, { "auxiliary_loss_clip": 0.01108768, "auxiliary_loss_mlp": 0.01030825, "balance_loss_clip": 1.04259372, "balance_loss_mlp": 1.02366614, "epoch": 0.7113569410208621, "flos": 24206704857600.0, "grad_norm": 1.7803555149264318, "language_loss": 0.74315351, "learning_rate": 8.11986590024401e-07, "loss": 0.76454949, "num_input_tokens_seen": 127156755, "step": 5916, "time_per_iteration": 2.634162187576294 }, { "auxiliary_loss_clip": 0.01149189, "auxiliary_loss_mlp": 0.01026786, "balance_loss_clip": 1.04847097, "balance_loss_mlp": 1.01910949, "epoch": 0.7114771839115013, "flos": 35439310924320.0, "grad_norm": 1.7318997072997173, "language_loss": 0.68950927, "learning_rate": 8.113600285500442e-07, "loss": 0.71126902, "num_input_tokens_seen": 127176965, "step": 5917, "time_per_iteration": 2.6356263160705566 }, { "auxiliary_loss_clip": 0.01171633, "auxiliary_loss_mlp": 0.01025235, "balance_loss_clip": 1.04870987, "balance_loss_mlp": 1.01808858, "epoch": 0.7115974268021403, "flos": 21099461066880.0, "grad_norm": 1.7025614328704632, "language_loss": 0.74178112, "learning_rate": 8.107336473914268e-07, "loss": 0.76374984, "num_input_tokens_seen": 127195595, "step": 5918, "time_per_iteration": 2.4572596549987793 }, { "auxiliary_loss_clip": 0.01044594, "auxiliary_loss_mlp": 0.01000191, "balance_loss_clip": 1.00910878, "balance_loss_mlp": 0.99878436, "epoch": 0.7117176696927794, "flos": 56752876965120.0, "grad_norm": 0.7797386551330278, "language_loss": 0.55736601, "learning_rate": 8.101074466435694e-07, "loss": 0.57781386, "num_input_tokens_seen": 127255070, "step": 5919, "time_per_iteration": 3.0677926540374756 }, { "auxiliary_loss_clip": 0.01151832, "auxiliary_loss_mlp": 0.01029185, "balance_loss_clip": 1.04615486, "balance_loss_mlp": 1.02172244, "epoch": 0.7118379125834186, "flos": 15925273026720.0, "grad_norm": 1.8146253303864206, "language_loss": 0.67953628, "learning_rate": 8.094814264014662e-07, "loss": 0.70134646, "num_input_tokens_seen": 127273825, "step": 5920, "time_per_iteration": 2.4996461868286133 }, { "auxiliary_loss_clip": 0.01176206, "auxiliary_loss_mlp": 0.01031312, "balance_loss_clip": 1.05067253, "balance_loss_mlp": 1.02377188, "epoch": 0.7119581554740576, "flos": 20193364040160.0, "grad_norm": 1.9706154367140596, "language_loss": 0.81006283, "learning_rate": 8.088555867600844e-07, "loss": 0.832138, "num_input_tokens_seen": 127289990, "step": 5921, "time_per_iteration": 2.464329719543457 }, { "auxiliary_loss_clip": 0.01127349, "auxiliary_loss_mlp": 0.01025206, "balance_loss_clip": 1.04406285, "balance_loss_mlp": 1.01882529, "epoch": 0.7120783983646967, "flos": 34715388694080.0, "grad_norm": 1.8238608337885944, "language_loss": 0.60276562, "learning_rate": 8.08229927814362e-07, "loss": 0.62429118, "num_input_tokens_seen": 127312880, "step": 5922, "time_per_iteration": 2.711820125579834 }, { "auxiliary_loss_clip": 0.01127822, "auxiliary_loss_mlp": 0.01028012, "balance_loss_clip": 1.04382181, "balance_loss_mlp": 1.02116299, "epoch": 0.7121986412553358, "flos": 26359118695200.0, "grad_norm": 1.7501488048228608, "language_loss": 0.65023875, "learning_rate": 8.076044496592134e-07, "loss": 0.67179704, "num_input_tokens_seen": 127334730, "step": 5923, "time_per_iteration": 2.6098861694335938 }, { "auxiliary_loss_clip": 0.01143946, "auxiliary_loss_mlp": 0.01024546, "balance_loss_clip": 1.04853845, "balance_loss_mlp": 1.0169704, "epoch": 0.7123188841459749, "flos": 11145347018880.0, "grad_norm": 2.3135327519654187, "language_loss": 0.78320074, "learning_rate": 8.069791523895204e-07, "loss": 0.80488575, "num_input_tokens_seen": 127351180, "step": 5924, "time_per_iteration": 3.257434844970703 }, { "auxiliary_loss_clip": 0.01120285, "auxiliary_loss_mlp": 0.01027808, "balance_loss_clip": 1.04228663, "balance_loss_mlp": 1.02112365, "epoch": 0.7124391270366139, "flos": 20811673656000.0, "grad_norm": 1.7218638039863812, "language_loss": 0.77337766, "learning_rate": 8.063540361001422e-07, "loss": 0.79485857, "num_input_tokens_seen": 127369750, "step": 5925, "time_per_iteration": 2.569796323776245 }, { "auxiliary_loss_clip": 0.01123943, "auxiliary_loss_mlp": 0.0102678, "balance_loss_clip": 1.04462695, "balance_loss_mlp": 1.01957691, "epoch": 0.7125593699272531, "flos": 17603737874400.0, "grad_norm": 2.741157973126758, "language_loss": 0.79519248, "learning_rate": 8.057291008859069e-07, "loss": 0.81669974, "num_input_tokens_seen": 127387910, "step": 5926, "time_per_iteration": 2.5338950157165527 }, { "auxiliary_loss_clip": 0.01153191, "auxiliary_loss_mlp": 0.01029325, "balance_loss_clip": 1.04643345, "balance_loss_mlp": 1.02246439, "epoch": 0.7126796128178922, "flos": 28654061393760.0, "grad_norm": 1.9307435555607135, "language_loss": 0.68393087, "learning_rate": 8.051043468416187e-07, "loss": 0.70575607, "num_input_tokens_seen": 127409160, "step": 5927, "time_per_iteration": 2.5488297939300537 }, { "auxiliary_loss_clip": 0.01173657, "auxiliary_loss_mlp": 0.01026911, "balance_loss_clip": 1.05208373, "balance_loss_mlp": 1.02040768, "epoch": 0.7127998557085312, "flos": 16034441420160.0, "grad_norm": 2.870572155872737, "language_loss": 0.82062066, "learning_rate": 8.044797740620506e-07, "loss": 0.84262627, "num_input_tokens_seen": 127427765, "step": 5928, "time_per_iteration": 2.450066566467285 }, { "auxiliary_loss_clip": 0.01108287, "auxiliary_loss_mlp": 0.01024957, "balance_loss_clip": 1.043648, "balance_loss_mlp": 1.01816773, "epoch": 0.7129200985991703, "flos": 23403275254560.0, "grad_norm": 2.004014679060387, "language_loss": 0.78684235, "learning_rate": 8.038553826419494e-07, "loss": 0.80817473, "num_input_tokens_seen": 127446475, "step": 5929, "time_per_iteration": 2.5829551219940186 }, { "auxiliary_loss_clip": 0.01171494, "auxiliary_loss_mlp": 0.01022779, "balance_loss_clip": 1.04892445, "balance_loss_mlp": 1.01596642, "epoch": 0.7130403414898094, "flos": 21397448894400.0, "grad_norm": 1.5877263967926911, "language_loss": 0.80715871, "learning_rate": 8.032311726760364e-07, "loss": 0.82910132, "num_input_tokens_seen": 127467695, "step": 5930, "time_per_iteration": 2.491694688796997 }, { "auxiliary_loss_clip": 0.01121267, "auxiliary_loss_mlp": 0.01027807, "balance_loss_clip": 1.04475284, "balance_loss_mlp": 1.0204432, "epoch": 0.7131605843804485, "flos": 74739053803680.0, "grad_norm": 1.7532677890767479, "language_loss": 0.68909836, "learning_rate": 8.026071442590022e-07, "loss": 0.71058911, "num_input_tokens_seen": 127494590, "step": 5931, "time_per_iteration": 2.951150894165039 }, { "auxiliary_loss_clip": 0.01158387, "auxiliary_loss_mlp": 0.01024125, "balance_loss_clip": 1.05061555, "balance_loss_mlp": 1.01715684, "epoch": 0.7132808272710875, "flos": 18368742718560.0, "grad_norm": 1.9268109697036508, "language_loss": 0.80906564, "learning_rate": 8.019832974855134e-07, "loss": 0.83089077, "num_input_tokens_seen": 127512550, "step": 5932, "time_per_iteration": 2.4685757160186768 }, { "auxiliary_loss_clip": 0.01128032, "auxiliary_loss_mlp": 0.01026898, "balance_loss_clip": 1.04591727, "balance_loss_mlp": 1.01952517, "epoch": 0.7134010701617267, "flos": 23253383416800.0, "grad_norm": 2.1920062330778443, "language_loss": 0.82289463, "learning_rate": 8.013596324502052e-07, "loss": 0.84444392, "num_input_tokens_seen": 127531015, "step": 5933, "time_per_iteration": 2.5771896839141846 }, { "auxiliary_loss_clip": 0.01153252, "auxiliary_loss_mlp": 0.0102505, "balance_loss_clip": 1.04941201, "balance_loss_mlp": 1.0182941, "epoch": 0.7135213130523658, "flos": 23653140744000.0, "grad_norm": 1.807612932478177, "language_loss": 0.78723812, "learning_rate": 8.007361492476872e-07, "loss": 0.80902117, "num_input_tokens_seen": 127550340, "step": 5934, "time_per_iteration": 2.5262680053710938 }, { "auxiliary_loss_clip": 0.01135305, "auxiliary_loss_mlp": 0.01025698, "balance_loss_clip": 1.04246938, "balance_loss_mlp": 1.01846826, "epoch": 0.7136415559430048, "flos": 24790648331040.0, "grad_norm": 1.437004508015465, "language_loss": 0.79442322, "learning_rate": 8.001128479725426e-07, "loss": 0.81603324, "num_input_tokens_seen": 127572245, "step": 5935, "time_per_iteration": 2.5987730026245117 }, { "auxiliary_loss_clip": 0.01107824, "auxiliary_loss_mlp": 0.01024026, "balance_loss_clip": 1.04255009, "balance_loss_mlp": 1.01699841, "epoch": 0.713761798833644, "flos": 18296957492160.0, "grad_norm": 1.6851393316810908, "language_loss": 0.80961221, "learning_rate": 7.994897287193248e-07, "loss": 0.83093071, "num_input_tokens_seen": 127591625, "step": 5936, "time_per_iteration": 2.573317050933838 }, { "auxiliary_loss_clip": 0.01159549, "auxiliary_loss_mlp": 0.01027506, "balance_loss_clip": 1.04618442, "balance_loss_mlp": 1.0203979, "epoch": 0.713882041724283, "flos": 15558265578720.0, "grad_norm": 2.4104260838899427, "language_loss": 0.83617055, "learning_rate": 7.988667915825605e-07, "loss": 0.85804105, "num_input_tokens_seen": 127608690, "step": 5937, "time_per_iteration": 3.24662184715271 }, { "auxiliary_loss_clip": 0.0114424, "auxiliary_loss_mlp": 0.01020047, "balance_loss_clip": 1.04739976, "balance_loss_mlp": 1.01279902, "epoch": 0.7140022846149221, "flos": 24061015304160.0, "grad_norm": 2.0788673984841486, "language_loss": 0.75726777, "learning_rate": 7.982440366567491e-07, "loss": 0.77891064, "num_input_tokens_seen": 127627180, "step": 5938, "time_per_iteration": 2.55754017829895 }, { "auxiliary_loss_clip": 0.01149241, "auxiliary_loss_mlp": 0.01023533, "balance_loss_clip": 1.04442787, "balance_loss_mlp": 1.01665497, "epoch": 0.7141225275055613, "flos": 27891714404640.0, "grad_norm": 2.216165341044474, "language_loss": 0.75119847, "learning_rate": 7.97621464036361e-07, "loss": 0.77292621, "num_input_tokens_seen": 127648940, "step": 5939, "time_per_iteration": 2.6175930500030518 }, { "auxiliary_loss_clip": 0.01158743, "auxiliary_loss_mlp": 0.0102843, "balance_loss_clip": 1.04726458, "balance_loss_mlp": 1.02107453, "epoch": 0.7142427703962003, "flos": 19682606554560.0, "grad_norm": 1.683119449492102, "language_loss": 0.67804283, "learning_rate": 7.969990738158417e-07, "loss": 0.69991452, "num_input_tokens_seen": 127667350, "step": 5940, "time_per_iteration": 3.1832683086395264 }, { "auxiliary_loss_clip": 0.01158372, "auxiliary_loss_mlp": 0.01028775, "balance_loss_clip": 1.04996276, "balance_loss_mlp": 1.02140808, "epoch": 0.7143630132868394, "flos": 21032380962240.0, "grad_norm": 2.1610492856509187, "language_loss": 0.85269463, "learning_rate": 7.963768660896062e-07, "loss": 0.87456608, "num_input_tokens_seen": 127685760, "step": 5941, "time_per_iteration": 3.241307497024536 }, { "auxiliary_loss_clip": 0.0115882, "auxiliary_loss_mlp": 0.0102526, "balance_loss_clip": 1.04828799, "balance_loss_mlp": 1.01841688, "epoch": 0.7144832561774785, "flos": 24129927173760.0, "grad_norm": 1.8599118095125955, "language_loss": 0.82371116, "learning_rate": 7.957548409520432e-07, "loss": 0.84555197, "num_input_tokens_seen": 127704985, "step": 5942, "time_per_iteration": 2.555194139480591 }, { "auxiliary_loss_clip": 0.01126909, "auxiliary_loss_mlp": 0.01026076, "balance_loss_clip": 1.04319584, "balance_loss_mlp": 1.01880693, "epoch": 0.7146034990681176, "flos": 16325820527040.0, "grad_norm": 1.8723669926243856, "language_loss": 0.84037822, "learning_rate": 7.951329984975135e-07, "loss": 0.86190808, "num_input_tokens_seen": 127721925, "step": 5943, "time_per_iteration": 2.5318665504455566 }, { "auxiliary_loss_clip": 0.01033839, "auxiliary_loss_mlp": 0.01002964, "balance_loss_clip": 1.00779963, "balance_loss_mlp": 1.00157499, "epoch": 0.7147237419587567, "flos": 69627175300800.0, "grad_norm": 0.7118412042018097, "language_loss": 0.54374737, "learning_rate": 7.94511338820349e-07, "loss": 0.56411541, "num_input_tokens_seen": 127784230, "step": 5944, "time_per_iteration": 3.154637336730957 }, { "auxiliary_loss_clip": 0.01145725, "auxiliary_loss_mlp": 0.00762763, "balance_loss_clip": 1.04859173, "balance_loss_mlp": 1.00059271, "epoch": 0.7148439848493958, "flos": 22266809259360.0, "grad_norm": 2.2168680396130913, "language_loss": 0.78162974, "learning_rate": 7.938898620148575e-07, "loss": 0.80071461, "num_input_tokens_seen": 127801990, "step": 5945, "time_per_iteration": 2.510688304901123 }, { "auxiliary_loss_clip": 0.01143079, "auxiliary_loss_mlp": 0.01031656, "balance_loss_clip": 1.04668903, "balance_loss_mlp": 1.0246346, "epoch": 0.7149642277400349, "flos": 17931386722560.0, "grad_norm": 2.0801082045743016, "language_loss": 0.70791131, "learning_rate": 7.932685681753135e-07, "loss": 0.72965872, "num_input_tokens_seen": 127819270, "step": 5946, "time_per_iteration": 2.49833083152771 }, { "auxiliary_loss_clip": 0.01168474, "auxiliary_loss_mlp": 0.01024927, "balance_loss_clip": 1.04864097, "balance_loss_mlp": 1.01738441, "epoch": 0.7150844706306739, "flos": 31681941479520.0, "grad_norm": 1.8609447247843065, "language_loss": 0.62771428, "learning_rate": 7.92647457395969e-07, "loss": 0.64964831, "num_input_tokens_seen": 127841095, "step": 5947, "time_per_iteration": 2.5591959953308105 }, { "auxiliary_loss_clip": 0.01106977, "auxiliary_loss_mlp": 0.01032548, "balance_loss_clip": 1.03848803, "balance_loss_mlp": 1.02530265, "epoch": 0.7152047135213131, "flos": 10926220058880.0, "grad_norm": 2.8093868620934934, "language_loss": 0.73443037, "learning_rate": 7.920265297710444e-07, "loss": 0.75582564, "num_input_tokens_seen": 127858485, "step": 5948, "time_per_iteration": 2.610968589782715 }, { "auxiliary_loss_clip": 0.01159845, "auxiliary_loss_mlp": 0.01021611, "balance_loss_clip": 1.04901016, "balance_loss_mlp": 1.01494169, "epoch": 0.7153249564119522, "flos": 20995644300480.0, "grad_norm": 2.2539351934050567, "language_loss": 0.73312402, "learning_rate": 7.914057853947363e-07, "loss": 0.7549386, "num_input_tokens_seen": 127877665, "step": 5949, "time_per_iteration": 3.211733102798462 }, { "auxiliary_loss_clip": 0.01126066, "auxiliary_loss_mlp": 0.01033566, "balance_loss_clip": 1.04326916, "balance_loss_mlp": 1.0258348, "epoch": 0.7154451993025912, "flos": 24243118266720.0, "grad_norm": 1.9087399021682214, "language_loss": 0.62147176, "learning_rate": 7.907852243612089e-07, "loss": 0.64306808, "num_input_tokens_seen": 127898070, "step": 5950, "time_per_iteration": 2.6048519611358643 }, { "auxiliary_loss_clip": 0.0114043, "auxiliary_loss_mlp": 0.01028929, "balance_loss_clip": 1.04539561, "balance_loss_mlp": 1.02168691, "epoch": 0.7155654421932304, "flos": 23330951273760.0, "grad_norm": 2.0467472001055333, "language_loss": 0.72453034, "learning_rate": 7.901648467646009e-07, "loss": 0.74622393, "num_input_tokens_seen": 127917010, "step": 5951, "time_per_iteration": 2.5369503498077393 }, { "auxiliary_loss_clip": 0.01172325, "auxiliary_loss_mlp": 0.01027328, "balance_loss_clip": 1.04850066, "balance_loss_mlp": 1.02004123, "epoch": 0.7156856850838694, "flos": 22711887401760.0, "grad_norm": 1.5393420148099468, "language_loss": 0.72419369, "learning_rate": 7.895446526990244e-07, "loss": 0.74619019, "num_input_tokens_seen": 127937025, "step": 5952, "time_per_iteration": 2.5026776790618896 }, { "auxiliary_loss_clip": 0.011245, "auxiliary_loss_mlp": 0.01026697, "balance_loss_clip": 1.0438, "balance_loss_mlp": 1.01964331, "epoch": 0.7158059279745085, "flos": 19865427856320.0, "grad_norm": 1.6442981486499484, "language_loss": 0.75378454, "learning_rate": 7.889246422585609e-07, "loss": 0.77529651, "num_input_tokens_seen": 127956410, "step": 5953, "time_per_iteration": 2.581803798675537 }, { "auxiliary_loss_clip": 0.01168437, "auxiliary_loss_mlp": 0.01023081, "balance_loss_clip": 1.0471915, "balance_loss_mlp": 1.01611042, "epoch": 0.7159261708651476, "flos": 24134775963360.0, "grad_norm": 1.6954961380754834, "language_loss": 0.73610842, "learning_rate": 7.883048155372675e-07, "loss": 0.75802362, "num_input_tokens_seen": 127974925, "step": 5954, "time_per_iteration": 2.5043070316314697 }, { "auxiliary_loss_clip": 0.0114692, "auxiliary_loss_mlp": 0.01028398, "balance_loss_clip": 1.04716659, "balance_loss_mlp": 1.0213021, "epoch": 0.7160464137557867, "flos": 16983201407040.0, "grad_norm": 2.481024336594289, "language_loss": 0.71282828, "learning_rate": 7.876851726291698e-07, "loss": 0.73458147, "num_input_tokens_seen": 127993225, "step": 5955, "time_per_iteration": 2.5096797943115234 }, { "auxiliary_loss_clip": 0.01132569, "auxiliary_loss_mlp": 0.01027411, "balance_loss_clip": 1.04350364, "balance_loss_mlp": 1.02033281, "epoch": 0.7161666566464258, "flos": 25228255745760.0, "grad_norm": 1.8144015191969582, "language_loss": 0.78467691, "learning_rate": 7.870657136282666e-07, "loss": 0.80627668, "num_input_tokens_seen": 128012085, "step": 5956, "time_per_iteration": 2.610142946243286 }, { "auxiliary_loss_clip": 0.0115042, "auxiliary_loss_mlp": 0.01027577, "balance_loss_clip": 1.04588068, "balance_loss_mlp": 1.0205909, "epoch": 0.7162868995370649, "flos": 26468394839520.0, "grad_norm": 1.4529229871178322, "language_loss": 0.81830895, "learning_rate": 7.86446438628531e-07, "loss": 0.84008896, "num_input_tokens_seen": 128033155, "step": 5957, "time_per_iteration": 2.545954465866089 }, { "auxiliary_loss_clip": 0.01065512, "auxiliary_loss_mlp": 0.01002074, "balance_loss_clip": 1.00875843, "balance_loss_mlp": 1.00070333, "epoch": 0.716407142427704, "flos": 69998923787520.0, "grad_norm": 0.7639448658004914, "language_loss": 0.56956846, "learning_rate": 7.858273477239059e-07, "loss": 0.59024429, "num_input_tokens_seen": 128101575, "step": 5958, "time_per_iteration": 3.10640811920166 }, { "auxiliary_loss_clip": 0.01101846, "auxiliary_loss_mlp": 0.01030159, "balance_loss_clip": 1.04205883, "balance_loss_mlp": 1.02273178, "epoch": 0.716527385318343, "flos": 20740463100960.0, "grad_norm": 1.8552336811153924, "language_loss": 0.71271175, "learning_rate": 7.852084410083067e-07, "loss": 0.7340318, "num_input_tokens_seen": 128120395, "step": 5959, "time_per_iteration": 2.60551118850708 }, { "auxiliary_loss_clip": 0.01137042, "auxiliary_loss_mlp": 0.01022548, "balance_loss_clip": 1.04618287, "balance_loss_mlp": 1.01642036, "epoch": 0.7166476282089821, "flos": 25371969866400.0, "grad_norm": 1.7032827283178402, "language_loss": 0.63727808, "learning_rate": 7.84589718575621e-07, "loss": 0.65887403, "num_input_tokens_seen": 128140840, "step": 5960, "time_per_iteration": 2.5856385231018066 }, { "auxiliary_loss_clip": 0.0114135, "auxiliary_loss_mlp": 0.01025746, "balance_loss_clip": 1.04098642, "balance_loss_mlp": 1.01860297, "epoch": 0.7167678710996213, "flos": 24133734371520.0, "grad_norm": 2.1801929785446323, "language_loss": 0.69172347, "learning_rate": 7.83971180519708e-07, "loss": 0.7133944, "num_input_tokens_seen": 128159695, "step": 5961, "time_per_iteration": 2.5467689037323 }, { "auxiliary_loss_clip": 0.01175787, "auxiliary_loss_mlp": 0.01025273, "balance_loss_clip": 1.05164802, "balance_loss_mlp": 1.01731527, "epoch": 0.7168881139902603, "flos": 30226590374400.0, "grad_norm": 2.90683371582269, "language_loss": 0.75600982, "learning_rate": 7.833528269344008e-07, "loss": 0.77802044, "num_input_tokens_seen": 128179600, "step": 5962, "time_per_iteration": 3.3366079330444336 }, { "auxiliary_loss_clip": 0.01129542, "auxiliary_loss_mlp": 0.0102544, "balance_loss_clip": 1.04710877, "balance_loss_mlp": 1.0178349, "epoch": 0.7170083568808994, "flos": 14606416733280.0, "grad_norm": 2.500862305852207, "language_loss": 0.77640873, "learning_rate": 7.827346579135023e-07, "loss": 0.79795861, "num_input_tokens_seen": 128196940, "step": 5963, "time_per_iteration": 2.6168203353881836 }, { "auxiliary_loss_clip": 0.01138597, "auxiliary_loss_mlp": 0.01023303, "balance_loss_clip": 1.04450405, "balance_loss_mlp": 1.01617098, "epoch": 0.7171285997715385, "flos": 23331094941600.0, "grad_norm": 1.8016700459940573, "language_loss": 0.82978201, "learning_rate": 7.821166735507885e-07, "loss": 0.85140103, "num_input_tokens_seen": 128215970, "step": 5964, "time_per_iteration": 2.5343410968780518 }, { "auxiliary_loss_clip": 0.01170581, "auxiliary_loss_mlp": 0.01018678, "balance_loss_clip": 1.0497129, "balance_loss_mlp": 1.01224101, "epoch": 0.7172488426621776, "flos": 16543546725600.0, "grad_norm": 1.5902620347846173, "language_loss": 0.68486178, "learning_rate": 7.81498873940007e-07, "loss": 0.70675433, "num_input_tokens_seen": 128233185, "step": 5965, "time_per_iteration": 2.4589152336120605 }, { "auxiliary_loss_clip": 0.01160512, "auxiliary_loss_mlp": 0.01026724, "balance_loss_clip": 1.04545712, "balance_loss_mlp": 1.01907969, "epoch": 0.7173690855528166, "flos": 26541616744320.0, "grad_norm": 2.1572761878589977, "language_loss": 0.76900852, "learning_rate": 7.808812591748768e-07, "loss": 0.7908808, "num_input_tokens_seen": 128253565, "step": 5966, "time_per_iteration": 4.010831117630005 }, { "auxiliary_loss_clip": 0.01126922, "auxiliary_loss_mlp": 0.01027921, "balance_loss_clip": 1.0449214, "balance_loss_mlp": 1.01979089, "epoch": 0.7174893284434558, "flos": 22784103631680.0, "grad_norm": 1.9487826603997072, "language_loss": 0.6467064, "learning_rate": 7.802638293490915e-07, "loss": 0.66825485, "num_input_tokens_seen": 128273210, "step": 5967, "time_per_iteration": 2.563575267791748 }, { "auxiliary_loss_clip": 0.01146112, "auxiliary_loss_mlp": 0.01025865, "balance_loss_clip": 1.04518926, "balance_loss_mlp": 1.01910543, "epoch": 0.7176095713340949, "flos": 23293568106720.0, "grad_norm": 1.8156143160845526, "language_loss": 0.77028638, "learning_rate": 7.796465845563123e-07, "loss": 0.79200613, "num_input_tokens_seen": 128292085, "step": 5968, "time_per_iteration": 2.550828218460083 }, { "auxiliary_loss_clip": 0.01139631, "auxiliary_loss_mlp": 0.00761756, "balance_loss_clip": 1.04779005, "balance_loss_mlp": 1.00062799, "epoch": 0.7177298142247339, "flos": 25591635580800.0, "grad_norm": 2.23642890685601, "language_loss": 0.79668552, "learning_rate": 7.790295248901766e-07, "loss": 0.81569934, "num_input_tokens_seen": 128313215, "step": 5969, "time_per_iteration": 2.5758895874023438 }, { "auxiliary_loss_clip": 0.01156591, "auxiliary_loss_mlp": 0.0102643, "balance_loss_clip": 1.04739261, "balance_loss_mlp": 1.01878905, "epoch": 0.7178500571153731, "flos": 31652783296320.0, "grad_norm": 1.804433501432797, "language_loss": 0.62361598, "learning_rate": 7.784126504442902e-07, "loss": 0.64544618, "num_input_tokens_seen": 128336445, "step": 5970, "time_per_iteration": 2.6139373779296875 }, { "auxiliary_loss_clip": 0.01123068, "auxiliary_loss_mlp": 0.01025399, "balance_loss_clip": 1.04661131, "balance_loss_mlp": 1.01837122, "epoch": 0.7179703000060121, "flos": 19427245770240.0, "grad_norm": 1.2994604756732626, "language_loss": 0.67770028, "learning_rate": 7.777959613122351e-07, "loss": 0.69918489, "num_input_tokens_seen": 128356270, "step": 5971, "time_per_iteration": 2.549039363861084 }, { "auxiliary_loss_clip": 0.0113674, "auxiliary_loss_mlp": 0.01023935, "balance_loss_clip": 1.0473814, "balance_loss_mlp": 1.01669574, "epoch": 0.7180905428966512, "flos": 28839253214880.0, "grad_norm": 1.7280452132761088, "language_loss": 0.7798183, "learning_rate": 7.771794575875604e-07, "loss": 0.8014251, "num_input_tokens_seen": 128378140, "step": 5972, "time_per_iteration": 2.56657075881958 }, { "auxiliary_loss_clip": 0.01159962, "auxiliary_loss_mlp": 0.01030529, "balance_loss_clip": 1.04961765, "balance_loss_mlp": 1.02332532, "epoch": 0.7182107857872904, "flos": 20047566735840.0, "grad_norm": 2.4450170018960415, "language_loss": 0.78144985, "learning_rate": 7.765631393637888e-07, "loss": 0.80335474, "num_input_tokens_seen": 128396335, "step": 5973, "time_per_iteration": 2.482959032058716 }, { "auxiliary_loss_clip": 0.01153483, "auxiliary_loss_mlp": 0.01028102, "balance_loss_clip": 1.04550016, "balance_loss_mlp": 1.02035379, "epoch": 0.7183310286779294, "flos": 22747690222560.0, "grad_norm": 2.699333518228367, "language_loss": 0.48735851, "learning_rate": 7.75947006734417e-07, "loss": 0.50917435, "num_input_tokens_seen": 128414115, "step": 5974, "time_per_iteration": 2.500009536743164 }, { "auxiliary_loss_clip": 0.01170648, "auxiliary_loss_mlp": 0.01022927, "balance_loss_clip": 1.04798269, "balance_loss_mlp": 1.01583958, "epoch": 0.7184512715685685, "flos": 17158264645440.0, "grad_norm": 1.8968100330836946, "language_loss": 0.82617986, "learning_rate": 7.753310597929101e-07, "loss": 0.84811568, "num_input_tokens_seen": 128430755, "step": 5975, "time_per_iteration": 2.4132163524627686 }, { "auxiliary_loss_clip": 0.01064454, "auxiliary_loss_mlp": 0.01001329, "balance_loss_clip": 1.0081594, "balance_loss_mlp": 0.99995822, "epoch": 0.7185715144592076, "flos": 65509622630400.0, "grad_norm": 0.7576036057034662, "language_loss": 0.55138493, "learning_rate": 7.747152986327095e-07, "loss": 0.57204276, "num_input_tokens_seen": 128491300, "step": 5976, "time_per_iteration": 3.72633957862854 }, { "auxiliary_loss_clip": 0.01119309, "auxiliary_loss_mlp": 0.01032399, "balance_loss_clip": 1.04417419, "balance_loss_mlp": 1.02604866, "epoch": 0.7186917573498467, "flos": 16180526060160.0, "grad_norm": 1.6171871647864111, "language_loss": 0.67936623, "learning_rate": 7.740997233472228e-07, "loss": 0.70088333, "num_input_tokens_seen": 128508920, "step": 5977, "time_per_iteration": 2.568176031112671 }, { "auxiliary_loss_clip": 0.01142407, "auxiliary_loss_mlp": 0.01025656, "balance_loss_clip": 1.04447865, "balance_loss_mlp": 1.01895666, "epoch": 0.7188120002404857, "flos": 29242278985440.0, "grad_norm": 3.0543113579187926, "language_loss": 0.70373845, "learning_rate": 7.734843340298329e-07, "loss": 0.7254191, "num_input_tokens_seen": 128528745, "step": 5978, "time_per_iteration": 2.5935261249542236 }, { "auxiliary_loss_clip": 0.01148139, "auxiliary_loss_mlp": 0.01032707, "balance_loss_clip": 1.0455184, "balance_loss_mlp": 1.0253191, "epoch": 0.7189322431311249, "flos": 33401165688480.0, "grad_norm": 1.8928350574452533, "language_loss": 0.74902391, "learning_rate": 7.72869130773895e-07, "loss": 0.7708323, "num_input_tokens_seen": 128549345, "step": 5979, "time_per_iteration": 2.6738510131835938 }, { "auxiliary_loss_clip": 0.01055519, "auxiliary_loss_mlp": 0.01001044, "balance_loss_clip": 1.00753856, "balance_loss_mlp": 0.99961978, "epoch": 0.719052486021764, "flos": 61351274681760.0, "grad_norm": 0.797073429693867, "language_loss": 0.59421659, "learning_rate": 7.722541136727343e-07, "loss": 0.61478221, "num_input_tokens_seen": 128605360, "step": 5980, "time_per_iteration": 2.9710514545440674 }, { "auxiliary_loss_clip": 0.01157072, "auxiliary_loss_mlp": 0.01025835, "balance_loss_clip": 1.04891062, "balance_loss_mlp": 1.0186882, "epoch": 0.719172728912403, "flos": 15596798088480.0, "grad_norm": 3.8322548133562093, "language_loss": 0.80790067, "learning_rate": 7.716392828196483e-07, "loss": 0.82972968, "num_input_tokens_seen": 128623160, "step": 5981, "time_per_iteration": 2.486152410507202 }, { "auxiliary_loss_clip": 0.01155812, "auxiliary_loss_mlp": 0.01028378, "balance_loss_clip": 1.04700685, "balance_loss_mlp": 1.02138007, "epoch": 0.7192929718030422, "flos": 15553165370400.0, "grad_norm": 2.5817696547325037, "language_loss": 0.77420676, "learning_rate": 7.710246383079064e-07, "loss": 0.7960487, "num_input_tokens_seen": 128638545, "step": 5982, "time_per_iteration": 2.4509732723236084 }, { "auxiliary_loss_clip": 0.01142762, "auxiliary_loss_mlp": 0.01026113, "balance_loss_clip": 1.04253948, "balance_loss_mlp": 1.01912153, "epoch": 0.7194132146936812, "flos": 21862490478240.0, "grad_norm": 2.527416502211908, "language_loss": 0.91693354, "learning_rate": 7.704101802307492e-07, "loss": 0.93862236, "num_input_tokens_seen": 128650845, "step": 5983, "time_per_iteration": 2.4969234466552734 }, { "auxiliary_loss_clip": 0.01121837, "auxiliary_loss_mlp": 0.01028192, "balance_loss_clip": 1.04391313, "balance_loss_mlp": 1.02118886, "epoch": 0.7195334575843203, "flos": 27338904547200.0, "grad_norm": 2.1935011747337607, "language_loss": 0.87046552, "learning_rate": 7.697959086813912e-07, "loss": 0.89196581, "num_input_tokens_seen": 128667010, "step": 5984, "time_per_iteration": 2.616210460662842 }, { "auxiliary_loss_clip": 0.0112082, "auxiliary_loss_mlp": 0.01025313, "balance_loss_clip": 1.04177165, "balance_loss_mlp": 1.01847935, "epoch": 0.7196537004749595, "flos": 18770619146400.0, "grad_norm": 1.8046293688773227, "language_loss": 0.80303288, "learning_rate": 7.691818237530145e-07, "loss": 0.82449412, "num_input_tokens_seen": 128685870, "step": 5985, "time_per_iteration": 2.5364089012145996 }, { "auxiliary_loss_clip": 0.0112887, "auxiliary_loss_mlp": 0.01025834, "balance_loss_clip": 1.04546368, "balance_loss_mlp": 1.019135, "epoch": 0.7197739433655985, "flos": 24531013428480.0, "grad_norm": 1.7336056726894058, "language_loss": 0.77291167, "learning_rate": 7.685679255387774e-07, "loss": 0.79445875, "num_input_tokens_seen": 128704185, "step": 5986, "time_per_iteration": 2.6200969219207764 }, { "auxiliary_loss_clip": 0.01141163, "auxiliary_loss_mlp": 0.01028501, "balance_loss_clip": 1.04694819, "balance_loss_mlp": 1.02154517, "epoch": 0.7198941862562376, "flos": 18040591032960.0, "grad_norm": 1.8654054524536423, "language_loss": 0.76611918, "learning_rate": 7.679542141318065e-07, "loss": 0.78781581, "num_input_tokens_seen": 128721290, "step": 5987, "time_per_iteration": 2.4933109283447266 }, { "auxiliary_loss_clip": 0.01129372, "auxiliary_loss_mlp": 0.01024432, "balance_loss_clip": 1.04244173, "balance_loss_mlp": 1.01723504, "epoch": 0.7200144291468767, "flos": 29022397769280.0, "grad_norm": 2.4996472442032354, "language_loss": 0.75481188, "learning_rate": 7.673406896252013e-07, "loss": 0.7763499, "num_input_tokens_seen": 128742665, "step": 5988, "time_per_iteration": 3.3217175006866455 }, { "auxiliary_loss_clip": 0.01126967, "auxiliary_loss_mlp": 0.01029902, "balance_loss_clip": 1.04236686, "balance_loss_mlp": 1.02254415, "epoch": 0.7201346720375158, "flos": 25374268551840.0, "grad_norm": 1.5062087139900457, "language_loss": 0.78456986, "learning_rate": 7.667273521120347e-07, "loss": 0.80613858, "num_input_tokens_seen": 128762225, "step": 5989, "time_per_iteration": 2.585645914077759 }, { "auxiliary_loss_clip": 0.01131074, "auxiliary_loss_mlp": 0.01023988, "balance_loss_clip": 1.04422021, "balance_loss_mlp": 1.01733303, "epoch": 0.7202549149281549, "flos": 14355617402880.0, "grad_norm": 1.9771823136191877, "language_loss": 0.79798818, "learning_rate": 7.661142016853468e-07, "loss": 0.81953871, "num_input_tokens_seen": 128779585, "step": 5990, "time_per_iteration": 2.5432257652282715 }, { "auxiliary_loss_clip": 0.011097, "auxiliary_loss_mlp": 0.01026089, "balance_loss_clip": 1.0412029, "balance_loss_mlp": 1.01938951, "epoch": 0.7203751578187939, "flos": 23001686162400.0, "grad_norm": 1.8325939263472641, "language_loss": 0.74882782, "learning_rate": 7.655012384381543e-07, "loss": 0.77018571, "num_input_tokens_seen": 128799070, "step": 5991, "time_per_iteration": 2.5961596965789795 }, { "auxiliary_loss_clip": 0.01141487, "auxiliary_loss_mlp": 0.01026193, "balance_loss_clip": 1.04838276, "balance_loss_mlp": 1.01862276, "epoch": 0.7204954007094331, "flos": 23692427509920.0, "grad_norm": 1.854211813331104, "language_loss": 0.815391, "learning_rate": 7.648884624634415e-07, "loss": 0.83706784, "num_input_tokens_seen": 128817620, "step": 5992, "time_per_iteration": 4.142524719238281 }, { "auxiliary_loss_clip": 0.01155402, "auxiliary_loss_mlp": 0.01023755, "balance_loss_clip": 1.04829574, "balance_loss_mlp": 1.01704955, "epoch": 0.7206156436000721, "flos": 16253029625760.0, "grad_norm": 2.0456462984216377, "language_loss": 0.88702261, "learning_rate": 7.642758738541683e-07, "loss": 0.90881419, "num_input_tokens_seen": 128834200, "step": 5993, "time_per_iteration": 2.4421753883361816 }, { "auxiliary_loss_clip": 0.01054483, "auxiliary_loss_mlp": 0.0099935, "balance_loss_clip": 1.00839615, "balance_loss_mlp": 0.99797881, "epoch": 0.7207358864907112, "flos": 54377816305440.0, "grad_norm": 0.7843228533018278, "language_loss": 0.60787278, "learning_rate": 7.636634727032621e-07, "loss": 0.62841111, "num_input_tokens_seen": 128891305, "step": 5994, "time_per_iteration": 2.9531538486480713 }, { "auxiliary_loss_clip": 0.01129148, "auxiliary_loss_mlp": 0.01030587, "balance_loss_clip": 1.03991365, "balance_loss_mlp": 1.02276087, "epoch": 0.7208561293813504, "flos": 19135543410720.0, "grad_norm": 1.93260438097037, "language_loss": 0.78669047, "learning_rate": 7.630512591036231e-07, "loss": 0.80828786, "num_input_tokens_seen": 128910615, "step": 5995, "time_per_iteration": 2.548553705215454 }, { "auxiliary_loss_clip": 0.01157858, "auxiliary_loss_mlp": 0.01027598, "balance_loss_clip": 1.04739022, "balance_loss_mlp": 1.02023101, "epoch": 0.7209763722719894, "flos": 17748529503840.0, "grad_norm": 2.1688569227623424, "language_loss": 0.64735836, "learning_rate": 7.624392331481255e-07, "loss": 0.66921294, "num_input_tokens_seen": 128928270, "step": 5996, "time_per_iteration": 2.477348566055298 }, { "auxiliary_loss_clip": 0.01053416, "auxiliary_loss_mlp": 0.01002603, "balance_loss_clip": 1.00868988, "balance_loss_mlp": 1.00129151, "epoch": 0.7210966151626285, "flos": 66819499683840.0, "grad_norm": 0.7434913604784067, "language_loss": 0.51854622, "learning_rate": 7.618273949296115e-07, "loss": 0.53910643, "num_input_tokens_seen": 128987780, "step": 5997, "time_per_iteration": 3.002774238586426 }, { "auxiliary_loss_clip": 0.01135034, "auxiliary_loss_mlp": 0.01029467, "balance_loss_clip": 1.04219127, "balance_loss_mlp": 1.02182579, "epoch": 0.7212168580532676, "flos": 21141872608320.0, "grad_norm": 1.8163445565346827, "language_loss": 0.68640864, "learning_rate": 7.612157445408987e-07, "loss": 0.70805365, "num_input_tokens_seen": 129005590, "step": 5998, "time_per_iteration": 2.5263707637786865 }, { "auxiliary_loss_clip": 0.01149843, "auxiliary_loss_mlp": 0.01022355, "balance_loss_clip": 1.05048943, "balance_loss_mlp": 1.01494658, "epoch": 0.7213371009439067, "flos": 22345741960800.0, "grad_norm": 4.574445569090875, "language_loss": 0.74636018, "learning_rate": 7.606042820747716e-07, "loss": 0.76808214, "num_input_tokens_seen": 129021995, "step": 5999, "time_per_iteration": 2.5247485637664795 }, { "auxiliary_loss_clip": 0.01149571, "auxiliary_loss_mlp": 0.01025738, "balance_loss_clip": 1.04962838, "balance_loss_mlp": 1.01887178, "epoch": 0.7214573438345457, "flos": 18515904867360.0, "grad_norm": 2.046491886708975, "language_loss": 0.85587114, "learning_rate": 7.599930076239889e-07, "loss": 0.87762427, "num_input_tokens_seen": 129039280, "step": 6000, "time_per_iteration": 2.5254132747650146 }, { "auxiliary_loss_clip": 0.01121335, "auxiliary_loss_mlp": 0.00761861, "balance_loss_clip": 1.04483581, "balance_loss_mlp": 1.00056481, "epoch": 0.7215775867251849, "flos": 35736113492160.0, "grad_norm": 2.948643882803082, "language_loss": 0.70414323, "learning_rate": 7.593819212812818e-07, "loss": 0.72297519, "num_input_tokens_seen": 129060860, "step": 6001, "time_per_iteration": 3.4447975158691406 }, { "auxiliary_loss_clip": 0.0115645, "auxiliary_loss_mlp": 0.01025866, "balance_loss_clip": 1.04950595, "balance_loss_mlp": 1.01907659, "epoch": 0.721697829615824, "flos": 20372414061120.0, "grad_norm": 2.2865495379511653, "language_loss": 0.71596944, "learning_rate": 7.587710231393508e-07, "loss": 0.73779261, "num_input_tokens_seen": 129079215, "step": 6002, "time_per_iteration": 2.486353635787964 }, { "auxiliary_loss_clip": 0.01077157, "auxiliary_loss_mlp": 0.01024283, "balance_loss_clip": 1.03885055, "balance_loss_mlp": 1.01707411, "epoch": 0.721818072506463, "flos": 20229813366240.0, "grad_norm": 1.9975909519051198, "language_loss": 0.838889, "learning_rate": 7.581603132908685e-07, "loss": 0.85990345, "num_input_tokens_seen": 129097185, "step": 6003, "time_per_iteration": 2.6243624687194824 }, { "auxiliary_loss_clip": 0.01123563, "auxiliary_loss_mlp": 0.01023814, "balance_loss_clip": 1.04441381, "balance_loss_mlp": 1.01602101, "epoch": 0.7219383153971022, "flos": 18186891174720.0, "grad_norm": 1.9208815234023753, "language_loss": 0.78394973, "learning_rate": 7.575497918284795e-07, "loss": 0.8054235, "num_input_tokens_seen": 129114730, "step": 6004, "time_per_iteration": 2.532862424850464 }, { "auxiliary_loss_clip": 0.01172548, "auxiliary_loss_mlp": 0.01029078, "balance_loss_clip": 1.04787266, "balance_loss_mlp": 1.02167213, "epoch": 0.7220585582877412, "flos": 17342127539040.0, "grad_norm": 2.199449763711036, "language_loss": 0.7457118, "learning_rate": 7.569394588447984e-07, "loss": 0.76772809, "num_input_tokens_seen": 129131745, "step": 6005, "time_per_iteration": 2.4269137382507324 }, { "auxiliary_loss_clip": 0.01146195, "auxiliary_loss_mlp": 0.01024288, "balance_loss_clip": 1.04408288, "balance_loss_mlp": 1.01740408, "epoch": 0.7221788011783803, "flos": 16976341267680.0, "grad_norm": 2.529029184887823, "language_loss": 0.78435582, "learning_rate": 7.563293144324146e-07, "loss": 0.80606067, "num_input_tokens_seen": 129147295, "step": 6006, "time_per_iteration": 2.4462392330169678 }, { "auxiliary_loss_clip": 0.01168338, "auxiliary_loss_mlp": 0.01021674, "balance_loss_clip": 1.04844689, "balance_loss_mlp": 1.01545763, "epoch": 0.7222990440690195, "flos": 26286363710880.0, "grad_norm": 1.7490550879497873, "language_loss": 0.79881316, "learning_rate": 7.557193586838834e-07, "loss": 0.82071328, "num_input_tokens_seen": 129162660, "step": 6007, "time_per_iteration": 2.480748176574707 }, { "auxiliary_loss_clip": 0.01143153, "auxiliary_loss_mlp": 0.01028031, "balance_loss_clip": 1.0462482, "balance_loss_mlp": 1.02096152, "epoch": 0.7224192869596585, "flos": 17601690607680.0, "grad_norm": 2.236151083047554, "language_loss": 0.70173347, "learning_rate": 7.551095916917371e-07, "loss": 0.7234453, "num_input_tokens_seen": 129179990, "step": 6008, "time_per_iteration": 2.4769952297210693 }, { "auxiliary_loss_clip": 0.0113852, "auxiliary_loss_mlp": 0.01033891, "balance_loss_clip": 1.04769707, "balance_loss_mlp": 1.02534962, "epoch": 0.7225395298502976, "flos": 12932333754720.0, "grad_norm": 2.5829619359971554, "language_loss": 0.66468304, "learning_rate": 7.545000135484758e-07, "loss": 0.68640715, "num_input_tokens_seen": 129197425, "step": 6009, "time_per_iteration": 2.5242090225219727 }, { "auxiliary_loss_clip": 0.0117171, "auxiliary_loss_mlp": 0.00762518, "balance_loss_clip": 1.04942441, "balance_loss_mlp": 1.00054693, "epoch": 0.7226597727409367, "flos": 29643903994560.0, "grad_norm": 2.0664415535406055, "language_loss": 0.62368739, "learning_rate": 7.538906243465714e-07, "loss": 0.64302969, "num_input_tokens_seen": 129217560, "step": 6010, "time_per_iteration": 2.5404231548309326 }, { "auxiliary_loss_clip": 0.01173655, "auxiliary_loss_mlp": 0.01030534, "balance_loss_clip": 1.0501039, "balance_loss_mlp": 1.0226692, "epoch": 0.7227800156315758, "flos": 13771638012480.0, "grad_norm": 2.0394579917221893, "language_loss": 0.78902066, "learning_rate": 7.5328142417847e-07, "loss": 0.81106257, "num_input_tokens_seen": 129234325, "step": 6011, "time_per_iteration": 2.442434787750244 }, { "auxiliary_loss_clip": 0.01151995, "auxiliary_loss_mlp": 0.01021902, "balance_loss_clip": 1.04484653, "balance_loss_mlp": 1.0148176, "epoch": 0.7229002585222148, "flos": 20301885928320.0, "grad_norm": 1.8435876225043555, "language_loss": 0.69350386, "learning_rate": 7.526724131365838e-07, "loss": 0.71524286, "num_input_tokens_seen": 129255280, "step": 6012, "time_per_iteration": 2.4953503608703613 }, { "auxiliary_loss_clip": 0.01144673, "auxiliary_loss_mlp": 0.01030993, "balance_loss_clip": 1.05033529, "balance_loss_mlp": 1.02315474, "epoch": 0.723020501412854, "flos": 16581253145280.0, "grad_norm": 1.6589683654036642, "language_loss": 0.7029115, "learning_rate": 7.520635913133017e-07, "loss": 0.72466809, "num_input_tokens_seen": 129273910, "step": 6013, "time_per_iteration": 2.4838616847991943 }, { "auxiliary_loss_clip": 0.01161519, "auxiliary_loss_mlp": 0.01031054, "balance_loss_clip": 1.04716182, "balance_loss_mlp": 1.02323937, "epoch": 0.7231407443034931, "flos": 28548305111520.0, "grad_norm": 2.709071484169308, "language_loss": 0.82628977, "learning_rate": 7.514549588009798e-07, "loss": 0.84821552, "num_input_tokens_seen": 129294785, "step": 6014, "time_per_iteration": 3.2990493774414062 }, { "auxiliary_loss_clip": 0.01147811, "auxiliary_loss_mlp": 0.01023827, "balance_loss_clip": 1.04802465, "balance_loss_mlp": 1.01629329, "epoch": 0.7232609871941321, "flos": 30008540923200.0, "grad_norm": 1.9817526957859286, "language_loss": 0.70328462, "learning_rate": 7.508465156919492e-07, "loss": 0.72500098, "num_input_tokens_seen": 129318295, "step": 6015, "time_per_iteration": 2.5941009521484375 }, { "auxiliary_loss_clip": 0.01142766, "auxiliary_loss_mlp": 0.01024035, "balance_loss_clip": 1.0455519, "balance_loss_mlp": 1.01635218, "epoch": 0.7233812300847713, "flos": 16654008129600.0, "grad_norm": 3.6315146112882832, "language_loss": 0.61341035, "learning_rate": 7.502382620785083e-07, "loss": 0.63507843, "num_input_tokens_seen": 129334845, "step": 6016, "time_per_iteration": 2.4823317527770996 }, { "auxiliary_loss_clip": 0.01033362, "auxiliary_loss_mlp": 0.01003418, "balance_loss_clip": 1.01376736, "balance_loss_mlp": 1.00226748, "epoch": 0.7235014729754103, "flos": 67258795195680.0, "grad_norm": 0.8085022159793679, "language_loss": 0.62503445, "learning_rate": 7.496301980529289e-07, "loss": 0.64540231, "num_input_tokens_seen": 129398055, "step": 6017, "time_per_iteration": 3.1756787300109863 }, { "auxiliary_loss_clip": 0.01171566, "auxiliary_loss_mlp": 0.01027917, "balance_loss_clip": 1.04923153, "balance_loss_mlp": 1.02137232, "epoch": 0.7236217158660494, "flos": 26943241753440.0, "grad_norm": 2.2419584325708364, "language_loss": 0.74855924, "learning_rate": 7.490223237074547e-07, "loss": 0.77055407, "num_input_tokens_seen": 129417765, "step": 6018, "time_per_iteration": 4.040271997451782 }, { "auxiliary_loss_clip": 0.01127146, "auxiliary_loss_mlp": 0.01027505, "balance_loss_clip": 1.04235673, "balance_loss_mlp": 1.02014351, "epoch": 0.7237419587566886, "flos": 29423376273120.0, "grad_norm": 1.9507569070709914, "language_loss": 0.66326547, "learning_rate": 7.484146391342989e-07, "loss": 0.68481195, "num_input_tokens_seen": 129437560, "step": 6019, "time_per_iteration": 2.65087628364563 }, { "auxiliary_loss_clip": 0.01135377, "auxiliary_loss_mlp": 0.01022234, "balance_loss_clip": 1.04374814, "balance_loss_mlp": 1.0153054, "epoch": 0.7238622016473276, "flos": 17821499989920.0, "grad_norm": 2.249910620411647, "language_loss": 0.57019734, "learning_rate": 7.478071444256484e-07, "loss": 0.59177345, "num_input_tokens_seen": 129455320, "step": 6020, "time_per_iteration": 2.5034379959106445 }, { "auxiliary_loss_clip": 0.01134391, "auxiliary_loss_mlp": 0.01027932, "balance_loss_clip": 1.0440073, "balance_loss_mlp": 1.02058542, "epoch": 0.7239824445379667, "flos": 25739120982240.0, "grad_norm": 1.9087298999429336, "language_loss": 0.79670227, "learning_rate": 7.471998396736579e-07, "loss": 0.8183254, "num_input_tokens_seen": 129475700, "step": 6021, "time_per_iteration": 2.6045002937316895 }, { "auxiliary_loss_clip": 0.01130836, "auxiliary_loss_mlp": 0.01024203, "balance_loss_clip": 1.04530144, "balance_loss_mlp": 1.01745319, "epoch": 0.7241026874286057, "flos": 23148920145120.0, "grad_norm": 1.9984548618956057, "language_loss": 0.76150811, "learning_rate": 7.465927249704549e-07, "loss": 0.78305846, "num_input_tokens_seen": 129493585, "step": 6022, "time_per_iteration": 2.5781235694885254 }, { "auxiliary_loss_clip": 0.01154102, "auxiliary_loss_mlp": 0.01024559, "balance_loss_clip": 1.04611576, "balance_loss_mlp": 1.0173769, "epoch": 0.7242229303192449, "flos": 20266909197600.0, "grad_norm": 1.9785153927779935, "language_loss": 0.77465802, "learning_rate": 7.459858004081398e-07, "loss": 0.79644465, "num_input_tokens_seen": 129511555, "step": 6023, "time_per_iteration": 2.4674785137176514 }, { "auxiliary_loss_clip": 0.0102602, "auxiliary_loss_mlp": 0.0100375, "balance_loss_clip": 1.00657129, "balance_loss_mlp": 1.00246847, "epoch": 0.724343173209884, "flos": 62311671763680.0, "grad_norm": 0.6554118827676895, "language_loss": 0.58058411, "learning_rate": 7.453790660787815e-07, "loss": 0.60088181, "num_input_tokens_seen": 129579650, "step": 6024, "time_per_iteration": 3.228658437728882 }, { "auxiliary_loss_clip": 0.01145945, "auxiliary_loss_mlp": 0.01032236, "balance_loss_clip": 1.04826832, "balance_loss_mlp": 1.02426076, "epoch": 0.724463416100523, "flos": 35006408631360.0, "grad_norm": 2.5539851265755664, "language_loss": 0.63476139, "learning_rate": 7.447725220744214e-07, "loss": 0.6565432, "num_input_tokens_seen": 129601895, "step": 6025, "time_per_iteration": 2.6097164154052734 }, { "auxiliary_loss_clip": 0.01171984, "auxiliary_loss_mlp": 0.01028569, "balance_loss_clip": 1.04878521, "balance_loss_mlp": 1.02136862, "epoch": 0.7245836589911622, "flos": 21871972555680.0, "grad_norm": 2.1722887638373103, "language_loss": 0.77093017, "learning_rate": 7.441661684870717e-07, "loss": 0.79293573, "num_input_tokens_seen": 129622150, "step": 6026, "time_per_iteration": 2.5023674964904785 }, { "auxiliary_loss_clip": 0.01170807, "auxiliary_loss_mlp": 0.0102744, "balance_loss_clip": 1.04848969, "balance_loss_mlp": 1.01984084, "epoch": 0.7247039018818012, "flos": 23006499035040.0, "grad_norm": 1.5681958619511158, "language_loss": 0.81675184, "learning_rate": 7.435600054087152e-07, "loss": 0.83873433, "num_input_tokens_seen": 129644315, "step": 6027, "time_per_iteration": 3.2381041049957275 }, { "auxiliary_loss_clip": 0.01174639, "auxiliary_loss_mlp": 0.01027718, "balance_loss_clip": 1.05068827, "balance_loss_mlp": 1.02059817, "epoch": 0.7248241447724403, "flos": 31722593089920.0, "grad_norm": 1.8282020015124374, "language_loss": 0.74097055, "learning_rate": 7.42954032931308e-07, "loss": 0.76299411, "num_input_tokens_seen": 129665355, "step": 6028, "time_per_iteration": 2.5181028842926025 }, { "auxiliary_loss_clip": 0.0114635, "auxiliary_loss_mlp": 0.01026729, "balance_loss_clip": 1.04747963, "balance_loss_mlp": 1.01990402, "epoch": 0.7249443876630794, "flos": 34896988819200.0, "grad_norm": 1.7606154234147793, "language_loss": 0.74682516, "learning_rate": 7.423482511467733e-07, "loss": 0.76855594, "num_input_tokens_seen": 129686125, "step": 6029, "time_per_iteration": 2.6126365661621094 }, { "auxiliary_loss_clip": 0.01090144, "auxiliary_loss_mlp": 0.01028348, "balance_loss_clip": 1.04193282, "balance_loss_mlp": 1.02106392, "epoch": 0.7250646305537185, "flos": 26359298280000.0, "grad_norm": 2.4685158512374814, "language_loss": 0.64907467, "learning_rate": 7.417426601470099e-07, "loss": 0.67025959, "num_input_tokens_seen": 129706485, "step": 6030, "time_per_iteration": 2.629239320755005 }, { "auxiliary_loss_clip": 0.01159924, "auxiliary_loss_mlp": 0.01024786, "balance_loss_clip": 1.04835975, "balance_loss_mlp": 1.01717448, "epoch": 0.7251848734443576, "flos": 30081619160160.0, "grad_norm": 2.3254988433158257, "language_loss": 0.7838589, "learning_rate": 7.411372600238841e-07, "loss": 0.80570596, "num_input_tokens_seen": 129727100, "step": 6031, "time_per_iteration": 2.536306142807007 }, { "auxiliary_loss_clip": 0.01172241, "auxiliary_loss_mlp": 0.01025692, "balance_loss_clip": 1.04915285, "balance_loss_mlp": 1.01871192, "epoch": 0.7253051163349967, "flos": 17785266165600.0, "grad_norm": 1.8131802766337932, "language_loss": 0.74062824, "learning_rate": 7.405320508692346e-07, "loss": 0.76260757, "num_input_tokens_seen": 129745840, "step": 6032, "time_per_iteration": 2.4339380264282227 }, { "auxiliary_loss_clip": 0.01167848, "auxiliary_loss_mlp": 0.01031008, "balance_loss_clip": 1.04942727, "balance_loss_mlp": 1.02426958, "epoch": 0.7254253592256358, "flos": 12641349734400.0, "grad_norm": 1.8220067841254746, "language_loss": 0.75542843, "learning_rate": 7.399270327748727e-07, "loss": 0.777417, "num_input_tokens_seen": 129763500, "step": 6033, "time_per_iteration": 2.457484483718872 }, { "auxiliary_loss_clip": 0.01129347, "auxiliary_loss_mlp": 0.00761151, "balance_loss_clip": 1.04389513, "balance_loss_mlp": 1.00053704, "epoch": 0.7255456021162748, "flos": 27199213126080.0, "grad_norm": 1.7049900749174953, "language_loss": 0.74499857, "learning_rate": 7.39322205832577e-07, "loss": 0.7639035, "num_input_tokens_seen": 129784390, "step": 6034, "time_per_iteration": 2.621302843093872 }, { "auxiliary_loss_clip": 0.01137318, "auxiliary_loss_mlp": 0.01023511, "balance_loss_clip": 1.0447042, "balance_loss_mlp": 1.01665998, "epoch": 0.725665845006914, "flos": 21288208667040.0, "grad_norm": 1.9136581353671382, "language_loss": 0.80575514, "learning_rate": 7.387175701341009e-07, "loss": 0.82736343, "num_input_tokens_seen": 129803060, "step": 6035, "time_per_iteration": 2.508227825164795 }, { "auxiliary_loss_clip": 0.01155657, "auxiliary_loss_mlp": 0.01023241, "balance_loss_clip": 1.04665446, "balance_loss_mlp": 1.01597559, "epoch": 0.7257860878975531, "flos": 16033687164000.0, "grad_norm": 2.367986914084262, "language_loss": 0.72244084, "learning_rate": 7.381131257711659e-07, "loss": 0.74422979, "num_input_tokens_seen": 129820165, "step": 6036, "time_per_iteration": 2.4670257568359375 }, { "auxiliary_loss_clip": 0.01140867, "auxiliary_loss_mlp": 0.01027989, "balance_loss_clip": 1.04901123, "balance_loss_mlp": 1.02081275, "epoch": 0.7259063307881921, "flos": 12129945743520.0, "grad_norm": 1.8995319711689116, "language_loss": 0.83551949, "learning_rate": 7.375088728354677e-07, "loss": 0.85720801, "num_input_tokens_seen": 129835195, "step": 6037, "time_per_iteration": 2.4756791591644287 }, { "auxiliary_loss_clip": 0.01131313, "auxiliary_loss_mlp": 0.01021191, "balance_loss_clip": 1.04370391, "balance_loss_mlp": 1.01413417, "epoch": 0.7260265736788313, "flos": 30443849652480.0, "grad_norm": 1.6488281391284145, "language_loss": 0.67437112, "learning_rate": 7.369048114186691e-07, "loss": 0.69589615, "num_input_tokens_seen": 129856240, "step": 6038, "time_per_iteration": 2.629857063293457 }, { "auxiliary_loss_clip": 0.01132422, "auxiliary_loss_mlp": 0.00761519, "balance_loss_clip": 1.04483461, "balance_loss_mlp": 1.00053287, "epoch": 0.7261468165694703, "flos": 21142267694880.0, "grad_norm": 2.338006812297607, "language_loss": 0.82988191, "learning_rate": 7.363009416124055e-07, "loss": 0.8488214, "num_input_tokens_seen": 129875565, "step": 6039, "time_per_iteration": 2.582075834274292 }, { "auxiliary_loss_clip": 0.01131836, "auxiliary_loss_mlp": 0.01026145, "balance_loss_clip": 1.04445195, "balance_loss_mlp": 1.01904309, "epoch": 0.7262670594601094, "flos": 22306311527040.0, "grad_norm": 2.1796345015541476, "language_loss": 0.62712097, "learning_rate": 7.356972635082852e-07, "loss": 0.64870083, "num_input_tokens_seen": 129894420, "step": 6040, "time_per_iteration": 2.5346736907958984 }, { "auxiliary_loss_clip": 0.01115205, "auxiliary_loss_mlp": 0.01029941, "balance_loss_clip": 1.04773664, "balance_loss_mlp": 1.02238655, "epoch": 0.7263873023507486, "flos": 25335053619840.0, "grad_norm": 2.100615083098996, "language_loss": 0.75744522, "learning_rate": 7.35093777197884e-07, "loss": 0.77889669, "num_input_tokens_seen": 129914490, "step": 6041, "time_per_iteration": 3.3365557193756104 }, { "auxiliary_loss_clip": 0.01141143, "auxiliary_loss_mlp": 0.01025794, "balance_loss_clip": 1.04653704, "balance_loss_mlp": 1.01905882, "epoch": 0.7265075452413876, "flos": 23878625005920.0, "grad_norm": 2.180592060759986, "language_loss": 0.8601526, "learning_rate": 7.344904827727525e-07, "loss": 0.88182193, "num_input_tokens_seen": 129931670, "step": 6042, "time_per_iteration": 2.5248992443084717 }, { "auxiliary_loss_clip": 0.0112594, "auxiliary_loss_mlp": 0.01025822, "balance_loss_clip": 1.04031217, "balance_loss_mlp": 1.01866031, "epoch": 0.7266277881320267, "flos": 28724553609600.0, "grad_norm": 2.882954842280191, "language_loss": 0.73427081, "learning_rate": 7.338873803244076e-07, "loss": 0.75578845, "num_input_tokens_seen": 129946905, "step": 6043, "time_per_iteration": 3.3362011909484863 }, { "auxiliary_loss_clip": 0.01137382, "auxiliary_loss_mlp": 0.01027597, "balance_loss_clip": 1.04526806, "balance_loss_mlp": 1.02070689, "epoch": 0.7267480310226658, "flos": 24863511066240.0, "grad_norm": 2.162032529745579, "language_loss": 0.80550539, "learning_rate": 7.332844699443401e-07, "loss": 0.82715523, "num_input_tokens_seen": 129965505, "step": 6044, "time_per_iteration": 3.3199856281280518 }, { "auxiliary_loss_clip": 0.01106316, "auxiliary_loss_mlp": 0.01021712, "balance_loss_clip": 1.04116488, "balance_loss_mlp": 1.01529837, "epoch": 0.7268682739133049, "flos": 27198494786880.0, "grad_norm": 2.7222485795429785, "language_loss": 0.75497431, "learning_rate": 7.326817517240121e-07, "loss": 0.77625459, "num_input_tokens_seen": 129987210, "step": 6045, "time_per_iteration": 2.63297438621521 }, { "auxiliary_loss_clip": 0.0115541, "auxiliary_loss_mlp": 0.00761639, "balance_loss_clip": 1.04575324, "balance_loss_mlp": 1.00053859, "epoch": 0.7269885168039439, "flos": 33508143147360.0, "grad_norm": 1.953713765854072, "language_loss": 0.82774746, "learning_rate": 7.320792257548545e-07, "loss": 0.84691799, "num_input_tokens_seen": 130008385, "step": 6046, "time_per_iteration": 2.6212735176086426 }, { "auxiliary_loss_clip": 0.01146853, "auxiliary_loss_mlp": 0.01026858, "balance_loss_clip": 1.04605961, "balance_loss_mlp": 1.01967001, "epoch": 0.7271087596945831, "flos": 24313754150400.0, "grad_norm": 1.9661263336706116, "language_loss": 0.75789434, "learning_rate": 7.314768921282704e-07, "loss": 0.7796315, "num_input_tokens_seen": 130029040, "step": 6047, "time_per_iteration": 2.5391366481781006 }, { "auxiliary_loss_clip": 0.01157759, "auxiliary_loss_mlp": 0.01027181, "balance_loss_clip": 1.04659379, "balance_loss_mlp": 1.02066922, "epoch": 0.7272290025852222, "flos": 23805151682400.0, "grad_norm": 2.5297091441423443, "language_loss": 0.71629119, "learning_rate": 7.30874750935633e-07, "loss": 0.73814058, "num_input_tokens_seen": 130048725, "step": 6048, "time_per_iteration": 2.5099024772644043 }, { "auxiliary_loss_clip": 0.01127077, "auxiliary_loss_mlp": 0.01027625, "balance_loss_clip": 1.04492092, "balance_loss_mlp": 1.02084243, "epoch": 0.7273492454758612, "flos": 16720369895040.0, "grad_norm": 1.9263114748470629, "language_loss": 0.79351521, "learning_rate": 7.30272802268286e-07, "loss": 0.81506222, "num_input_tokens_seen": 130065720, "step": 6049, "time_per_iteration": 2.5130460262298584 }, { "auxiliary_loss_clip": 0.01074479, "auxiliary_loss_mlp": 0.01021733, "balance_loss_clip": 1.03768873, "balance_loss_mlp": 1.01532269, "epoch": 0.7274694883665004, "flos": 28031333991840.0, "grad_norm": 1.7836032419603336, "language_loss": 0.76066476, "learning_rate": 7.29671046217547e-07, "loss": 0.78162688, "num_input_tokens_seen": 130084830, "step": 6050, "time_per_iteration": 2.669555425643921 }, { "auxiliary_loss_clip": 0.01129651, "auxiliary_loss_mlp": 0.01023076, "balance_loss_clip": 1.04376698, "balance_loss_mlp": 1.0164361, "epoch": 0.7275897312571394, "flos": 30372710931360.0, "grad_norm": 5.446392578478732, "language_loss": 0.81421429, "learning_rate": 7.290694828746988e-07, "loss": 0.83574152, "num_input_tokens_seen": 130104495, "step": 6051, "time_per_iteration": 2.622053384780884 }, { "auxiliary_loss_clip": 0.01131687, "auxiliary_loss_mlp": 0.01026999, "balance_loss_clip": 1.04305816, "balance_loss_mlp": 1.01975369, "epoch": 0.7277099741477785, "flos": 19204778532960.0, "grad_norm": 1.9511659360903502, "language_loss": 0.8565048, "learning_rate": 7.284681123310004e-07, "loss": 0.87809157, "num_input_tokens_seen": 130123210, "step": 6052, "time_per_iteration": 2.5683321952819824 }, { "auxiliary_loss_clip": 0.01157773, "auxiliary_loss_mlp": 0.01026704, "balance_loss_clip": 1.04819727, "balance_loss_mlp": 1.01981354, "epoch": 0.7278302170384175, "flos": 20667887701440.0, "grad_norm": 1.6660852167553253, "language_loss": 0.79702687, "learning_rate": 7.27866934677678e-07, "loss": 0.81887162, "num_input_tokens_seen": 130142880, "step": 6053, "time_per_iteration": 3.183371067047119 }, { "auxiliary_loss_clip": 0.01112412, "auxiliary_loss_mlp": 0.01027583, "balance_loss_clip": 1.04405403, "balance_loss_mlp": 1.02072871, "epoch": 0.7279504599290567, "flos": 19093203703200.0, "grad_norm": 1.623513284633398, "language_loss": 0.78201348, "learning_rate": 7.272659500059297e-07, "loss": 0.80341345, "num_input_tokens_seen": 130160220, "step": 6054, "time_per_iteration": 2.5555927753448486 }, { "auxiliary_loss_clip": 0.01151687, "auxiliary_loss_mlp": 0.0103075, "balance_loss_clip": 1.04753292, "balance_loss_mlp": 1.02355564, "epoch": 0.7280707028196958, "flos": 19062177838080.0, "grad_norm": 2.054431144600042, "language_loss": 0.80260807, "learning_rate": 7.266651584069264e-07, "loss": 0.82443237, "num_input_tokens_seen": 130177885, "step": 6055, "time_per_iteration": 2.4758689403533936 }, { "auxiliary_loss_clip": 0.01160581, "auxiliary_loss_mlp": 0.01027738, "balance_loss_clip": 1.05095506, "balance_loss_mlp": 1.02098811, "epoch": 0.7281909457103348, "flos": 37196313386880.0, "grad_norm": 1.6956559248530718, "language_loss": 0.57292056, "learning_rate": 7.260645599718045e-07, "loss": 0.59480375, "num_input_tokens_seen": 130204240, "step": 6056, "time_per_iteration": 2.639875650405884 }, { "auxiliary_loss_clip": 0.01145315, "auxiliary_loss_mlp": 0.010325, "balance_loss_clip": 1.04712546, "balance_loss_mlp": 1.02505505, "epoch": 0.728311188600974, "flos": 20667097528320.0, "grad_norm": 3.3607756486640703, "language_loss": 0.67067873, "learning_rate": 7.254641547916767e-07, "loss": 0.6924569, "num_input_tokens_seen": 130221735, "step": 6057, "time_per_iteration": 2.505603551864624 }, { "auxiliary_loss_clip": 0.01172524, "auxiliary_loss_mlp": 0.01023965, "balance_loss_clip": 1.05144739, "balance_loss_mlp": 1.01665187, "epoch": 0.728431431491613, "flos": 28840689893280.0, "grad_norm": 1.7810929357325787, "language_loss": 0.69416666, "learning_rate": 7.248639429576226e-07, "loss": 0.71613157, "num_input_tokens_seen": 130241190, "step": 6058, "time_per_iteration": 2.5090043544769287 }, { "auxiliary_loss_clip": 0.01157809, "auxiliary_loss_mlp": 0.01023518, "balance_loss_clip": 1.04714727, "balance_loss_mlp": 1.01665401, "epoch": 0.7285516743822521, "flos": 25991859828480.0, "grad_norm": 1.6609897820766906, "language_loss": 0.72060359, "learning_rate": 7.242639245606959e-07, "loss": 0.74241686, "num_input_tokens_seen": 130260980, "step": 6059, "time_per_iteration": 2.5319643020629883 }, { "auxiliary_loss_clip": 0.01145932, "auxiliary_loss_mlp": 0.0102409, "balance_loss_clip": 1.04385352, "balance_loss_mlp": 1.01710439, "epoch": 0.7286719172728913, "flos": 16399724854080.0, "grad_norm": 1.6695309137460932, "language_loss": 0.8209213, "learning_rate": 7.236640996919168e-07, "loss": 0.84262151, "num_input_tokens_seen": 130280025, "step": 6060, "time_per_iteration": 2.5111076831817627 }, { "auxiliary_loss_clip": 0.01161103, "auxiliary_loss_mlp": 0.01022269, "balance_loss_clip": 1.04928684, "balance_loss_mlp": 1.01587653, "epoch": 0.7287921601635303, "flos": 22018164946560.0, "grad_norm": 1.6594248989871454, "language_loss": 0.70623457, "learning_rate": 7.230644684422782e-07, "loss": 0.72806829, "num_input_tokens_seen": 130300255, "step": 6061, "time_per_iteration": 2.490004062652588 }, { "auxiliary_loss_clip": 0.0112605, "auxiliary_loss_mlp": 0.01027524, "balance_loss_clip": 1.04307199, "balance_loss_mlp": 1.02031767, "epoch": 0.7289124030541694, "flos": 24600930972960.0, "grad_norm": 2.4003889522801782, "language_loss": 0.81937307, "learning_rate": 7.224650309027451e-07, "loss": 0.84090877, "num_input_tokens_seen": 130320005, "step": 6062, "time_per_iteration": 2.5818381309509277 }, { "auxiliary_loss_clip": 0.0116078, "auxiliary_loss_mlp": 0.0102601, "balance_loss_clip": 1.04954612, "balance_loss_mlp": 1.01894689, "epoch": 0.7290326459448085, "flos": 21393641696640.0, "grad_norm": 2.2567461931227326, "language_loss": 0.685812, "learning_rate": 7.218657871642506e-07, "loss": 0.70767987, "num_input_tokens_seen": 130338810, "step": 6063, "time_per_iteration": 2.4841487407684326 }, { "auxiliary_loss_clip": 0.01174705, "auxiliary_loss_mlp": 0.01023767, "balance_loss_clip": 1.0502106, "balance_loss_mlp": 1.01628065, "epoch": 0.7291528888354476, "flos": 18587690093760.0, "grad_norm": 2.110919653498846, "language_loss": 0.62275934, "learning_rate": 7.212667373177012e-07, "loss": 0.64474404, "num_input_tokens_seen": 130353805, "step": 6064, "time_per_iteration": 2.4315991401672363 }, { "auxiliary_loss_clip": 0.01128881, "auxiliary_loss_mlp": 0.01028263, "balance_loss_clip": 1.04484725, "balance_loss_mlp": 1.02103853, "epoch": 0.7292731317260867, "flos": 18951069928800.0, "grad_norm": 1.852705677661496, "language_loss": 0.75222766, "learning_rate": 7.206678814539704e-07, "loss": 0.77379906, "num_input_tokens_seen": 130372105, "step": 6065, "time_per_iteration": 2.5249993801116943 }, { "auxiliary_loss_clip": 0.01123135, "auxiliary_loss_mlp": 0.01023278, "balance_loss_clip": 1.0432142, "balance_loss_mlp": 1.01703489, "epoch": 0.7293933746167258, "flos": 21067573194720.0, "grad_norm": 1.6013636586544857, "language_loss": 0.72799861, "learning_rate": 7.20069219663904e-07, "loss": 0.74946278, "num_input_tokens_seen": 130391990, "step": 6066, "time_per_iteration": 2.577291965484619 }, { "auxiliary_loss_clip": 0.01158536, "auxiliary_loss_mlp": 0.0102668, "balance_loss_clip": 1.04703867, "balance_loss_mlp": 1.0196197, "epoch": 0.7295136175073649, "flos": 22453330008000.0, "grad_norm": 1.7804220293476969, "language_loss": 0.79505908, "learning_rate": 7.1947075203832e-07, "loss": 0.81691122, "num_input_tokens_seen": 130411970, "step": 6067, "time_per_iteration": 3.2379963397979736 }, { "auxiliary_loss_clip": 0.01064609, "auxiliary_loss_mlp": 0.01000101, "balance_loss_clip": 1.00839639, "balance_loss_mlp": 0.99876636, "epoch": 0.7296338603980039, "flos": 56125516275360.0, "grad_norm": 0.8609403962279333, "language_loss": 0.60178143, "learning_rate": 7.188724786680049e-07, "loss": 0.62242854, "num_input_tokens_seen": 130472440, "step": 6068, "time_per_iteration": 3.0710959434509277 }, { "auxiliary_loss_clip": 0.01144048, "auxiliary_loss_mlp": 0.01027576, "balance_loss_clip": 1.04537666, "balance_loss_mlp": 1.02056384, "epoch": 0.7297541032886431, "flos": 25228291662720.0, "grad_norm": 1.6829125216167058, "language_loss": 0.75622976, "learning_rate": 7.182743996437162e-07, "loss": 0.777946, "num_input_tokens_seen": 130491975, "step": 6069, "time_per_iteration": 4.091414451599121 }, { "auxiliary_loss_clip": 0.01137966, "auxiliary_loss_mlp": 0.01027007, "balance_loss_clip": 1.04509878, "balance_loss_mlp": 1.01977122, "epoch": 0.7298743461792822, "flos": 26467604666400.0, "grad_norm": 2.0385248002510434, "language_loss": 0.6873529, "learning_rate": 7.176765150561819e-07, "loss": 0.70900261, "num_input_tokens_seen": 130510580, "step": 6070, "time_per_iteration": 2.57285213470459 }, { "auxiliary_loss_clip": 0.0117122, "auxiliary_loss_mlp": 0.01023276, "balance_loss_clip": 1.04774952, "balance_loss_mlp": 1.01547718, "epoch": 0.7299945890699212, "flos": 19569056292000.0, "grad_norm": 2.2144137693049433, "language_loss": 0.79912579, "learning_rate": 7.170788249961002e-07, "loss": 0.82107079, "num_input_tokens_seen": 130529090, "step": 6071, "time_per_iteration": 2.4665114879608154 }, { "auxiliary_loss_clip": 0.01170692, "auxiliary_loss_mlp": 0.01022223, "balance_loss_clip": 1.04968977, "balance_loss_mlp": 1.01548505, "epoch": 0.7301148319605604, "flos": 22928967095040.0, "grad_norm": 1.9708826961784673, "language_loss": 0.88224834, "learning_rate": 7.164813295541418e-07, "loss": 0.90417749, "num_input_tokens_seen": 130548655, "step": 6072, "time_per_iteration": 2.465200662612915 }, { "auxiliary_loss_clip": 0.01145285, "auxiliary_loss_mlp": 0.01029106, "balance_loss_clip": 1.04486787, "balance_loss_mlp": 1.02194798, "epoch": 0.7302350748511994, "flos": 25369707097920.0, "grad_norm": 1.922955561447098, "language_loss": 0.70156521, "learning_rate": 7.15884028820944e-07, "loss": 0.72330916, "num_input_tokens_seen": 130567710, "step": 6073, "time_per_iteration": 2.546083688735962 }, { "auxiliary_loss_clip": 0.011246, "auxiliary_loss_mlp": 0.0102581, "balance_loss_clip": 1.04265785, "balance_loss_mlp": 1.01894021, "epoch": 0.7303553177418385, "flos": 27819174922080.0, "grad_norm": 2.001638570169012, "language_loss": 0.60233748, "learning_rate": 7.152869228871185e-07, "loss": 0.62384152, "num_input_tokens_seen": 130590195, "step": 6074, "time_per_iteration": 2.586576223373413 }, { "auxiliary_loss_clip": 0.01139765, "auxiliary_loss_mlp": 0.01027742, "balance_loss_clip": 1.04580092, "balance_loss_mlp": 1.02054739, "epoch": 0.7304755606324776, "flos": 24426514239840.0, "grad_norm": 1.9475332979645685, "language_loss": 0.72215283, "learning_rate": 7.146900118432457e-07, "loss": 0.74382782, "num_input_tokens_seen": 130609940, "step": 6075, "time_per_iteration": 2.5558886528015137 }, { "auxiliary_loss_clip": 0.01082614, "auxiliary_loss_mlp": 0.01025224, "balance_loss_clip": 1.03592515, "balance_loss_mlp": 1.01857781, "epoch": 0.7305958035231167, "flos": 23840487582720.0, "grad_norm": 1.648199297502183, "language_loss": 0.86037678, "learning_rate": 7.140932957798753e-07, "loss": 0.88145512, "num_input_tokens_seen": 130628380, "step": 6076, "time_per_iteration": 2.714519739151001 }, { "auxiliary_loss_clip": 0.01146076, "auxiliary_loss_mlp": 0.01027006, "balance_loss_clip": 1.04549873, "balance_loss_mlp": 1.02020526, "epoch": 0.7307160464137558, "flos": 16726942698720.0, "grad_norm": 8.121502354362821, "language_loss": 0.71140528, "learning_rate": 7.134967747875309e-07, "loss": 0.73313618, "num_input_tokens_seen": 130646590, "step": 6077, "time_per_iteration": 2.4932825565338135 }, { "auxiliary_loss_clip": 0.01151433, "auxiliary_loss_mlp": 0.01025304, "balance_loss_clip": 1.04566991, "balance_loss_mlp": 1.01835084, "epoch": 0.7308362893043949, "flos": 21798283730400.0, "grad_norm": 2.2370155929902356, "language_loss": 0.81731009, "learning_rate": 7.129004489567014e-07, "loss": 0.83907747, "num_input_tokens_seen": 130664070, "step": 6078, "time_per_iteration": 2.4938578605651855 }, { "auxiliary_loss_clip": 0.01132534, "auxiliary_loss_mlp": 0.01021799, "balance_loss_clip": 1.04493284, "balance_loss_mlp": 1.01458097, "epoch": 0.730956532195034, "flos": 10707380434560.0, "grad_norm": 3.236137868909914, "language_loss": 0.77989322, "learning_rate": 7.123043183778512e-07, "loss": 0.80143654, "num_input_tokens_seen": 130681400, "step": 6079, "time_per_iteration": 3.2621538639068604 }, { "auxiliary_loss_clip": 0.01136458, "auxiliary_loss_mlp": 0.01029398, "balance_loss_clip": 1.04870009, "balance_loss_mlp": 1.02256405, "epoch": 0.731076775085673, "flos": 19791990449760.0, "grad_norm": 1.505269738959121, "language_loss": 0.65145999, "learning_rate": 7.117083831414114e-07, "loss": 0.67311853, "num_input_tokens_seen": 130700675, "step": 6080, "time_per_iteration": 2.7006704807281494 }, { "auxiliary_loss_clip": 0.01169139, "auxiliary_loss_mlp": 0.01031366, "balance_loss_clip": 1.04766798, "balance_loss_mlp": 1.02494371, "epoch": 0.7311970179763122, "flos": 20447036727360.0, "grad_norm": 1.8110437865290705, "language_loss": 0.6968441, "learning_rate": 7.11112643337787e-07, "loss": 0.71884918, "num_input_tokens_seen": 130719720, "step": 6081, "time_per_iteration": 2.555036783218384 }, { "auxiliary_loss_clip": 0.01147739, "auxiliary_loss_mlp": 0.01029683, "balance_loss_clip": 1.05055022, "balance_loss_mlp": 1.02253389, "epoch": 0.7313172608669513, "flos": 18513821683680.0, "grad_norm": 2.189172405404264, "language_loss": 0.76288784, "learning_rate": 7.10517099057349e-07, "loss": 0.78466201, "num_input_tokens_seen": 130736670, "step": 6082, "time_per_iteration": 2.565152168273926 }, { "auxiliary_loss_clip": 0.01145243, "auxiliary_loss_mlp": 0.01025198, "balance_loss_clip": 1.04691243, "balance_loss_mlp": 1.01768756, "epoch": 0.7314375037575903, "flos": 16180741561920.0, "grad_norm": 2.213965347695896, "language_loss": 0.61178738, "learning_rate": 7.099217503904411e-07, "loss": 0.63349181, "num_input_tokens_seen": 130754525, "step": 6083, "time_per_iteration": 2.473264217376709 }, { "auxiliary_loss_clip": 0.0114638, "auxiliary_loss_mlp": 0.01026423, "balance_loss_clip": 1.04660022, "balance_loss_mlp": 1.01992857, "epoch": 0.7315577466482295, "flos": 17967943799520.0, "grad_norm": 1.8214921538027713, "language_loss": 0.90053737, "learning_rate": 7.093265974273788e-07, "loss": 0.92226535, "num_input_tokens_seen": 130772420, "step": 6084, "time_per_iteration": 2.529799699783325 }, { "auxiliary_loss_clip": 0.01157924, "auxiliary_loss_mlp": 0.01022767, "balance_loss_clip": 1.04590464, "balance_loss_mlp": 1.01635313, "epoch": 0.7316779895388685, "flos": 18405443463360.0, "grad_norm": 1.7726352095696452, "language_loss": 0.71952581, "learning_rate": 7.087316402584447e-07, "loss": 0.74133271, "num_input_tokens_seen": 130791245, "step": 6085, "time_per_iteration": 2.5046215057373047 }, { "auxiliary_loss_clip": 0.01168361, "auxiliary_loss_mlp": 0.010283, "balance_loss_clip": 1.04691124, "balance_loss_mlp": 1.02133191, "epoch": 0.7317982324295076, "flos": 17928297864000.0, "grad_norm": 1.7616195049976413, "language_loss": 0.86381775, "learning_rate": 7.081368789738953e-07, "loss": 0.88578439, "num_input_tokens_seen": 130808445, "step": 6086, "time_per_iteration": 2.456843852996826 }, { "auxiliary_loss_clip": 0.01133991, "auxiliary_loss_mlp": 0.01023369, "balance_loss_clip": 1.04068089, "balance_loss_mlp": 1.01663065, "epoch": 0.7319184753201466, "flos": 27229843904640.0, "grad_norm": 1.9018510226368979, "language_loss": 0.78199762, "learning_rate": 7.075423136639537e-07, "loss": 0.80357122, "num_input_tokens_seen": 130827700, "step": 6087, "time_per_iteration": 2.5687613487243652 }, { "auxiliary_loss_clip": 0.01124766, "auxiliary_loss_mlp": 0.01026482, "balance_loss_clip": 1.04359651, "balance_loss_mlp": 1.01912665, "epoch": 0.7320387182107858, "flos": 37448549395680.0, "grad_norm": 1.8496300708435174, "language_loss": 0.74667132, "learning_rate": 7.069479444188149e-07, "loss": 0.76818383, "num_input_tokens_seen": 130848290, "step": 6088, "time_per_iteration": 2.7486791610717773 }, { "auxiliary_loss_clip": 0.01133546, "auxiliary_loss_mlp": 0.01029829, "balance_loss_clip": 1.04395592, "balance_loss_mlp": 1.02287054, "epoch": 0.7321589611014249, "flos": 17859026824800.0, "grad_norm": 1.9332391070534334, "language_loss": 0.81999171, "learning_rate": 7.063537713286453e-07, "loss": 0.84162545, "num_input_tokens_seen": 130865970, "step": 6089, "time_per_iteration": 2.514209032058716 }, { "auxiliary_loss_clip": 0.01149348, "auxiliary_loss_mlp": 0.01027408, "balance_loss_clip": 1.04535139, "balance_loss_mlp": 1.02005839, "epoch": 0.7322792039920639, "flos": 26100597218400.0, "grad_norm": 1.88015976278763, "language_loss": 0.80806947, "learning_rate": 7.057597944835803e-07, "loss": 0.82983708, "num_input_tokens_seen": 130885245, "step": 6090, "time_per_iteration": 2.574435234069824 }, { "auxiliary_loss_clip": 0.01133192, "auxiliary_loss_mlp": 0.01027402, "balance_loss_clip": 1.04214454, "balance_loss_mlp": 1.02069616, "epoch": 0.7323994468827031, "flos": 25369096509600.0, "grad_norm": 1.6464666670015469, "language_loss": 0.74727398, "learning_rate": 7.051660139737253e-07, "loss": 0.76887989, "num_input_tokens_seen": 130903465, "step": 6091, "time_per_iteration": 2.6268246173858643 }, { "auxiliary_loss_clip": 0.01156063, "auxiliary_loss_mlp": 0.00762095, "balance_loss_clip": 1.05016184, "balance_loss_mlp": 1.00056243, "epoch": 0.7325196897733421, "flos": 26907079763040.0, "grad_norm": 1.8931353886533295, "language_loss": 0.77136785, "learning_rate": 7.045724298891565e-07, "loss": 0.7905494, "num_input_tokens_seen": 130922935, "step": 6092, "time_per_iteration": 2.5423145294189453 }, { "auxiliary_loss_clip": 0.01156375, "auxiliary_loss_mlp": 0.01023917, "balance_loss_clip": 1.04883587, "balance_loss_mlp": 1.01679444, "epoch": 0.7326399326639812, "flos": 25775785810080.0, "grad_norm": 2.033373936358249, "language_loss": 0.69220936, "learning_rate": 7.039790423199192e-07, "loss": 0.71401227, "num_input_tokens_seen": 130942575, "step": 6093, "time_per_iteration": 3.305753707885742 }, { "auxiliary_loss_clip": 0.01147777, "auxiliary_loss_mlp": 0.01027062, "balance_loss_clip": 1.04711795, "balance_loss_mlp": 1.01937604, "epoch": 0.7327601755546204, "flos": 21032273211360.0, "grad_norm": 2.0604842680849274, "language_loss": 0.77878028, "learning_rate": 7.033858513560322e-07, "loss": 0.80052865, "num_input_tokens_seen": 130958870, "step": 6094, "time_per_iteration": 2.5336480140686035 }, { "auxiliary_loss_clip": 0.01157595, "auxiliary_loss_mlp": 0.01028764, "balance_loss_clip": 1.04746819, "balance_loss_mlp": 1.02192163, "epoch": 0.7328804184452594, "flos": 16289227533120.0, "grad_norm": 2.2734589123399735, "language_loss": 0.76482749, "learning_rate": 7.027928570874794e-07, "loss": 0.78669107, "num_input_tokens_seen": 130977060, "step": 6095, "time_per_iteration": 3.310020923614502 }, { "auxiliary_loss_clip": 0.0116859, "auxiliary_loss_mlp": 0.01026209, "balance_loss_clip": 1.04763126, "balance_loss_mlp": 1.01897264, "epoch": 0.7330006613358985, "flos": 17858236651680.0, "grad_norm": 2.045682501056403, "language_loss": 0.85565829, "learning_rate": 7.022000596042194e-07, "loss": 0.87760621, "num_input_tokens_seen": 130994160, "step": 6096, "time_per_iteration": 3.1507694721221924 }, { "auxiliary_loss_clip": 0.01129957, "auxiliary_loss_mlp": 0.01021505, "balance_loss_clip": 1.0409286, "balance_loss_mlp": 1.01460552, "epoch": 0.7331209042265376, "flos": 22492078019520.0, "grad_norm": 1.9922122922295968, "language_loss": 0.81775773, "learning_rate": 7.016074589961784e-07, "loss": 0.83927232, "num_input_tokens_seen": 131012725, "step": 6097, "time_per_iteration": 2.5699150562286377 }, { "auxiliary_loss_clip": 0.01140186, "auxiliary_loss_mlp": 0.01027018, "balance_loss_clip": 1.04688823, "balance_loss_mlp": 1.02020204, "epoch": 0.7332411471171767, "flos": 33072762584160.0, "grad_norm": 1.7295576297009576, "language_loss": 0.67064166, "learning_rate": 7.01015055353253e-07, "loss": 0.69231367, "num_input_tokens_seen": 131035150, "step": 6098, "time_per_iteration": 2.639752149581909 }, { "auxiliary_loss_clip": 0.01105608, "auxiliary_loss_mlp": 0.01030765, "balance_loss_clip": 1.04299045, "balance_loss_mlp": 1.02329016, "epoch": 0.7333613900078157, "flos": 22743021017760.0, "grad_norm": 1.6684517253191755, "language_loss": 0.77965689, "learning_rate": 7.004228487653123e-07, "loss": 0.80102062, "num_input_tokens_seen": 131055955, "step": 6099, "time_per_iteration": 2.6110005378723145 }, { "auxiliary_loss_clip": 0.01123678, "auxiliary_loss_mlp": 0.01024028, "balance_loss_clip": 1.03896868, "balance_loss_mlp": 1.01731062, "epoch": 0.7334816328984549, "flos": 22346137047360.0, "grad_norm": 1.8815838986827884, "language_loss": 0.78103578, "learning_rate": 6.998308393221906e-07, "loss": 0.80251276, "num_input_tokens_seen": 131074360, "step": 6100, "time_per_iteration": 2.567695379257202 }, { "auxiliary_loss_clip": 0.01131447, "auxiliary_loss_mlp": 0.01025346, "balance_loss_clip": 1.04563403, "balance_loss_mlp": 1.01782048, "epoch": 0.733601875789094, "flos": 20736153065760.0, "grad_norm": 2.610791958905453, "language_loss": 0.71099645, "learning_rate": 6.992390271136977e-07, "loss": 0.73256439, "num_input_tokens_seen": 131090070, "step": 6101, "time_per_iteration": 2.5550410747528076 }, { "auxiliary_loss_clip": 0.01148335, "auxiliary_loss_mlp": 0.01025408, "balance_loss_clip": 1.04516006, "balance_loss_mlp": 1.01796043, "epoch": 0.733722118679733, "flos": 22564366083360.0, "grad_norm": 1.6917885431988882, "language_loss": 0.85518569, "learning_rate": 6.986474122296094e-07, "loss": 0.87692308, "num_input_tokens_seen": 131109185, "step": 6102, "time_per_iteration": 2.5082173347473145 }, { "auxiliary_loss_clip": 0.01173282, "auxiliary_loss_mlp": 0.01029009, "balance_loss_clip": 1.04906571, "balance_loss_mlp": 1.02112913, "epoch": 0.7338423615703722, "flos": 20084195646720.0, "grad_norm": 1.8644377338378173, "language_loss": 0.7241841, "learning_rate": 6.980559947596751e-07, "loss": 0.746207, "num_input_tokens_seen": 131127725, "step": 6103, "time_per_iteration": 2.4693541526794434 }, { "auxiliary_loss_clip": 0.01115205, "auxiliary_loss_mlp": 0.010292, "balance_loss_clip": 1.04389584, "balance_loss_mlp": 1.02266431, "epoch": 0.7339626044610112, "flos": 21687678658560.0, "grad_norm": 2.343296851367662, "language_loss": 0.75540423, "learning_rate": 6.974647747936109e-07, "loss": 0.7768482, "num_input_tokens_seen": 131146110, "step": 6104, "time_per_iteration": 2.60451078414917 }, { "auxiliary_loss_clip": 0.01168692, "auxiliary_loss_mlp": 0.00762276, "balance_loss_clip": 1.04812443, "balance_loss_mlp": 1.00053608, "epoch": 0.7340828473516503, "flos": 15268251316320.0, "grad_norm": 2.0827519671161703, "language_loss": 0.82434142, "learning_rate": 6.968737524211039e-07, "loss": 0.84365118, "num_input_tokens_seen": 131162920, "step": 6105, "time_per_iteration": 3.195301055908203 }, { "auxiliary_loss_clip": 0.01156459, "auxiliary_loss_mlp": 0.01028631, "balance_loss_clip": 1.04779589, "balance_loss_mlp": 1.02167535, "epoch": 0.7342030902422895, "flos": 22930116437760.0, "grad_norm": 2.946510575465181, "language_loss": 0.79987133, "learning_rate": 6.962829277318132e-07, "loss": 0.82172221, "num_input_tokens_seen": 131182515, "step": 6106, "time_per_iteration": 2.495668411254883 }, { "auxiliary_loss_clip": 0.0115843, "auxiliary_loss_mlp": 0.01022734, "balance_loss_clip": 1.05001378, "balance_loss_mlp": 1.01597786, "epoch": 0.7343233331329285, "flos": 25847894289120.0, "grad_norm": 1.7773659166054525, "language_loss": 0.83732724, "learning_rate": 6.956923008153652e-07, "loss": 0.85913891, "num_input_tokens_seen": 131202280, "step": 6107, "time_per_iteration": 2.514636516571045 }, { "auxiliary_loss_clip": 0.01156326, "auxiliary_loss_mlp": 0.01023828, "balance_loss_clip": 1.04526424, "balance_loss_mlp": 1.01719713, "epoch": 0.7344435760235676, "flos": 18478988620800.0, "grad_norm": 2.032147532236801, "language_loss": 0.84528768, "learning_rate": 6.951018717613593e-07, "loss": 0.86708921, "num_input_tokens_seen": 131221295, "step": 6108, "time_per_iteration": 2.488630771636963 }, { "auxiliary_loss_clip": 0.01157621, "auxiliary_loss_mlp": 0.01024849, "balance_loss_clip": 1.04979861, "balance_loss_mlp": 1.01809573, "epoch": 0.7345638189142067, "flos": 17640043532640.0, "grad_norm": 1.7347576772257145, "language_loss": 0.78208578, "learning_rate": 6.945116406593614e-07, "loss": 0.80391049, "num_input_tokens_seen": 131240150, "step": 6109, "time_per_iteration": 2.4604287147521973 }, { "auxiliary_loss_clip": 0.01118854, "auxiliary_loss_mlp": 0.01030505, "balance_loss_clip": 1.04537833, "balance_loss_mlp": 1.02332878, "epoch": 0.7346840618048458, "flos": 20260228643040.0, "grad_norm": 2.8609188307043185, "language_loss": 0.742643, "learning_rate": 6.939216075989089e-07, "loss": 0.76413667, "num_input_tokens_seen": 131258080, "step": 6110, "time_per_iteration": 2.5797674655914307 }, { "auxiliary_loss_clip": 0.01139906, "auxiliary_loss_mlp": 0.01019162, "balance_loss_clip": 1.0438571, "balance_loss_mlp": 1.01249814, "epoch": 0.7348043046954849, "flos": 29023187942400.0, "grad_norm": 2.190527598233124, "language_loss": 0.65965587, "learning_rate": 6.933317726695109e-07, "loss": 0.68124652, "num_input_tokens_seen": 131279310, "step": 6111, "time_per_iteration": 2.5719683170318604 }, { "auxiliary_loss_clip": 0.0112815, "auxiliary_loss_mlp": 0.01029429, "balance_loss_clip": 1.04854608, "balance_loss_mlp": 1.02259576, "epoch": 0.734924547586124, "flos": 17931207137760.0, "grad_norm": 2.9685366907039685, "language_loss": 0.79902917, "learning_rate": 6.92742135960644e-07, "loss": 0.82060492, "num_input_tokens_seen": 131297010, "step": 6112, "time_per_iteration": 2.5129919052124023 }, { "auxiliary_loss_clip": 0.01057424, "auxiliary_loss_mlp": 0.00999904, "balance_loss_clip": 1.00998449, "balance_loss_mlp": 0.99866432, "epoch": 0.7350447904767631, "flos": 63588332017440.0, "grad_norm": 0.8113953609979389, "language_loss": 0.5568881, "learning_rate": 6.921526975617556e-07, "loss": 0.57746136, "num_input_tokens_seen": 131356470, "step": 6113, "time_per_iteration": 3.1287760734558105 }, { "auxiliary_loss_clip": 0.01141718, "auxiliary_loss_mlp": 0.01027036, "balance_loss_clip": 1.04456151, "balance_loss_mlp": 1.01984179, "epoch": 0.7351650333674021, "flos": 21580018777440.0, "grad_norm": 1.851485536954995, "language_loss": 0.75438046, "learning_rate": 6.915634575622631e-07, "loss": 0.77606797, "num_input_tokens_seen": 131374985, "step": 6114, "time_per_iteration": 2.6032378673553467 }, { "auxiliary_loss_clip": 0.01168763, "auxiliary_loss_mlp": 0.01025286, "balance_loss_clip": 1.04724312, "balance_loss_mlp": 1.01843452, "epoch": 0.7352852762580413, "flos": 18186352420320.0, "grad_norm": 2.2938076663982967, "language_loss": 0.71045232, "learning_rate": 6.909744160515532e-07, "loss": 0.73239279, "num_input_tokens_seen": 131393125, "step": 6115, "time_per_iteration": 2.4265027046203613 }, { "auxiliary_loss_clip": 0.01140587, "auxiliary_loss_mlp": 0.01023546, "balance_loss_clip": 1.04585004, "balance_loss_mlp": 1.01658463, "epoch": 0.7354055191486804, "flos": 38910078217920.0, "grad_norm": 1.781893259326624, "language_loss": 0.6918776, "learning_rate": 6.903855731189849e-07, "loss": 0.71351892, "num_input_tokens_seen": 131415760, "step": 6116, "time_per_iteration": 2.6788740158081055 }, { "auxiliary_loss_clip": 0.01151612, "auxiliary_loss_mlp": 0.0103058, "balance_loss_clip": 1.0477016, "balance_loss_mlp": 1.02306974, "epoch": 0.7355257620393194, "flos": 16289981789280.0, "grad_norm": 2.1565605480573447, "language_loss": 0.81988525, "learning_rate": 6.897969288538825e-07, "loss": 0.84170717, "num_input_tokens_seen": 131433705, "step": 6117, "time_per_iteration": 2.4872608184814453 }, { "auxiliary_loss_clip": 0.01140326, "auxiliary_loss_mlp": 0.01023621, "balance_loss_clip": 1.04664254, "balance_loss_mlp": 1.01687694, "epoch": 0.7356460049299585, "flos": 18114243941280.0, "grad_norm": 1.7518383152178951, "language_loss": 0.81191623, "learning_rate": 6.892084833455452e-07, "loss": 0.8335557, "num_input_tokens_seen": 131453275, "step": 6118, "time_per_iteration": 2.517573833465576 }, { "auxiliary_loss_clip": 0.01153668, "auxiliary_loss_mlp": 0.01023912, "balance_loss_clip": 1.04671025, "balance_loss_mlp": 1.01691151, "epoch": 0.7357662478205976, "flos": 21325196747520.0, "grad_norm": 1.4372398223159575, "language_loss": 0.83754081, "learning_rate": 6.886202366832384e-07, "loss": 0.85931659, "num_input_tokens_seen": 131474960, "step": 6119, "time_per_iteration": 3.256742477416992 }, { "auxiliary_loss_clip": 0.0111628, "auxiliary_loss_mlp": 0.0102715, "balance_loss_clip": 1.04732108, "balance_loss_mlp": 1.02014387, "epoch": 0.7358864907112367, "flos": 14246844096000.0, "grad_norm": 1.9502765963187088, "language_loss": 0.73515332, "learning_rate": 6.880321889561987e-07, "loss": 0.75658762, "num_input_tokens_seen": 131492935, "step": 6120, "time_per_iteration": 2.5911567211151123 }, { "auxiliary_loss_clip": 0.01121349, "auxiliary_loss_mlp": 0.0102546, "balance_loss_clip": 1.04291368, "balance_loss_mlp": 1.01734459, "epoch": 0.7360067336018757, "flos": 22309687721280.0, "grad_norm": 1.9817350883335754, "language_loss": 0.653364, "learning_rate": 6.874443402536338e-07, "loss": 0.67483211, "num_input_tokens_seen": 131512025, "step": 6121, "time_per_iteration": 3.356264114379883 }, { "auxiliary_loss_clip": 0.01145334, "auxiliary_loss_mlp": 0.01024673, "balance_loss_clip": 1.0456475, "balance_loss_mlp": 1.01752591, "epoch": 0.7361269764925149, "flos": 25554611583360.0, "grad_norm": 2.129105125183848, "language_loss": 0.80335307, "learning_rate": 6.868566906647177e-07, "loss": 0.8250531, "num_input_tokens_seen": 131532975, "step": 6122, "time_per_iteration": 3.2600483894348145 }, { "auxiliary_loss_clip": 0.01157844, "auxiliary_loss_mlp": 0.01028682, "balance_loss_clip": 1.04742575, "balance_loss_mlp": 1.02175939, "epoch": 0.736247219383154, "flos": 20376508594560.0, "grad_norm": 1.8660546382140912, "language_loss": 0.83080947, "learning_rate": 6.862692402785984e-07, "loss": 0.85267472, "num_input_tokens_seen": 131553225, "step": 6123, "time_per_iteration": 2.5162765979766846 }, { "auxiliary_loss_clip": 0.01041756, "auxiliary_loss_mlp": 0.01005585, "balance_loss_clip": 1.02195907, "balance_loss_mlp": 1.00432706, "epoch": 0.736367462273793, "flos": 70339538658240.0, "grad_norm": 0.6789597240716547, "language_loss": 0.49655509, "learning_rate": 6.856819891843899e-07, "loss": 0.51702851, "num_input_tokens_seen": 131617930, "step": 6124, "time_per_iteration": 3.2369823455810547 }, { "auxiliary_loss_clip": 0.01104221, "auxiliary_loss_mlp": 0.01028478, "balance_loss_clip": 1.04473364, "balance_loss_mlp": 1.0211885, "epoch": 0.7364877051644322, "flos": 22412714314560.0, "grad_norm": 1.859469558450278, "language_loss": 0.71546257, "learning_rate": 6.8509493747118e-07, "loss": 0.73678952, "num_input_tokens_seen": 131636740, "step": 6125, "time_per_iteration": 2.6222479343414307 }, { "auxiliary_loss_clip": 0.01174179, "auxiliary_loss_mlp": 0.01020161, "balance_loss_clip": 1.05131078, "balance_loss_mlp": 1.01319325, "epoch": 0.7366079480550712, "flos": 12130268996160.0, "grad_norm": 2.37912931549207, "language_loss": 0.88237298, "learning_rate": 6.845080852280221e-07, "loss": 0.90431637, "num_input_tokens_seen": 131653810, "step": 6126, "time_per_iteration": 2.4416515827178955 }, { "auxiliary_loss_clip": 0.01130177, "auxiliary_loss_mlp": 0.01027574, "balance_loss_clip": 1.04412901, "balance_loss_mlp": 1.02131879, "epoch": 0.7367281909457103, "flos": 15049339858080.0, "grad_norm": 1.6695931798969235, "language_loss": 0.74175543, "learning_rate": 6.839214325439409e-07, "loss": 0.76333296, "num_input_tokens_seen": 131671505, "step": 6127, "time_per_iteration": 2.56913423538208 }, { "auxiliary_loss_clip": 0.01139375, "auxiliary_loss_mlp": 0.01024483, "balance_loss_clip": 1.0492574, "balance_loss_mlp": 1.01775372, "epoch": 0.7368484338363495, "flos": 23510755550880.0, "grad_norm": 1.5995133035780542, "language_loss": 0.7167896, "learning_rate": 6.833349795079327e-07, "loss": 0.73842812, "num_input_tokens_seen": 131690615, "step": 6128, "time_per_iteration": 2.5385024547576904 }, { "auxiliary_loss_clip": 0.01127379, "auxiliary_loss_mlp": 0.01025037, "balance_loss_clip": 1.0457027, "balance_loss_mlp": 1.01790583, "epoch": 0.7369686767269885, "flos": 27417837248640.0, "grad_norm": 2.2516967679603876, "language_loss": 0.6880523, "learning_rate": 6.827487262089613e-07, "loss": 0.70957649, "num_input_tokens_seen": 131711120, "step": 6129, "time_per_iteration": 2.626296043395996 }, { "auxiliary_loss_clip": 0.0103962, "auxiliary_loss_mlp": 0.01001593, "balance_loss_clip": 1.00831187, "balance_loss_mlp": 1.00028789, "epoch": 0.7370889196176276, "flos": 70293355836000.0, "grad_norm": 1.0440792020504157, "language_loss": 0.56860983, "learning_rate": 6.821626727359606e-07, "loss": 0.58902192, "num_input_tokens_seen": 131776680, "step": 6130, "time_per_iteration": 3.200995445251465 }, { "auxiliary_loss_clip": 0.0114557, "auxiliary_loss_mlp": 0.01024782, "balance_loss_clip": 1.04964375, "balance_loss_mlp": 1.01714635, "epoch": 0.7372091625082667, "flos": 18040842451680.0, "grad_norm": 2.3941120775697966, "language_loss": 0.77105868, "learning_rate": 6.815768191778348e-07, "loss": 0.79276222, "num_input_tokens_seen": 131794760, "step": 6131, "time_per_iteration": 3.2357120513916016 }, { "auxiliary_loss_clip": 0.0115138, "auxiliary_loss_mlp": 0.01029545, "balance_loss_clip": 1.04555774, "balance_loss_mlp": 1.02201653, "epoch": 0.7373294053989058, "flos": 33726336266400.0, "grad_norm": 1.6881478846808007, "language_loss": 0.73085141, "learning_rate": 6.809911656234569e-07, "loss": 0.75266069, "num_input_tokens_seen": 131816735, "step": 6132, "time_per_iteration": 2.619645118713379 }, { "auxiliary_loss_clip": 0.01130695, "auxiliary_loss_mlp": 0.01023013, "balance_loss_clip": 1.04147303, "balance_loss_mlp": 1.01673365, "epoch": 0.7374496482895448, "flos": 21506329952160.0, "grad_norm": 2.0599030239389933, "language_loss": 0.78284305, "learning_rate": 6.804057121616707e-07, "loss": 0.80438012, "num_input_tokens_seen": 131834940, "step": 6133, "time_per_iteration": 2.5552282333374023 }, { "auxiliary_loss_clip": 0.01159284, "auxiliary_loss_mlp": 0.01023708, "balance_loss_clip": 1.04779434, "balance_loss_mlp": 1.0164268, "epoch": 0.737569891180184, "flos": 24936912555840.0, "grad_norm": 2.176878680872135, "language_loss": 0.72150695, "learning_rate": 6.798204588812888e-07, "loss": 0.7433368, "num_input_tokens_seen": 131854355, "step": 6134, "time_per_iteration": 2.526707410812378 }, { "auxiliary_loss_clip": 0.01090767, "auxiliary_loss_mlp": 0.00762295, "balance_loss_clip": 1.03896809, "balance_loss_mlp": 1.00055242, "epoch": 0.7376901340708231, "flos": 20664547424160.0, "grad_norm": 1.674556033603089, "language_loss": 0.75440371, "learning_rate": 6.792354058710937e-07, "loss": 0.77293432, "num_input_tokens_seen": 131871825, "step": 6135, "time_per_iteration": 2.612382411956787 }, { "auxiliary_loss_clip": 0.01165784, "auxiliary_loss_mlp": 0.01020643, "balance_loss_clip": 1.04852927, "balance_loss_mlp": 1.01437855, "epoch": 0.7378103769614621, "flos": 23805798187680.0, "grad_norm": 1.7947274261892914, "language_loss": 0.64915007, "learning_rate": 6.786505532198374e-07, "loss": 0.67101431, "num_input_tokens_seen": 131890770, "step": 6136, "time_per_iteration": 2.5257139205932617 }, { "auxiliary_loss_clip": 0.0117148, "auxiliary_loss_mlp": 0.01023087, "balance_loss_clip": 1.04920733, "balance_loss_mlp": 1.01591647, "epoch": 0.7379306198521013, "flos": 22237220072640.0, "grad_norm": 2.3305082610567016, "language_loss": 0.85246509, "learning_rate": 6.780659010162411e-07, "loss": 0.87441081, "num_input_tokens_seen": 131909720, "step": 6137, "time_per_iteration": 2.4544200897216797 }, { "auxiliary_loss_clip": 0.01131437, "auxiliary_loss_mlp": 0.01021917, "balance_loss_clip": 1.04576921, "balance_loss_mlp": 1.01529789, "epoch": 0.7380508627427403, "flos": 14903111550240.0, "grad_norm": 1.649383587001359, "language_loss": 0.83113486, "learning_rate": 6.774814493489975e-07, "loss": 0.8526684, "num_input_tokens_seen": 131927395, "step": 6138, "time_per_iteration": 2.5202181339263916 }, { "auxiliary_loss_clip": 0.01152811, "auxiliary_loss_mlp": 0.01023221, "balance_loss_clip": 1.04718018, "balance_loss_mlp": 1.01664054, "epoch": 0.7381711056333794, "flos": 21685846893600.0, "grad_norm": 1.7074884962218566, "language_loss": 0.6577636, "learning_rate": 6.768971983067655e-07, "loss": 0.67952394, "num_input_tokens_seen": 131947725, "step": 6139, "time_per_iteration": 2.4889655113220215 }, { "auxiliary_loss_clip": 0.01065262, "auxiliary_loss_mlp": 0.00999635, "balance_loss_clip": 1.0086441, "balance_loss_mlp": 0.9984194, "epoch": 0.7382913485240186, "flos": 52404272904000.0, "grad_norm": 1.0106490223299296, "language_loss": 0.67865932, "learning_rate": 6.763131479781772e-07, "loss": 0.69930828, "num_input_tokens_seen": 131997485, "step": 6140, "time_per_iteration": 2.884490728378296 }, { "auxiliary_loss_clip": 0.0113558, "auxiliary_loss_mlp": 0.01021259, "balance_loss_clip": 1.04636431, "balance_loss_mlp": 1.01479161, "epoch": 0.7384115914146576, "flos": 21798822484800.0, "grad_norm": 2.2455365876315674, "language_loss": 0.76495111, "learning_rate": 6.757292984518316e-07, "loss": 0.78651953, "num_input_tokens_seen": 132016885, "step": 6141, "time_per_iteration": 2.496567487716675 }, { "auxiliary_loss_clip": 0.0105616, "auxiliary_loss_mlp": 0.0100066, "balance_loss_clip": 1.0086875, "balance_loss_mlp": 0.9993664, "epoch": 0.7385318343052967, "flos": 61494342297120.0, "grad_norm": 0.7433352914613717, "language_loss": 0.5648765, "learning_rate": 6.751456498162981e-07, "loss": 0.58544469, "num_input_tokens_seen": 132075920, "step": 6142, "time_per_iteration": 2.979783296585083 }, { "auxiliary_loss_clip": 0.01153817, "auxiliary_loss_mlp": 0.01024459, "balance_loss_clip": 1.04382133, "balance_loss_mlp": 1.01780713, "epoch": 0.7386520771959358, "flos": 17013760351680.0, "grad_norm": 2.4274725342225847, "language_loss": 0.85589194, "learning_rate": 6.745622021601174e-07, "loss": 0.8776747, "num_input_tokens_seen": 132092945, "step": 6143, "time_per_iteration": 2.4429099559783936 }, { "auxiliary_loss_clip": 0.01130414, "auxiliary_loss_mlp": 0.01022478, "balance_loss_clip": 1.04422951, "balance_loss_mlp": 1.01553106, "epoch": 0.7387723200865749, "flos": 18770762814240.0, "grad_norm": 1.8832052876418703, "language_loss": 0.69386458, "learning_rate": 6.739789555717954e-07, "loss": 0.71539354, "num_input_tokens_seen": 132109920, "step": 6144, "time_per_iteration": 2.522799253463745 }, { "auxiliary_loss_clip": 0.01166897, "auxiliary_loss_mlp": 0.01023318, "balance_loss_clip": 1.0468564, "balance_loss_mlp": 1.01653183, "epoch": 0.738892562977214, "flos": 22525546237920.0, "grad_norm": 1.9459524655693383, "language_loss": 0.77331519, "learning_rate": 6.733959101398124e-07, "loss": 0.79521728, "num_input_tokens_seen": 132128050, "step": 6145, "time_per_iteration": 3.231771469116211 }, { "auxiliary_loss_clip": 0.01139694, "auxiliary_loss_mlp": 0.01023954, "balance_loss_clip": 1.04372323, "balance_loss_mlp": 1.01718247, "epoch": 0.7390128058678531, "flos": 21501481162560.0, "grad_norm": 1.5915664911863288, "language_loss": 0.81381333, "learning_rate": 6.728130659526143e-07, "loss": 0.83544981, "num_input_tokens_seen": 132145860, "step": 6146, "time_per_iteration": 2.5052623748779297 }, { "auxiliary_loss_clip": 0.01144335, "auxiliary_loss_mlp": 0.01027009, "balance_loss_clip": 1.04599345, "balance_loss_mlp": 1.02016044, "epoch": 0.7391330487584922, "flos": 25776180896640.0, "grad_norm": 2.292073776897378, "language_loss": 0.70842171, "learning_rate": 6.7223042309862e-07, "loss": 0.73013514, "num_input_tokens_seen": 132166060, "step": 6147, "time_per_iteration": 3.356375217437744 }, { "auxiliary_loss_clip": 0.01152314, "auxiliary_loss_mlp": 0.01026273, "balance_loss_clip": 1.04449594, "balance_loss_mlp": 1.0192306, "epoch": 0.7392532916491312, "flos": 28366740903360.0, "grad_norm": 1.9061068988940153, "language_loss": 0.73844588, "learning_rate": 6.716479816662144e-07, "loss": 0.76023173, "num_input_tokens_seen": 132187790, "step": 6148, "time_per_iteration": 3.268537759780884 }, { "auxiliary_loss_clip": 0.01144621, "auxiliary_loss_mlp": 0.01025007, "balance_loss_clip": 1.04315579, "balance_loss_mlp": 1.0181675, "epoch": 0.7393735345397703, "flos": 23585881054560.0, "grad_norm": 2.591943541962601, "language_loss": 0.72855937, "learning_rate": 6.710657417437531e-07, "loss": 0.75025558, "num_input_tokens_seen": 132207495, "step": 6149, "time_per_iteration": 2.5636696815490723 }, { "auxiliary_loss_clip": 0.01142157, "auxiliary_loss_mlp": 0.01023225, "balance_loss_clip": 1.04590487, "balance_loss_mlp": 1.01689541, "epoch": 0.7394937774304094, "flos": 19974775834560.0, "grad_norm": 2.3905902123468463, "language_loss": 0.80177683, "learning_rate": 6.704837034195628e-07, "loss": 0.82343066, "num_input_tokens_seen": 132225960, "step": 6150, "time_per_iteration": 2.494439125061035 }, { "auxiliary_loss_clip": 0.01148229, "auxiliary_loss_mlp": 0.01028352, "balance_loss_clip": 1.04562593, "balance_loss_mlp": 1.02124166, "epoch": 0.7396140203210485, "flos": 23478041588640.0, "grad_norm": 1.6286640315787162, "language_loss": 0.84458911, "learning_rate": 6.699018667819376e-07, "loss": 0.86635494, "num_input_tokens_seen": 132245360, "step": 6151, "time_per_iteration": 2.5283265113830566 }, { "auxiliary_loss_clip": 0.01150085, "auxiliary_loss_mlp": 0.01021758, "balance_loss_clip": 1.04403698, "balance_loss_mlp": 1.01400661, "epoch": 0.7397342632116876, "flos": 25555437673440.0, "grad_norm": 1.5662199225214524, "language_loss": 0.72915947, "learning_rate": 6.693202319191415e-07, "loss": 0.75087792, "num_input_tokens_seen": 132267095, "step": 6152, "time_per_iteration": 2.525683879852295 }, { "auxiliary_loss_clip": 0.01169721, "auxiliary_loss_mlp": 0.01027744, "balance_loss_clip": 1.05059731, "balance_loss_mlp": 1.02089548, "epoch": 0.7398545061023267, "flos": 24755025095040.0, "grad_norm": 1.940807051383751, "language_loss": 0.74697459, "learning_rate": 6.687387989194084e-07, "loss": 0.76894927, "num_input_tokens_seen": 132286610, "step": 6153, "time_per_iteration": 2.485213279724121 }, { "auxiliary_loss_clip": 0.01139023, "auxiliary_loss_mlp": 0.01022559, "balance_loss_clip": 1.04709411, "balance_loss_mlp": 1.01581502, "epoch": 0.7399747489929658, "flos": 16508605911840.0, "grad_norm": 2.1509733821419466, "language_loss": 0.79520494, "learning_rate": 6.681575678709404e-07, "loss": 0.81682074, "num_input_tokens_seen": 132305300, "step": 6154, "time_per_iteration": 2.4679226875305176 }, { "auxiliary_loss_clip": 0.01152379, "auxiliary_loss_mlp": 0.01029103, "balance_loss_clip": 1.04508483, "balance_loss_mlp": 1.02199769, "epoch": 0.7400949918836048, "flos": 24097069543680.0, "grad_norm": 1.9127734410415855, "language_loss": 0.70637912, "learning_rate": 6.67576538861911e-07, "loss": 0.72819394, "num_input_tokens_seen": 132323875, "step": 6155, "time_per_iteration": 2.5089566707611084 }, { "auxiliary_loss_clip": 0.0113886, "auxiliary_loss_mlp": 0.01028422, "balance_loss_clip": 1.04530418, "balance_loss_mlp": 1.02185678, "epoch": 0.740215234774244, "flos": 21802521931680.0, "grad_norm": 1.545715272376469, "language_loss": 0.82043415, "learning_rate": 6.669957119804612e-07, "loss": 0.842107, "num_input_tokens_seen": 132345510, "step": 6156, "time_per_iteration": 2.523635149002075 }, { "auxiliary_loss_clip": 0.01149148, "auxiliary_loss_mlp": 0.01023094, "balance_loss_clip": 1.04480803, "balance_loss_mlp": 1.01601243, "epoch": 0.7403354776648831, "flos": 18733200062400.0, "grad_norm": 2.712031167777262, "language_loss": 0.72441179, "learning_rate": 6.66415087314702e-07, "loss": 0.74613422, "num_input_tokens_seen": 132360465, "step": 6157, "time_per_iteration": 3.2172772884368896 }, { "auxiliary_loss_clip": 0.01141271, "auxiliary_loss_mlp": 0.0102306, "balance_loss_clip": 1.04279304, "balance_loss_mlp": 1.0158987, "epoch": 0.7404557205555221, "flos": 16909584415680.0, "grad_norm": 2.9320820002140078, "language_loss": 0.72845596, "learning_rate": 6.65834664952714e-07, "loss": 0.75009918, "num_input_tokens_seen": 132377915, "step": 6158, "time_per_iteration": 2.4848885536193848 }, { "auxiliary_loss_clip": 0.01128419, "auxiliary_loss_mlp": 0.01023727, "balance_loss_clip": 1.04307723, "balance_loss_mlp": 1.01735246, "epoch": 0.7405759634461613, "flos": 21214412090880.0, "grad_norm": 1.7826709342070683, "language_loss": 0.75772774, "learning_rate": 6.652544449825457e-07, "loss": 0.77924919, "num_input_tokens_seen": 132398170, "step": 6159, "time_per_iteration": 2.5652849674224854 }, { "auxiliary_loss_clip": 0.01146641, "auxiliary_loss_mlp": 0.01026863, "balance_loss_clip": 1.04521, "balance_loss_mlp": 1.01941836, "epoch": 0.7406962063368003, "flos": 20480109859200.0, "grad_norm": 1.5873561569469798, "language_loss": 0.76605284, "learning_rate": 6.646744274922182e-07, "loss": 0.78778791, "num_input_tokens_seen": 132416615, "step": 6160, "time_per_iteration": 2.5191030502319336 }, { "auxiliary_loss_clip": 0.01141887, "auxiliary_loss_mlp": 0.01025158, "balance_loss_clip": 1.04353178, "balance_loss_mlp": 1.01817203, "epoch": 0.7408164492274394, "flos": 19791918615840.0, "grad_norm": 3.006688529950598, "language_loss": 0.75107002, "learning_rate": 6.640946125697171e-07, "loss": 0.77274048, "num_input_tokens_seen": 132434145, "step": 6161, "time_per_iteration": 2.4831671714782715 }, { "auxiliary_loss_clip": 0.01157041, "auxiliary_loss_mlp": 0.0102511, "balance_loss_clip": 1.04605317, "balance_loss_mlp": 1.01760566, "epoch": 0.7409366921180786, "flos": 29204859901440.0, "grad_norm": 1.8012862454575227, "language_loss": 0.75686878, "learning_rate": 6.635150003030017e-07, "loss": 0.77869022, "num_input_tokens_seen": 132452670, "step": 6162, "time_per_iteration": 2.53540301322937 }, { "auxiliary_loss_clip": 0.01111393, "auxiliary_loss_mlp": 0.0102192, "balance_loss_clip": 1.0406363, "balance_loss_mlp": 1.01471722, "epoch": 0.7410569350087176, "flos": 22930008686880.0, "grad_norm": 2.2572560485796123, "language_loss": 0.8633604, "learning_rate": 6.629355907799981e-07, "loss": 0.88469356, "num_input_tokens_seen": 132472475, "step": 6163, "time_per_iteration": 2.63421630859375 }, { "auxiliary_loss_clip": 0.01156569, "auxiliary_loss_mlp": 0.01028064, "balance_loss_clip": 1.0464561, "balance_loss_mlp": 1.02036893, "epoch": 0.7411771778993567, "flos": 30440401624320.0, "grad_norm": 1.863632714077149, "language_loss": 0.69169855, "learning_rate": 6.623563840886015e-07, "loss": 0.71354491, "num_input_tokens_seen": 132493400, "step": 6164, "time_per_iteration": 2.644510507583618 }, { "auxiliary_loss_clip": 0.01149672, "auxiliary_loss_mlp": 0.01024132, "balance_loss_clip": 1.04393804, "balance_loss_mlp": 1.01769185, "epoch": 0.7412974207899958, "flos": 20522054480160.0, "grad_norm": 1.8114233607104173, "language_loss": 0.69451153, "learning_rate": 6.617773803166795e-07, "loss": 0.71624959, "num_input_tokens_seen": 132511725, "step": 6165, "time_per_iteration": 2.5834197998046875 }, { "auxiliary_loss_clip": 0.01149018, "auxiliary_loss_mlp": 0.00762599, "balance_loss_clip": 1.04823363, "balance_loss_mlp": 1.00058556, "epoch": 0.7414176636806349, "flos": 22090704429120.0, "grad_norm": 2.310765702528199, "language_loss": 0.81808841, "learning_rate": 6.611985795520634e-07, "loss": 0.83720458, "num_input_tokens_seen": 132530270, "step": 6166, "time_per_iteration": 2.6158013343811035 }, { "auxiliary_loss_clip": 0.01137536, "auxiliary_loss_mlp": 0.01028229, "balance_loss_clip": 1.04590309, "balance_loss_mlp": 1.0206238, "epoch": 0.7415379065712739, "flos": 25155249342720.0, "grad_norm": 2.672457239579851, "language_loss": 0.76643813, "learning_rate": 6.606199818825588e-07, "loss": 0.78809577, "num_input_tokens_seen": 132550725, "step": 6167, "time_per_iteration": 2.5745348930358887 }, { "auxiliary_loss_clip": 0.01146069, "auxiliary_loss_mlp": 0.01022274, "balance_loss_clip": 1.04390168, "balance_loss_mlp": 1.01580977, "epoch": 0.7416581494619131, "flos": 16871734328160.0, "grad_norm": 2.485351245982481, "language_loss": 0.81470633, "learning_rate": 6.600415873959377e-07, "loss": 0.83638978, "num_input_tokens_seen": 132568600, "step": 6168, "time_per_iteration": 2.515672445297241 }, { "auxiliary_loss_clip": 0.01099518, "auxiliary_loss_mlp": 0.00761305, "balance_loss_clip": 1.03990221, "balance_loss_mlp": 1.00054789, "epoch": 0.7417783923525522, "flos": 28438885299360.0, "grad_norm": 2.50547120694065, "language_loss": 0.64509803, "learning_rate": 6.594633961799437e-07, "loss": 0.6637063, "num_input_tokens_seen": 132587640, "step": 6169, "time_per_iteration": 2.668973922729492 }, { "auxiliary_loss_clip": 0.01135728, "auxiliary_loss_mlp": 0.01021812, "balance_loss_clip": 1.04421353, "balance_loss_mlp": 1.01538408, "epoch": 0.7418986352431912, "flos": 20084303397600.0, "grad_norm": 1.6788002132444095, "language_loss": 0.8154155, "learning_rate": 6.588854083222857e-07, "loss": 0.83699089, "num_input_tokens_seen": 132607075, "step": 6170, "time_per_iteration": 2.5762264728546143 }, { "auxiliary_loss_clip": 0.01145793, "auxiliary_loss_mlp": 0.01029982, "balance_loss_clip": 1.0467906, "balance_loss_mlp": 1.02244544, "epoch": 0.7420188781338304, "flos": 18259574325120.0, "grad_norm": 2.031481582962752, "language_loss": 0.8037529, "learning_rate": 6.583076239106444e-07, "loss": 0.82551062, "num_input_tokens_seen": 132625580, "step": 6171, "time_per_iteration": 3.297210454940796 }, { "auxiliary_loss_clip": 0.01148518, "auxiliary_loss_mlp": 0.01022847, "balance_loss_clip": 1.04656935, "balance_loss_mlp": 1.01565611, "epoch": 0.7421391210244694, "flos": 13771997182080.0, "grad_norm": 6.991407112080641, "language_loss": 0.75371897, "learning_rate": 6.577300430326707e-07, "loss": 0.77543259, "num_input_tokens_seen": 132640525, "step": 6172, "time_per_iteration": 2.5280449390411377 }, { "auxiliary_loss_clip": 0.01124866, "auxiliary_loss_mlp": 0.01025906, "balance_loss_clip": 1.04402173, "balance_loss_mlp": 1.01923966, "epoch": 0.7422593639151085, "flos": 15961686435840.0, "grad_norm": 2.185600785431701, "language_loss": 0.7182104, "learning_rate": 6.571526657759821e-07, "loss": 0.73971808, "num_input_tokens_seen": 132656265, "step": 6173, "time_per_iteration": 4.060878276824951 }, { "auxiliary_loss_clip": 0.01148451, "auxiliary_loss_mlp": 0.01020764, "balance_loss_clip": 1.04396319, "balance_loss_mlp": 1.01383209, "epoch": 0.7423796068057477, "flos": 30114404956320.0, "grad_norm": 1.5245063866476107, "language_loss": 0.70451593, "learning_rate": 6.565754922281663e-07, "loss": 0.72620809, "num_input_tokens_seen": 132678510, "step": 6174, "time_per_iteration": 2.561194896697998 }, { "auxiliary_loss_clip": 0.01138461, "auxiliary_loss_mlp": 0.01027276, "balance_loss_clip": 1.04275048, "balance_loss_mlp": 1.02016473, "epoch": 0.7424998496963867, "flos": 20521910812320.0, "grad_norm": 1.8790176841058281, "language_loss": 0.78388518, "learning_rate": 6.559985224767801e-07, "loss": 0.80554247, "num_input_tokens_seen": 132696385, "step": 6175, "time_per_iteration": 2.5121028423309326 }, { "auxiliary_loss_clip": 0.01129885, "auxiliary_loss_mlp": 0.01025634, "balance_loss_clip": 1.04462397, "balance_loss_mlp": 1.0185951, "epoch": 0.7426200925870258, "flos": 21871577469120.0, "grad_norm": 3.7850011808183086, "language_loss": 0.75355238, "learning_rate": 6.55421756609349e-07, "loss": 0.77510756, "num_input_tokens_seen": 132714640, "step": 6176, "time_per_iteration": 2.532292604446411 }, { "auxiliary_loss_clip": 0.01156404, "auxiliary_loss_mlp": 0.01029956, "balance_loss_clip": 1.04961538, "balance_loss_mlp": 1.02270508, "epoch": 0.7427403354776649, "flos": 26432053264320.0, "grad_norm": 1.9983649940072021, "language_loss": 0.78899366, "learning_rate": 6.54845194713369e-07, "loss": 0.8108573, "num_input_tokens_seen": 132735590, "step": 6177, "time_per_iteration": 2.5187110900878906 }, { "auxiliary_loss_clip": 0.01152204, "auxiliary_loss_mlp": 0.0103254, "balance_loss_clip": 1.04699409, "balance_loss_mlp": 1.02546215, "epoch": 0.742860578368304, "flos": 19898393237280.0, "grad_norm": 3.149435998331331, "language_loss": 0.80146432, "learning_rate": 6.542688368763034e-07, "loss": 0.82331181, "num_input_tokens_seen": 132753995, "step": 6178, "time_per_iteration": 2.4580295085906982 }, { "auxiliary_loss_clip": 0.01154972, "auxiliary_loss_mlp": 0.01022265, "balance_loss_clip": 1.04702139, "balance_loss_mlp": 1.01513362, "epoch": 0.742980821258943, "flos": 24827205408000.0, "grad_norm": 1.8166728012303475, "language_loss": 0.77075958, "learning_rate": 6.536926831855854e-07, "loss": 0.79253197, "num_input_tokens_seen": 132773160, "step": 6179, "time_per_iteration": 2.5200161933898926 }, { "auxiliary_loss_clip": 0.01137859, "auxiliary_loss_mlp": 0.01026633, "balance_loss_clip": 1.04578567, "balance_loss_mlp": 1.02025843, "epoch": 0.7431010641495821, "flos": 25228650832320.0, "grad_norm": 2.2237898053761636, "language_loss": 0.72896218, "learning_rate": 6.531167337286165e-07, "loss": 0.75060713, "num_input_tokens_seen": 132793180, "step": 6180, "time_per_iteration": 2.5176050662994385 }, { "auxiliary_loss_clip": 0.01143302, "auxiliary_loss_mlp": 0.01024613, "balance_loss_clip": 1.04861224, "balance_loss_mlp": 1.01782393, "epoch": 0.7432213070402213, "flos": 21762373158720.0, "grad_norm": 1.5878520336952398, "language_loss": 0.7970925, "learning_rate": 6.52540988592768e-07, "loss": 0.81877166, "num_input_tokens_seen": 132814200, "step": 6181, "time_per_iteration": 2.541086196899414 }, { "auxiliary_loss_clip": 0.01143038, "auxiliary_loss_mlp": 0.01028541, "balance_loss_clip": 1.04424, "balance_loss_mlp": 1.0215131, "epoch": 0.7433415499308603, "flos": 14793835405920.0, "grad_norm": 2.5376732137447298, "language_loss": 0.83690846, "learning_rate": 6.519654478653814e-07, "loss": 0.85862428, "num_input_tokens_seen": 132832565, "step": 6182, "time_per_iteration": 2.4767403602600098 }, { "auxiliary_loss_clip": 0.01048316, "auxiliary_loss_mlp": 0.01000934, "balance_loss_clip": 1.00881469, "balance_loss_mlp": 0.99956912, "epoch": 0.7434617928214994, "flos": 67155589017600.0, "grad_norm": 0.7493260166481024, "language_loss": 0.56205428, "learning_rate": 6.51390111633763e-07, "loss": 0.58254683, "num_input_tokens_seen": 132897840, "step": 6183, "time_per_iteration": 3.8426947593688965 }, { "auxiliary_loss_clip": 0.01097598, "auxiliary_loss_mlp": 0.01026435, "balance_loss_clip": 1.03843784, "balance_loss_mlp": 1.01965821, "epoch": 0.7435820357121385, "flos": 27377580724800.0, "grad_norm": 1.6241591608122579, "language_loss": 0.76014602, "learning_rate": 6.508149799851932e-07, "loss": 0.78138638, "num_input_tokens_seen": 132919505, "step": 6184, "time_per_iteration": 2.6690733432769775 }, { "auxiliary_loss_clip": 0.01136988, "auxiliary_loss_mlp": 0.01021608, "balance_loss_clip": 1.04347825, "balance_loss_mlp": 1.01516497, "epoch": 0.7437022786027776, "flos": 23987649731520.0, "grad_norm": 1.8142723782944505, "language_loss": 0.61317325, "learning_rate": 6.502400530069183e-07, "loss": 0.63475925, "num_input_tokens_seen": 132939390, "step": 6185, "time_per_iteration": 2.556948661804199 }, { "auxiliary_loss_clip": 0.01127811, "auxiliary_loss_mlp": 0.01031646, "balance_loss_clip": 1.04459321, "balance_loss_mlp": 1.02281833, "epoch": 0.7438225214934167, "flos": 21866764596480.0, "grad_norm": 2.2421390048806384, "language_loss": 0.68446076, "learning_rate": 6.496653307861535e-07, "loss": 0.70605528, "num_input_tokens_seen": 132960060, "step": 6186, "time_per_iteration": 2.588486433029175 }, { "auxiliary_loss_clip": 0.01160834, "auxiliary_loss_mlp": 0.01026575, "balance_loss_clip": 1.04619074, "balance_loss_mlp": 1.0190146, "epoch": 0.7439427643840558, "flos": 20230100701920.0, "grad_norm": 1.7614595556691919, "language_loss": 0.65753371, "learning_rate": 6.490908134100857e-07, "loss": 0.67940778, "num_input_tokens_seen": 132978525, "step": 6187, "time_per_iteration": 2.4875810146331787 }, { "auxiliary_loss_clip": 0.01159561, "auxiliary_loss_mlp": 0.01029781, "balance_loss_clip": 1.04743147, "balance_loss_mlp": 1.02278304, "epoch": 0.7440630072746949, "flos": 20849918830080.0, "grad_norm": 2.520127598276664, "language_loss": 0.69258666, "learning_rate": 6.48516500965866e-07, "loss": 0.71448004, "num_input_tokens_seen": 132998460, "step": 6188, "time_per_iteration": 2.5214314460754395 }, { "auxiliary_loss_clip": 0.01160664, "auxiliary_loss_mlp": 0.01025116, "balance_loss_clip": 1.0457499, "balance_loss_mlp": 1.01815391, "epoch": 0.7441832501653339, "flos": 26503766656800.0, "grad_norm": 2.5366879387690346, "language_loss": 0.81648624, "learning_rate": 6.479423935406192e-07, "loss": 0.83834404, "num_input_tokens_seen": 133018445, "step": 6189, "time_per_iteration": 2.54134202003479 }, { "auxiliary_loss_clip": 0.01041508, "auxiliary_loss_mlp": 0.01001565, "balance_loss_clip": 1.01039147, "balance_loss_mlp": 1.00027108, "epoch": 0.7443034930559731, "flos": 68602858806720.0, "grad_norm": 0.8745150918005425, "language_loss": 0.62029982, "learning_rate": 6.473684912214357e-07, "loss": 0.6407305, "num_input_tokens_seen": 133082005, "step": 6190, "time_per_iteration": 3.275733232498169 }, { "auxiliary_loss_clip": 0.01157157, "auxiliary_loss_mlp": 0.01030698, "balance_loss_clip": 1.04830909, "balance_loss_mlp": 1.02378654, "epoch": 0.7444237359466122, "flos": 18654985700160.0, "grad_norm": 2.798097790256754, "language_loss": 0.69646704, "learning_rate": 6.467947940953778e-07, "loss": 0.71834558, "num_input_tokens_seen": 133100530, "step": 6191, "time_per_iteration": 2.483778238296509 }, { "auxiliary_loss_clip": 0.01138216, "auxiliary_loss_mlp": 0.01024956, "balance_loss_clip": 1.04296434, "balance_loss_mlp": 1.01857257, "epoch": 0.7445439788372512, "flos": 22817607767040.0, "grad_norm": 1.7495731951407216, "language_loss": 0.72267628, "learning_rate": 6.462213022494732e-07, "loss": 0.74430799, "num_input_tokens_seen": 133119775, "step": 6192, "time_per_iteration": 2.543358087539673 }, { "auxiliary_loss_clip": 0.010563, "auxiliary_loss_mlp": 0.01002904, "balance_loss_clip": 1.00811517, "balance_loss_mlp": 1.00162888, "epoch": 0.7446642217278904, "flos": 67045702284960.0, "grad_norm": 0.77350987841267, "language_loss": 0.61001933, "learning_rate": 6.456480157707201e-07, "loss": 0.63061136, "num_input_tokens_seen": 133184550, "step": 6193, "time_per_iteration": 3.050449848175049 }, { "auxiliary_loss_clip": 0.01122953, "auxiliary_loss_mlp": 0.01024073, "balance_loss_clip": 1.04362333, "balance_loss_mlp": 1.01701617, "epoch": 0.7447844646185294, "flos": 17417468544480.0, "grad_norm": 2.151350748621846, "language_loss": 0.85158992, "learning_rate": 6.450749347460866e-07, "loss": 0.87306011, "num_input_tokens_seen": 133201525, "step": 6194, "time_per_iteration": 2.545957326889038 }, { "auxiliary_loss_clip": 0.01171186, "auxiliary_loss_mlp": 0.01026535, "balance_loss_clip": 1.0477047, "balance_loss_mlp": 1.01970696, "epoch": 0.7449047075091685, "flos": 26615880240960.0, "grad_norm": 1.9949759290215974, "language_loss": 0.78703558, "learning_rate": 6.445020592625083e-07, "loss": 0.80901277, "num_input_tokens_seen": 133222175, "step": 6195, "time_per_iteration": 2.4886200428009033 }, { "auxiliary_loss_clip": 0.01168153, "auxiliary_loss_mlp": 0.0102275, "balance_loss_clip": 1.04655862, "balance_loss_mlp": 1.01557314, "epoch": 0.7450249503998077, "flos": 14170461498720.0, "grad_norm": 2.394814269538161, "language_loss": 0.79469395, "learning_rate": 6.4392938940689e-07, "loss": 0.81660306, "num_input_tokens_seen": 133237590, "step": 6196, "time_per_iteration": 2.4322774410247803 }, { "auxiliary_loss_clip": 0.01111113, "auxiliary_loss_mlp": 0.00762084, "balance_loss_clip": 1.04194069, "balance_loss_mlp": 1.0005815, "epoch": 0.7451451932904467, "flos": 19606690877760.0, "grad_norm": 3.561402972735996, "language_loss": 0.71306944, "learning_rate": 6.433569252661049e-07, "loss": 0.73180151, "num_input_tokens_seen": 133255590, "step": 6197, "time_per_iteration": 3.4475760459899902 }, { "auxiliary_loss_clip": 0.01119481, "auxiliary_loss_mlp": 0.01027486, "balance_loss_clip": 1.04133534, "balance_loss_mlp": 1.02092636, "epoch": 0.7452654361810858, "flos": 12495408762240.0, "grad_norm": 1.9943255166008902, "language_loss": 0.71352732, "learning_rate": 6.427846669269952e-07, "loss": 0.73499697, "num_input_tokens_seen": 133273210, "step": 6198, "time_per_iteration": 2.523451328277588 }, { "auxiliary_loss_clip": 0.01172447, "auxiliary_loss_mlp": 0.01024068, "balance_loss_clip": 1.05094051, "balance_loss_mlp": 1.01696062, "epoch": 0.7453856790717249, "flos": 22127333340000.0, "grad_norm": 2.0117701911225274, "language_loss": 0.81851125, "learning_rate": 6.422126144763729e-07, "loss": 0.84047639, "num_input_tokens_seen": 133292600, "step": 6199, "time_per_iteration": 3.2274844646453857 }, { "auxiliary_loss_clip": 0.01125228, "auxiliary_loss_mlp": 0.00763002, "balance_loss_clip": 1.03996682, "balance_loss_mlp": 1.00055122, "epoch": 0.745505921962364, "flos": 20010686406240.0, "grad_norm": 2.4993673633452493, "language_loss": 0.77217865, "learning_rate": 6.416407680010174e-07, "loss": 0.79106098, "num_input_tokens_seen": 133306960, "step": 6200, "time_per_iteration": 3.2539100646972656 }, { "auxiliary_loss_clip": 0.01125381, "auxiliary_loss_mlp": 0.01025464, "balance_loss_clip": 1.04556322, "balance_loss_mlp": 1.01852274, "epoch": 0.745626164853003, "flos": 24677888241600.0, "grad_norm": 2.0282426475329345, "language_loss": 0.81138396, "learning_rate": 6.410691275876774e-07, "loss": 0.83289242, "num_input_tokens_seen": 133326380, "step": 6201, "time_per_iteration": 2.620171308517456 }, { "auxiliary_loss_clip": 0.01145692, "auxiliary_loss_mlp": 0.01027409, "balance_loss_clip": 1.04566002, "balance_loss_mlp": 1.02046216, "epoch": 0.7457464077436422, "flos": 14538833791200.0, "grad_norm": 2.1625502479212577, "language_loss": 0.76627004, "learning_rate": 6.404976933230704e-07, "loss": 0.78800106, "num_input_tokens_seen": 133342900, "step": 6202, "time_per_iteration": 2.496666669845581 }, { "auxiliary_loss_clip": 0.01148092, "auxiliary_loss_mlp": 0.01029828, "balance_loss_clip": 1.04564881, "balance_loss_mlp": 1.02243662, "epoch": 0.7458666506342813, "flos": 34021199318400.0, "grad_norm": 1.7128671671922817, "language_loss": 0.72670966, "learning_rate": 6.399264652938813e-07, "loss": 0.7484889, "num_input_tokens_seen": 133363805, "step": 6203, "time_per_iteration": 2.6272037029266357 }, { "auxiliary_loss_clip": 0.01140427, "auxiliary_loss_mlp": 0.01020818, "balance_loss_clip": 1.0442214, "balance_loss_mlp": 1.01374006, "epoch": 0.7459868935249203, "flos": 24279028838400.0, "grad_norm": 1.7903399662211326, "language_loss": 0.74436438, "learning_rate": 6.393554435867679e-07, "loss": 0.76597679, "num_input_tokens_seen": 133384655, "step": 6204, "time_per_iteration": 2.5535359382629395 }, { "auxiliary_loss_clip": 0.0112386, "auxiliary_loss_mlp": 0.01021295, "balance_loss_clip": 1.04178774, "balance_loss_mlp": 1.01421392, "epoch": 0.7461071364155595, "flos": 21908781051360.0, "grad_norm": 2.189405823060587, "language_loss": 0.83670545, "learning_rate": 6.387846282883502e-07, "loss": 0.85815692, "num_input_tokens_seen": 133401185, "step": 6205, "time_per_iteration": 2.5617027282714844 }, { "auxiliary_loss_clip": 0.01169816, "auxiliary_loss_mlp": 0.01026918, "balance_loss_clip": 1.04845691, "balance_loss_mlp": 1.019593, "epoch": 0.7462273793061985, "flos": 22889716246080.0, "grad_norm": 1.9111779103662223, "language_loss": 0.76679122, "learning_rate": 6.38214019485223e-07, "loss": 0.78875858, "num_input_tokens_seen": 133420010, "step": 6206, "time_per_iteration": 2.4568727016448975 }, { "auxiliary_loss_clip": 0.01094306, "auxiliary_loss_mlp": 0.0102379, "balance_loss_clip": 1.03948116, "balance_loss_mlp": 1.01684594, "epoch": 0.7463476221968376, "flos": 19968454449600.0, "grad_norm": 1.6608161807181134, "language_loss": 0.71750093, "learning_rate": 6.376436172639461e-07, "loss": 0.73868191, "num_input_tokens_seen": 133437855, "step": 6207, "time_per_iteration": 2.6167068481445312 }, { "auxiliary_loss_clip": 0.01087598, "auxiliary_loss_mlp": 0.01028367, "balance_loss_clip": 1.04146695, "balance_loss_mlp": 1.02065396, "epoch": 0.7464678650874768, "flos": 16836613929600.0, "grad_norm": 2.402849210724651, "language_loss": 0.64815098, "learning_rate": 6.370734217110487e-07, "loss": 0.66931063, "num_input_tokens_seen": 133456600, "step": 6208, "time_per_iteration": 2.643841505050659 }, { "auxiliary_loss_clip": 0.01149063, "auxiliary_loss_mlp": 0.01025388, "balance_loss_clip": 1.05060744, "balance_loss_mlp": 1.01737142, "epoch": 0.7465881079781158, "flos": 48100875518880.0, "grad_norm": 11.121767454272145, "language_loss": 0.6405775, "learning_rate": 6.36503432913031e-07, "loss": 0.66232198, "num_input_tokens_seen": 133479745, "step": 6209, "time_per_iteration": 3.4880523681640625 }, { "auxiliary_loss_clip": 0.01154933, "auxiliary_loss_mlp": 0.01026442, "balance_loss_clip": 1.04750299, "balance_loss_mlp": 1.01907814, "epoch": 0.7467083508687549, "flos": 19677362678400.0, "grad_norm": 7.1256364045787715, "language_loss": 0.68906373, "learning_rate": 6.359336509563569e-07, "loss": 0.71087754, "num_input_tokens_seen": 133495765, "step": 6210, "time_per_iteration": 2.4905054569244385 }, { "auxiliary_loss_clip": 0.01114542, "auxiliary_loss_mlp": 0.01020776, "balance_loss_clip": 1.04226804, "balance_loss_mlp": 1.01372445, "epoch": 0.7468285937593939, "flos": 17895440233920.0, "grad_norm": 1.8245433058028087, "language_loss": 0.80606341, "learning_rate": 6.353640759274641e-07, "loss": 0.82741654, "num_input_tokens_seen": 133514655, "step": 6211, "time_per_iteration": 2.5187597274780273 }, { "auxiliary_loss_clip": 0.01154794, "auxiliary_loss_mlp": 0.01023925, "balance_loss_clip": 1.04529381, "balance_loss_mlp": 1.01683784, "epoch": 0.7469488366500331, "flos": 23141449417440.0, "grad_norm": 2.4291732074517665, "language_loss": 0.75049293, "learning_rate": 6.347947079127556e-07, "loss": 0.7722801, "num_input_tokens_seen": 133532555, "step": 6212, "time_per_iteration": 2.514775276184082 }, { "auxiliary_loss_clip": 0.01136366, "auxiliary_loss_mlp": 0.01030535, "balance_loss_clip": 1.04505801, "balance_loss_mlp": 1.02366555, "epoch": 0.7470690795406721, "flos": 16690852542240.0, "grad_norm": 3.3855138219483782, "language_loss": 0.76852095, "learning_rate": 6.342255469986053e-07, "loss": 0.79018998, "num_input_tokens_seen": 133551300, "step": 6213, "time_per_iteration": 2.4803152084350586 }, { "auxiliary_loss_clip": 0.01171044, "auxiliary_loss_mlp": 0.0102468, "balance_loss_clip": 1.04997993, "balance_loss_mlp": 1.01787364, "epoch": 0.7471893224313112, "flos": 25192704343680.0, "grad_norm": 1.673322818776939, "language_loss": 0.76405406, "learning_rate": 6.336565932713533e-07, "loss": 0.78601128, "num_input_tokens_seen": 133570725, "step": 6214, "time_per_iteration": 2.492680788040161 }, { "auxiliary_loss_clip": 0.01142892, "auxiliary_loss_mlp": 0.01027387, "balance_loss_clip": 1.04859114, "balance_loss_mlp": 1.01998079, "epoch": 0.7473095653219504, "flos": 22526228660160.0, "grad_norm": 1.6880727830622624, "language_loss": 0.77748364, "learning_rate": 6.330878468173088e-07, "loss": 0.79918641, "num_input_tokens_seen": 133590790, "step": 6215, "time_per_iteration": 2.531968355178833 }, { "auxiliary_loss_clip": 0.01148573, "auxiliary_loss_mlp": 0.01024531, "balance_loss_clip": 1.04500008, "balance_loss_mlp": 1.01728284, "epoch": 0.7474298082125894, "flos": 18113992522560.0, "grad_norm": 2.0610482974848474, "language_loss": 0.72766012, "learning_rate": 6.32519307722752e-07, "loss": 0.7493912, "num_input_tokens_seen": 133608685, "step": 6216, "time_per_iteration": 2.482543706893921 }, { "auxiliary_loss_clip": 0.01042725, "auxiliary_loss_mlp": 0.01001194, "balance_loss_clip": 1.02276707, "balance_loss_mlp": 0.99993086, "epoch": 0.7475500511032285, "flos": 62086546671360.0, "grad_norm": 0.9023350664774038, "language_loss": 0.55033118, "learning_rate": 6.31950976073929e-07, "loss": 0.57077038, "num_input_tokens_seen": 133662775, "step": 6217, "time_per_iteration": 3.125199317932129 }, { "auxiliary_loss_clip": 0.01111878, "auxiliary_loss_mlp": 0.01029878, "balance_loss_clip": 1.0434401, "balance_loss_mlp": 1.02223074, "epoch": 0.7476702939938676, "flos": 17785589418240.0, "grad_norm": 2.420866377238385, "language_loss": 0.81166381, "learning_rate": 6.31382851957055e-07, "loss": 0.83308136, "num_input_tokens_seen": 133679595, "step": 6218, "time_per_iteration": 2.5616612434387207 }, { "auxiliary_loss_clip": 0.01121937, "auxiliary_loss_mlp": 0.00761901, "balance_loss_clip": 1.04263854, "balance_loss_mlp": 1.00048614, "epoch": 0.7477905368845067, "flos": 27927948228960.0, "grad_norm": 1.8889399426443458, "language_loss": 0.71888137, "learning_rate": 6.308149354583143e-07, "loss": 0.73771971, "num_input_tokens_seen": 133699000, "step": 6219, "time_per_iteration": 2.6454715728759766 }, { "auxiliary_loss_clip": 0.0116206, "auxiliary_loss_mlp": 0.01025923, "balance_loss_clip": 1.04966545, "balance_loss_mlp": 1.018466, "epoch": 0.7479107797751458, "flos": 26870379018240.0, "grad_norm": 1.8452076948997254, "language_loss": 0.81682825, "learning_rate": 6.302472266638586e-07, "loss": 0.83870804, "num_input_tokens_seen": 133719540, "step": 6220, "time_per_iteration": 2.5290558338165283 }, { "auxiliary_loss_clip": 0.01178427, "auxiliary_loss_mlp": 0.0103177, "balance_loss_clip": 1.05103445, "balance_loss_mlp": 1.02407503, "epoch": 0.7480310226657849, "flos": 33943380042720.0, "grad_norm": 2.008081243407756, "language_loss": 0.69992447, "learning_rate": 6.296797256598101e-07, "loss": 0.72202641, "num_input_tokens_seen": 133741020, "step": 6221, "time_per_iteration": 2.564509868621826 }, { "auxiliary_loss_clip": 0.01117312, "auxiliary_loss_mlp": 0.01029535, "balance_loss_clip": 1.04146922, "balance_loss_mlp": 1.02285683, "epoch": 0.748151265556424, "flos": 24826558902720.0, "grad_norm": 1.663981638903778, "language_loss": 0.81154835, "learning_rate": 6.291124325322576e-07, "loss": 0.83301675, "num_input_tokens_seen": 133761145, "step": 6222, "time_per_iteration": 3.394357681274414 }, { "auxiliary_loss_clip": 0.01147469, "auxiliary_loss_mlp": 0.01028348, "balance_loss_clip": 1.04567504, "balance_loss_mlp": 1.02123129, "epoch": 0.748271508447063, "flos": 38399356649280.0, "grad_norm": 1.5546779462948803, "language_loss": 0.62295187, "learning_rate": 6.285453473672595e-07, "loss": 0.64471006, "num_input_tokens_seen": 133783715, "step": 6223, "time_per_iteration": 2.7579805850982666 }, { "auxiliary_loss_clip": 0.01169029, "auxiliary_loss_mlp": 0.01023334, "balance_loss_clip": 1.04799867, "balance_loss_mlp": 1.01622868, "epoch": 0.7483917513377022, "flos": 21541845437280.0, "grad_norm": 2.3003783615938556, "language_loss": 0.7531147, "learning_rate": 6.279784702508415e-07, "loss": 0.77503836, "num_input_tokens_seen": 133804465, "step": 6224, "time_per_iteration": 2.582749128341675 }, { "auxiliary_loss_clip": 0.01034632, "auxiliary_loss_mlp": 0.01005443, "balance_loss_clip": 1.00916052, "balance_loss_mlp": 1.00410748, "epoch": 0.7485119942283412, "flos": 62314545120480.0, "grad_norm": 0.7752152911564814, "language_loss": 0.58631003, "learning_rate": 6.274118012689979e-07, "loss": 0.60671079, "num_input_tokens_seen": 133866365, "step": 6225, "time_per_iteration": 4.133336544036865 }, { "auxiliary_loss_clip": 0.01133602, "auxiliary_loss_mlp": 0.01026375, "balance_loss_clip": 1.04242396, "balance_loss_mlp": 1.0190351, "epoch": 0.7486322371189803, "flos": 29937617703840.0, "grad_norm": 1.524014560035851, "language_loss": 0.6808784, "learning_rate": 6.268453405076943e-07, "loss": 0.70247817, "num_input_tokens_seen": 133888760, "step": 6226, "time_per_iteration": 2.684833526611328 }, { "auxiliary_loss_clip": 0.01142487, "auxiliary_loss_mlp": 0.01028414, "balance_loss_clip": 1.04469693, "balance_loss_mlp": 1.02149057, "epoch": 0.7487524800096195, "flos": 18949417748640.0, "grad_norm": 1.9579956341174654, "language_loss": 0.81834471, "learning_rate": 6.262790880528592e-07, "loss": 0.84005368, "num_input_tokens_seen": 133906380, "step": 6227, "time_per_iteration": 2.650149345397949 }, { "auxiliary_loss_clip": 0.01136951, "auxiliary_loss_mlp": 0.01026004, "balance_loss_clip": 1.04069519, "balance_loss_mlp": 1.01853514, "epoch": 0.7488727229002585, "flos": 18697397241600.0, "grad_norm": 2.2322242845334395, "language_loss": 0.79225665, "learning_rate": 6.257130439903951e-07, "loss": 0.81388617, "num_input_tokens_seen": 133922875, "step": 6228, "time_per_iteration": 2.667727470397949 }, { "auxiliary_loss_clip": 0.01173407, "auxiliary_loss_mlp": 0.01027373, "balance_loss_clip": 1.05067348, "balance_loss_mlp": 1.0200597, "epoch": 0.7489929657908976, "flos": 23623371972480.0, "grad_norm": 1.8149741625930345, "language_loss": 0.80998945, "learning_rate": 6.251472084061695e-07, "loss": 0.83199728, "num_input_tokens_seen": 133941795, "step": 6229, "time_per_iteration": 2.503632068634033 }, { "auxiliary_loss_clip": 0.01153227, "auxiliary_loss_mlp": 0.01022913, "balance_loss_clip": 1.04643321, "balance_loss_mlp": 1.01614451, "epoch": 0.7491132086815367, "flos": 20551535916000.0, "grad_norm": 1.976315520281815, "language_loss": 0.88829327, "learning_rate": 6.245815813860191e-07, "loss": 0.91005468, "num_input_tokens_seen": 133957305, "step": 6230, "time_per_iteration": 2.4941468238830566 }, { "auxiliary_loss_clip": 0.01172851, "auxiliary_loss_mlp": 0.01019764, "balance_loss_clip": 1.04854941, "balance_loss_mlp": 1.01252472, "epoch": 0.7492334515721758, "flos": 23003015089920.0, "grad_norm": 2.0018148281428214, "language_loss": 0.70641971, "learning_rate": 6.240161630157495e-07, "loss": 0.72834587, "num_input_tokens_seen": 133976660, "step": 6231, "time_per_iteration": 2.4518494606018066 }, { "auxiliary_loss_clip": 0.01174501, "auxiliary_loss_mlp": 0.01023938, "balance_loss_clip": 1.04929984, "balance_loss_mlp": 1.01621342, "epoch": 0.7493536944628149, "flos": 16398826930080.0, "grad_norm": 2.054790240192329, "language_loss": 0.69935286, "learning_rate": 6.23450953381133e-07, "loss": 0.7213372, "num_input_tokens_seen": 133994750, "step": 6232, "time_per_iteration": 2.426896572113037 }, { "auxiliary_loss_clip": 0.01133839, "auxiliary_loss_mlp": 0.01026319, "balance_loss_clip": 1.04304743, "balance_loss_mlp": 1.01969075, "epoch": 0.749473937353454, "flos": 15338563947360.0, "grad_norm": 2.162724861496442, "language_loss": 0.67705435, "learning_rate": 6.228859525679131e-07, "loss": 0.69865596, "num_input_tokens_seen": 134009165, "step": 6233, "time_per_iteration": 2.4787120819091797 }, { "auxiliary_loss_clip": 0.01155704, "auxiliary_loss_mlp": 0.01024041, "balance_loss_clip": 1.04732788, "balance_loss_mlp": 1.01704037, "epoch": 0.7495941802440931, "flos": 18951141762720.0, "grad_norm": 2.0297358037301265, "language_loss": 0.79871434, "learning_rate": 6.223211606617986e-07, "loss": 0.82051182, "num_input_tokens_seen": 134027585, "step": 6234, "time_per_iteration": 2.474682092666626 }, { "auxiliary_loss_clip": 0.01153146, "auxiliary_loss_mlp": 0.01025412, "balance_loss_clip": 1.04810143, "balance_loss_mlp": 1.01946604, "epoch": 0.7497144231347321, "flos": 22492473106080.0, "grad_norm": 1.7677271427174743, "language_loss": 0.84079456, "learning_rate": 6.217565777484701e-07, "loss": 0.86258018, "num_input_tokens_seen": 134046680, "step": 6235, "time_per_iteration": 3.230018377304077 }, { "auxiliary_loss_clip": 0.01137821, "auxiliary_loss_mlp": 0.00761828, "balance_loss_clip": 1.04468858, "balance_loss_mlp": 1.00053644, "epoch": 0.7498346660253713, "flos": 24243513353280.0, "grad_norm": 2.031931708432506, "language_loss": 0.80672431, "learning_rate": 6.211922039135722e-07, "loss": 0.82572079, "num_input_tokens_seen": 134066825, "step": 6236, "time_per_iteration": 2.560783624649048 }, { "auxiliary_loss_clip": 0.01172138, "auxiliary_loss_mlp": 0.01023012, "balance_loss_clip": 1.05083179, "balance_loss_mlp": 1.01612186, "epoch": 0.7499549089160104, "flos": 24387083806080.0, "grad_norm": 1.8993049957287427, "language_loss": 0.80771673, "learning_rate": 6.206280392427201e-07, "loss": 0.82966816, "num_input_tokens_seen": 134086410, "step": 6237, "time_per_iteration": 2.4683680534362793 }, { "auxiliary_loss_clip": 0.01148252, "auxiliary_loss_mlp": 0.01025233, "balance_loss_clip": 1.0447042, "balance_loss_mlp": 1.01852143, "epoch": 0.7500751518066494, "flos": 34057325391840.0, "grad_norm": 1.6817762208620795, "language_loss": 0.73686564, "learning_rate": 6.200640838214983e-07, "loss": 0.75860047, "num_input_tokens_seen": 134109185, "step": 6238, "time_per_iteration": 2.5852596759796143 }, { "auxiliary_loss_clip": 0.0116994, "auxiliary_loss_mlp": 0.01025697, "balance_loss_clip": 1.04870701, "balance_loss_mlp": 1.01891971, "epoch": 0.7501953946972886, "flos": 18843589632480.0, "grad_norm": 1.7979513703984515, "language_loss": 0.66390294, "learning_rate": 6.195003377354578e-07, "loss": 0.68585932, "num_input_tokens_seen": 134128455, "step": 6239, "time_per_iteration": 2.4290642738342285 }, { "auxiliary_loss_clip": 0.01151916, "auxiliary_loss_mlp": 0.01025354, "balance_loss_clip": 1.04470778, "balance_loss_mlp": 1.01862121, "epoch": 0.7503156375879276, "flos": 20257678538880.0, "grad_norm": 2.423152582143233, "language_loss": 0.73370957, "learning_rate": 6.189368010701183e-07, "loss": 0.75548226, "num_input_tokens_seen": 134145515, "step": 6240, "time_per_iteration": 2.485628843307495 }, { "auxiliary_loss_clip": 0.01161068, "auxiliary_loss_mlp": 0.01030215, "balance_loss_clip": 1.04598582, "balance_loss_mlp": 1.02352738, "epoch": 0.7504358804785667, "flos": 13480043403840.0, "grad_norm": 1.8053958148267915, "language_loss": 0.76265168, "learning_rate": 6.183734739109683e-07, "loss": 0.7845645, "num_input_tokens_seen": 134163335, "step": 6241, "time_per_iteration": 2.4504005908966064 }, { "auxiliary_loss_clip": 0.01162875, "auxiliary_loss_mlp": 0.01030593, "balance_loss_clip": 1.05001593, "balance_loss_mlp": 1.02276969, "epoch": 0.7505561233692057, "flos": 29461046775840.0, "grad_norm": 2.0021510510751783, "language_loss": 0.68860936, "learning_rate": 6.178103563434629e-07, "loss": 0.71054405, "num_input_tokens_seen": 134182335, "step": 6242, "time_per_iteration": 2.542931079864502 }, { "auxiliary_loss_clip": 0.01170438, "auxiliary_loss_mlp": 0.01023918, "balance_loss_clip": 1.04887915, "balance_loss_mlp": 1.01633084, "epoch": 0.7506763662598449, "flos": 20302460599680.0, "grad_norm": 2.0007278728582247, "language_loss": 0.84008151, "learning_rate": 6.172474484530283e-07, "loss": 0.86202508, "num_input_tokens_seen": 134201070, "step": 6243, "time_per_iteration": 2.446063280105591 }, { "auxiliary_loss_clip": 0.01128655, "auxiliary_loss_mlp": 0.01025937, "balance_loss_clip": 1.04055667, "balance_loss_mlp": 1.01852226, "epoch": 0.750796609150484, "flos": 37230966864960.0, "grad_norm": 1.6949099191166432, "language_loss": 0.76008236, "learning_rate": 6.166847503250563e-07, "loss": 0.78162831, "num_input_tokens_seen": 134223310, "step": 6244, "time_per_iteration": 2.6725339889526367 }, { "auxiliary_loss_clip": 0.01141131, "auxiliary_loss_mlp": 0.01024383, "balance_loss_clip": 1.04427731, "balance_loss_mlp": 1.01705146, "epoch": 0.750916852041123, "flos": 19609420566720.0, "grad_norm": 2.6458399099799745, "language_loss": 0.79239464, "learning_rate": 6.161222620449078e-07, "loss": 0.81404978, "num_input_tokens_seen": 134242085, "step": 6245, "time_per_iteration": 2.4940896034240723 }, { "auxiliary_loss_clip": 0.01128997, "auxiliary_loss_mlp": 0.01027698, "balance_loss_clip": 1.04348278, "balance_loss_mlp": 1.02084661, "epoch": 0.7510370949317622, "flos": 25112694133440.0, "grad_norm": 2.134341493388232, "language_loss": 0.80097729, "learning_rate": 6.155599836979117e-07, "loss": 0.82254422, "num_input_tokens_seen": 134260770, "step": 6246, "time_per_iteration": 2.5978925228118896 }, { "auxiliary_loss_clip": 0.01114869, "auxiliary_loss_mlp": 0.01035819, "balance_loss_clip": 1.04121006, "balance_loss_mlp": 1.02759361, "epoch": 0.7511573378224012, "flos": 19062285588960.0, "grad_norm": 2.318212115816824, "language_loss": 0.81923676, "learning_rate": 6.149979153693649e-07, "loss": 0.84074354, "num_input_tokens_seen": 134278025, "step": 6247, "time_per_iteration": 2.558126211166382 }, { "auxiliary_loss_clip": 0.01152518, "auxiliary_loss_mlp": 0.01024979, "balance_loss_clip": 1.04519427, "balance_loss_mlp": 1.01765108, "epoch": 0.7512775807130403, "flos": 19937680003200.0, "grad_norm": 4.069092860026141, "language_loss": 0.76799333, "learning_rate": 6.144360571445343e-07, "loss": 0.78976834, "num_input_tokens_seen": 134297170, "step": 6248, "time_per_iteration": 3.3034133911132812 }, { "auxiliary_loss_clip": 0.01153659, "auxiliary_loss_mlp": 0.01026264, "balance_loss_clip": 1.04660368, "balance_loss_mlp": 1.01950765, "epoch": 0.7513978236036795, "flos": 20739924346560.0, "grad_norm": 1.6261192974600376, "language_loss": 0.80354023, "learning_rate": 6.138744091086509e-07, "loss": 0.82533944, "num_input_tokens_seen": 134316755, "step": 6249, "time_per_iteration": 2.514004707336426 }, { "auxiliary_loss_clip": 0.01133707, "auxiliary_loss_mlp": 0.01026079, "balance_loss_clip": 1.04684711, "balance_loss_mlp": 1.01943636, "epoch": 0.7515180664943185, "flos": 27563167632480.0, "grad_norm": 2.1247077856163648, "language_loss": 0.72655481, "learning_rate": 6.133129713469183e-07, "loss": 0.74815273, "num_input_tokens_seen": 134335960, "step": 6250, "time_per_iteration": 2.6138758659362793 }, { "auxiliary_loss_clip": 0.01135665, "auxiliary_loss_mlp": 0.01022431, "balance_loss_clip": 1.04234684, "balance_loss_mlp": 1.01507545, "epoch": 0.7516383093849576, "flos": 33803185784160.0, "grad_norm": 1.8357718378663626, "language_loss": 0.64140749, "learning_rate": 6.127517439445053e-07, "loss": 0.66298842, "num_input_tokens_seen": 134356805, "step": 6251, "time_per_iteration": 3.4393439292907715 }, { "auxiliary_loss_clip": 0.0110305, "auxiliary_loss_mlp": 0.01025089, "balance_loss_clip": 1.04250455, "balance_loss_mlp": 1.01862788, "epoch": 0.7517585522755967, "flos": 29746176331680.0, "grad_norm": 2.098250022865222, "language_loss": 0.81598312, "learning_rate": 6.121907269865498e-07, "loss": 0.83726454, "num_input_tokens_seen": 134376295, "step": 6252, "time_per_iteration": 2.637850761413574 }, { "auxiliary_loss_clip": 0.01031284, "auxiliary_loss_mlp": 0.01002422, "balance_loss_clip": 1.01060307, "balance_loss_mlp": 1.00104511, "epoch": 0.7518787951662358, "flos": 69807985252800.0, "grad_norm": 0.9321354274919649, "language_loss": 0.67315304, "learning_rate": 6.116299205581577e-07, "loss": 0.69349015, "num_input_tokens_seen": 134431125, "step": 6253, "time_per_iteration": 3.109527349472046 }, { "auxiliary_loss_clip": 0.01177496, "auxiliary_loss_mlp": 0.01026248, "balance_loss_clip": 1.05149937, "balance_loss_mlp": 1.01862812, "epoch": 0.7519990380568748, "flos": 34203230447040.0, "grad_norm": 1.9561070846976554, "language_loss": 0.68516713, "learning_rate": 6.110693247444018e-07, "loss": 0.70720458, "num_input_tokens_seen": 134452960, "step": 6254, "time_per_iteration": 2.559239387512207 }, { "auxiliary_loss_clip": 0.01116209, "auxiliary_loss_mlp": 0.01024027, "balance_loss_clip": 1.04212809, "balance_loss_mlp": 1.01672518, "epoch": 0.752119280947514, "flos": 21725708330880.0, "grad_norm": 1.7826447312272944, "language_loss": 0.82755101, "learning_rate": 6.105089396303258e-07, "loss": 0.84895331, "num_input_tokens_seen": 134471350, "step": 6255, "time_per_iteration": 2.540940761566162 }, { "auxiliary_loss_clip": 0.01142778, "auxiliary_loss_mlp": 0.01023821, "balance_loss_clip": 1.0445857, "balance_loss_mlp": 1.01610804, "epoch": 0.7522395238381531, "flos": 32742779133600.0, "grad_norm": 2.3159574626922907, "language_loss": 0.7558037, "learning_rate": 6.099487653009383e-07, "loss": 0.77746975, "num_input_tokens_seen": 134490695, "step": 6256, "time_per_iteration": 2.605649948120117 }, { "auxiliary_loss_clip": 0.01154631, "auxiliary_loss_mlp": 0.01025243, "balance_loss_clip": 1.04619837, "balance_loss_mlp": 1.01882064, "epoch": 0.7523597667287921, "flos": 23476030238880.0, "grad_norm": 1.9673190266663398, "language_loss": 0.83036005, "learning_rate": 6.093888018412192e-07, "loss": 0.85215884, "num_input_tokens_seen": 134506885, "step": 6257, "time_per_iteration": 2.5123369693756104 }, { "auxiliary_loss_clip": 0.01056711, "auxiliary_loss_mlp": 0.01001411, "balance_loss_clip": 1.00851607, "balance_loss_mlp": 1.00009978, "epoch": 0.7524800096194313, "flos": 67346743054080.0, "grad_norm": 0.844391267802732, "language_loss": 0.54692566, "learning_rate": 6.088290493361125e-07, "loss": 0.56750691, "num_input_tokens_seen": 134571770, "step": 6258, "time_per_iteration": 3.2493207454681396 }, { "auxiliary_loss_clip": 0.01108741, "auxiliary_loss_mlp": 0.0102495, "balance_loss_clip": 1.04273331, "balance_loss_mlp": 1.01755631, "epoch": 0.7526002525100703, "flos": 13006058496960.0, "grad_norm": 3.3781230702056613, "language_loss": 0.71241695, "learning_rate": 6.082695078705322e-07, "loss": 0.7337538, "num_input_tokens_seen": 134589250, "step": 6259, "time_per_iteration": 2.575684070587158 }, { "auxiliary_loss_clip": 0.01148387, "auxiliary_loss_mlp": 0.01028608, "balance_loss_clip": 1.04441357, "balance_loss_mlp": 1.02071369, "epoch": 0.7527204954007094, "flos": 21397233392640.0, "grad_norm": 1.8282901930633606, "language_loss": 0.68519914, "learning_rate": 6.077101775293618e-07, "loss": 0.70696902, "num_input_tokens_seen": 134608075, "step": 6260, "time_per_iteration": 2.5323214530944824 }, { "auxiliary_loss_clip": 0.01158932, "auxiliary_loss_mlp": 0.01028923, "balance_loss_clip": 1.04707432, "balance_loss_mlp": 1.02140665, "epoch": 0.7528407382913486, "flos": 18947190897120.0, "grad_norm": 2.4209103714218085, "language_loss": 0.82450795, "learning_rate": 6.071510583974504e-07, "loss": 0.84638655, "num_input_tokens_seen": 134623260, "step": 6261, "time_per_iteration": 3.233184576034546 }, { "auxiliary_loss_clip": 0.01171779, "auxiliary_loss_mlp": 0.01026427, "balance_loss_clip": 1.04846573, "balance_loss_mlp": 1.01971245, "epoch": 0.7529609811819876, "flos": 15231801990240.0, "grad_norm": 2.3671762803844927, "language_loss": 0.72025174, "learning_rate": 6.065921505596161e-07, "loss": 0.74223381, "num_input_tokens_seen": 134641540, "step": 6262, "time_per_iteration": 2.471243143081665 }, { "auxiliary_loss_clip": 0.01127882, "auxiliary_loss_mlp": 0.0102539, "balance_loss_clip": 1.04596663, "balance_loss_mlp": 1.01786184, "epoch": 0.7530812240726267, "flos": 19354490785920.0, "grad_norm": 1.7115591430918458, "language_loss": 0.76641768, "learning_rate": 6.060334541006445e-07, "loss": 0.7879504, "num_input_tokens_seen": 134660035, "step": 6263, "time_per_iteration": 2.55572509765625 }, { "auxiliary_loss_clip": 0.0112722, "auxiliary_loss_mlp": 0.01024775, "balance_loss_clip": 1.03945291, "balance_loss_mlp": 1.01758981, "epoch": 0.7532014669632658, "flos": 27748251702720.0, "grad_norm": 1.5296258029359144, "language_loss": 0.69045025, "learning_rate": 6.05474969105289e-07, "loss": 0.71197021, "num_input_tokens_seen": 134683025, "step": 6264, "time_per_iteration": 2.635950803756714 }, { "auxiliary_loss_clip": 0.01158017, "auxiliary_loss_mlp": 0.01028771, "balance_loss_clip": 1.0480181, "balance_loss_mlp": 1.02109694, "epoch": 0.7533217098539049, "flos": 14137424283840.0, "grad_norm": 2.093012248255477, "language_loss": 0.73573756, "learning_rate": 6.049166956582725e-07, "loss": 0.75760543, "num_input_tokens_seen": 134701290, "step": 6265, "time_per_iteration": 2.4807686805725098 }, { "auxiliary_loss_clip": 0.0115317, "auxiliary_loss_mlp": 0.01026527, "balance_loss_clip": 1.04663658, "balance_loss_mlp": 1.01956773, "epoch": 0.753441952744544, "flos": 26429072156640.0, "grad_norm": 1.915505869924376, "language_loss": 0.87704444, "learning_rate": 6.043586338442841e-07, "loss": 0.8988415, "num_input_tokens_seen": 134720345, "step": 6266, "time_per_iteration": 2.565920352935791 }, { "auxiliary_loss_clip": 0.01164829, "auxiliary_loss_mlp": 0.01023775, "balance_loss_clip": 1.04727221, "balance_loss_mlp": 1.01774621, "epoch": 0.7535621956351831, "flos": 23878625005920.0, "grad_norm": 1.5753317661180808, "language_loss": 0.73115826, "learning_rate": 6.038007837479815e-07, "loss": 0.75304431, "num_input_tokens_seen": 134741450, "step": 6267, "time_per_iteration": 2.489628791809082 }, { "auxiliary_loss_clip": 0.01153311, "auxiliary_loss_mlp": 0.01023402, "balance_loss_clip": 1.04703248, "balance_loss_mlp": 1.01725054, "epoch": 0.7536824385258222, "flos": 21795877294080.0, "grad_norm": 8.662849109101385, "language_loss": 0.64135766, "learning_rate": 6.032431454539897e-07, "loss": 0.6631248, "num_input_tokens_seen": 134760295, "step": 6268, "time_per_iteration": 2.4902374744415283 }, { "auxiliary_loss_clip": 0.01129044, "auxiliary_loss_mlp": 0.0102523, "balance_loss_clip": 1.04484963, "balance_loss_mlp": 1.01852417, "epoch": 0.7538026814164612, "flos": 28911649029600.0, "grad_norm": 1.6989003539676744, "language_loss": 0.81509101, "learning_rate": 6.026857190469014e-07, "loss": 0.83663374, "num_input_tokens_seen": 134782050, "step": 6269, "time_per_iteration": 2.5908255577087402 }, { "auxiliary_loss_clip": 0.01143492, "auxiliary_loss_mlp": 0.01029959, "balance_loss_clip": 1.04518807, "balance_loss_mlp": 1.0227375, "epoch": 0.7539229243071004, "flos": 21104704943040.0, "grad_norm": 1.895343682833951, "language_loss": 0.73741192, "learning_rate": 6.0212850461128e-07, "loss": 0.75914639, "num_input_tokens_seen": 134801170, "step": 6270, "time_per_iteration": 2.523606061935425 }, { "auxiliary_loss_clip": 0.01143791, "auxiliary_loss_mlp": 0.01021616, "balance_loss_clip": 1.04413426, "balance_loss_mlp": 1.0142014, "epoch": 0.7540431671977395, "flos": 15158472334560.0, "grad_norm": 3.074623865874634, "language_loss": 0.74331927, "learning_rate": 6.015715022316516e-07, "loss": 0.76497328, "num_input_tokens_seen": 134819150, "step": 6271, "time_per_iteration": 2.483809232711792 }, { "auxiliary_loss_clip": 0.0111644, "auxiliary_loss_mlp": 0.01019856, "balance_loss_clip": 1.04273891, "balance_loss_mlp": 1.01256967, "epoch": 0.7541634100883785, "flos": 18770583229440.0, "grad_norm": 2.7255135086954247, "language_loss": 0.78037, "learning_rate": 6.010147119925154e-07, "loss": 0.80173296, "num_input_tokens_seen": 134836905, "step": 6272, "time_per_iteration": 2.5596649646759033 }, { "auxiliary_loss_clip": 0.01120341, "auxiliary_loss_mlp": 0.01030835, "balance_loss_clip": 1.04362607, "balance_loss_mlp": 1.02376008, "epoch": 0.7542836529790176, "flos": 20594773547520.0, "grad_norm": 2.0910544287081128, "language_loss": 0.66452426, "learning_rate": 6.004581339783348e-07, "loss": 0.68603605, "num_input_tokens_seen": 134855225, "step": 6273, "time_per_iteration": 2.534113883972168 }, { "auxiliary_loss_clip": 0.01159382, "auxiliary_loss_mlp": 0.01026037, "balance_loss_clip": 1.0470078, "balance_loss_mlp": 1.01845813, "epoch": 0.7544038958696567, "flos": 19095107302080.0, "grad_norm": 2.3745914970831494, "language_loss": 0.68571836, "learning_rate": 5.999017682735425e-07, "loss": 0.70757258, "num_input_tokens_seen": 134871615, "step": 6274, "time_per_iteration": 3.2884681224823 }, { "auxiliary_loss_clip": 0.0110688, "auxiliary_loss_mlp": 0.01024741, "balance_loss_clip": 1.04229164, "balance_loss_mlp": 1.01765394, "epoch": 0.7545241387602958, "flos": 31723311429120.0, "grad_norm": 2.156596625180831, "language_loss": 0.66550708, "learning_rate": 5.993456149625387e-07, "loss": 0.68682325, "num_input_tokens_seen": 134892765, "step": 6275, "time_per_iteration": 2.6795458793640137 }, { "auxiliary_loss_clip": 0.01116344, "auxiliary_loss_mlp": 0.01024935, "balance_loss_clip": 1.04143214, "balance_loss_mlp": 1.01819038, "epoch": 0.7546443816509348, "flos": 20296498384320.0, "grad_norm": 1.702738794170383, "language_loss": 0.82250941, "learning_rate": 5.987896741296909e-07, "loss": 0.8439222, "num_input_tokens_seen": 134910505, "step": 6276, "time_per_iteration": 2.5350403785705566 }, { "auxiliary_loss_clip": 0.01140834, "auxiliary_loss_mlp": 0.01028163, "balance_loss_clip": 1.04598975, "balance_loss_mlp": 1.02113843, "epoch": 0.754764624541574, "flos": 23696162873760.0, "grad_norm": 2.3364375914340054, "language_loss": 0.78189379, "learning_rate": 5.982339458593361e-07, "loss": 0.80358374, "num_input_tokens_seen": 134930445, "step": 6277, "time_per_iteration": 3.338660955429077 }, { "auxiliary_loss_clip": 0.01150567, "auxiliary_loss_mlp": 0.0076145, "balance_loss_clip": 1.04526341, "balance_loss_mlp": 1.00053799, "epoch": 0.7548848674322131, "flos": 25337208637440.0, "grad_norm": 1.5357220495704624, "language_loss": 0.83884561, "learning_rate": 5.976784302357767e-07, "loss": 0.85796577, "num_input_tokens_seen": 134951010, "step": 6278, "time_per_iteration": 2.5279252529144287 }, { "auxiliary_loss_clip": 0.01159475, "auxiliary_loss_mlp": 0.01028564, "balance_loss_clip": 1.04801047, "balance_loss_mlp": 1.02170706, "epoch": 0.7550051103228521, "flos": 19573150825440.0, "grad_norm": 1.760398557309218, "language_loss": 0.73098195, "learning_rate": 5.971231273432855e-07, "loss": 0.75286233, "num_input_tokens_seen": 134970495, "step": 6279, "time_per_iteration": 2.4933128356933594 }, { "auxiliary_loss_clip": 0.01055308, "auxiliary_loss_mlp": 0.01000364, "balance_loss_clip": 1.00805831, "balance_loss_mlp": 0.99910086, "epoch": 0.7551253532134913, "flos": 64150078809600.0, "grad_norm": 0.8103742845257487, "language_loss": 0.54612184, "learning_rate": 5.965680372661e-07, "loss": 0.56667852, "num_input_tokens_seen": 135028060, "step": 6280, "time_per_iteration": 3.007462501525879 }, { "auxiliary_loss_clip": 0.01143046, "auxiliary_loss_mlp": 0.01022819, "balance_loss_clip": 1.04687047, "balance_loss_mlp": 1.01656055, "epoch": 0.7552455961041303, "flos": 26067991007040.0, "grad_norm": 1.9449120487557938, "language_loss": 0.56788194, "learning_rate": 5.960131600884266e-07, "loss": 0.5895406, "num_input_tokens_seen": 135047330, "step": 6281, "time_per_iteration": 2.5679702758789062 }, { "auxiliary_loss_clip": 0.01128039, "auxiliary_loss_mlp": 0.01022037, "balance_loss_clip": 1.04312027, "balance_loss_mlp": 1.01570153, "epoch": 0.7553658389947694, "flos": 24498227632320.0, "grad_norm": 1.7852510692178574, "language_loss": 0.76115882, "learning_rate": 5.954584958944413e-07, "loss": 0.78265965, "num_input_tokens_seen": 135065995, "step": 6282, "time_per_iteration": 2.5770468711853027 }, { "auxiliary_loss_clip": 0.01128227, "auxiliary_loss_mlp": 0.00761652, "balance_loss_clip": 1.04187965, "balance_loss_mlp": 1.0006026, "epoch": 0.7554860818854086, "flos": 21799468990080.0, "grad_norm": 1.9296253348196062, "language_loss": 0.81640267, "learning_rate": 5.949040447682854e-07, "loss": 0.83530152, "num_input_tokens_seen": 135085820, "step": 6283, "time_per_iteration": 2.6094863414764404 }, { "auxiliary_loss_clip": 0.01144336, "auxiliary_loss_mlp": 0.01029295, "balance_loss_clip": 1.04401803, "balance_loss_mlp": 1.02195144, "epoch": 0.7556063247760476, "flos": 16362126185280.0, "grad_norm": 3.0953467162812496, "language_loss": 0.68490946, "learning_rate": 5.943498067940686e-07, "loss": 0.70664579, "num_input_tokens_seen": 135102845, "step": 6284, "time_per_iteration": 2.4930918216705322 }, { "auxiliary_loss_clip": 0.01138821, "auxiliary_loss_mlp": 0.01023733, "balance_loss_clip": 1.05000806, "balance_loss_mlp": 1.01702785, "epoch": 0.7557265676666867, "flos": 27235159614720.0, "grad_norm": 1.7984067645357646, "language_loss": 0.81487983, "learning_rate": 5.937957820558686e-07, "loss": 0.83650541, "num_input_tokens_seen": 135122190, "step": 6285, "time_per_iteration": 2.587841033935547 }, { "auxiliary_loss_clip": 0.01044043, "auxiliary_loss_mlp": 0.00999688, "balance_loss_clip": 1.00888777, "balance_loss_mlp": 0.99840015, "epoch": 0.7558468105573258, "flos": 62189142261120.0, "grad_norm": 0.8522420989276852, "language_loss": 0.65448093, "learning_rate": 5.932419706377296e-07, "loss": 0.67491829, "num_input_tokens_seen": 135180495, "step": 6286, "time_per_iteration": 3.0869290828704834 }, { "auxiliary_loss_clip": 0.01126731, "auxiliary_loss_mlp": 0.01020637, "balance_loss_clip": 1.04763484, "balance_loss_mlp": 1.01408648, "epoch": 0.7559670534479649, "flos": 33249082916160.0, "grad_norm": 1.844717613625305, "language_loss": 0.73883402, "learning_rate": 5.92688372623666e-07, "loss": 0.76030767, "num_input_tokens_seen": 135199200, "step": 6287, "time_per_iteration": 3.3779830932617188 }, { "auxiliary_loss_clip": 0.0115642, "auxiliary_loss_mlp": 0.01029269, "balance_loss_clip": 1.04568601, "balance_loss_mlp": 1.02173519, "epoch": 0.7560872963386039, "flos": 14064381963840.0, "grad_norm": 2.088569967779996, "language_loss": 0.73814368, "learning_rate": 5.921349880976574e-07, "loss": 0.76000059, "num_input_tokens_seen": 135217035, "step": 6288, "time_per_iteration": 2.4774727821350098 }, { "auxiliary_loss_clip": 0.01144874, "auxiliary_loss_mlp": 0.00761868, "balance_loss_clip": 1.04389369, "balance_loss_mlp": 1.00056434, "epoch": 0.7562075392292431, "flos": 20412311415360.0, "grad_norm": 3.5768865199884265, "language_loss": 0.81558341, "learning_rate": 5.915818171436515e-07, "loss": 0.83465081, "num_input_tokens_seen": 135236370, "step": 6289, "time_per_iteration": 2.513093948364258 }, { "auxiliary_loss_clip": 0.01139763, "auxiliary_loss_mlp": 0.01024587, "balance_loss_clip": 1.04122496, "balance_loss_mlp": 1.01740515, "epoch": 0.7563277821198822, "flos": 20376795930240.0, "grad_norm": 1.8264640325895558, "language_loss": 0.74722767, "learning_rate": 5.910288598455642e-07, "loss": 0.76887113, "num_input_tokens_seen": 135255720, "step": 6290, "time_per_iteration": 2.522231340408325 }, { "auxiliary_loss_clip": 0.01161226, "auxiliary_loss_mlp": 0.01024494, "balance_loss_clip": 1.04616547, "balance_loss_mlp": 1.01665616, "epoch": 0.7564480250105212, "flos": 18588264765120.0, "grad_norm": 3.2538851146336047, "language_loss": 0.74060512, "learning_rate": 5.90476116287278e-07, "loss": 0.76246238, "num_input_tokens_seen": 135273320, "step": 6291, "time_per_iteration": 2.4420132637023926 }, { "auxiliary_loss_clip": 0.01140838, "auxiliary_loss_mlp": 0.01022406, "balance_loss_clip": 1.04629517, "balance_loss_mlp": 1.01551259, "epoch": 0.7565682679011604, "flos": 21215525516640.0, "grad_norm": 1.973926180105223, "language_loss": 0.68069363, "learning_rate": 5.899235865526456e-07, "loss": 0.70232612, "num_input_tokens_seen": 135292615, "step": 6292, "time_per_iteration": 2.5243935585021973 }, { "auxiliary_loss_clip": 0.01119474, "auxiliary_loss_mlp": 0.01023031, "balance_loss_clip": 1.04313374, "balance_loss_mlp": 1.01657033, "epoch": 0.7566885107917994, "flos": 20449012160160.0, "grad_norm": 1.6542093507179825, "language_loss": 0.82228857, "learning_rate": 5.893712707254825e-07, "loss": 0.84371364, "num_input_tokens_seen": 135310075, "step": 6293, "time_per_iteration": 2.5332841873168945 }, { "auxiliary_loss_clip": 0.01107291, "auxiliary_loss_mlp": 0.01025167, "balance_loss_clip": 1.03954291, "balance_loss_mlp": 1.01797557, "epoch": 0.7568087536824385, "flos": 19025836262880.0, "grad_norm": 2.5408941197369908, "language_loss": 0.65809989, "learning_rate": 5.888191688895769e-07, "loss": 0.67942452, "num_input_tokens_seen": 135327335, "step": 6294, "time_per_iteration": 2.559194564819336 }, { "auxiliary_loss_clip": 0.01169669, "auxiliary_loss_mlp": 0.01025091, "balance_loss_clip": 1.04648173, "balance_loss_mlp": 1.01735163, "epoch": 0.7569289965730777, "flos": 15225444688320.0, "grad_norm": 3.210116685937714, "language_loss": 0.62698162, "learning_rate": 5.882672811286813e-07, "loss": 0.64892924, "num_input_tokens_seen": 135343615, "step": 6295, "time_per_iteration": 2.4091601371765137 }, { "auxiliary_loss_clip": 0.01172863, "auxiliary_loss_mlp": 0.01023256, "balance_loss_clip": 1.04908681, "balance_loss_mlp": 1.0162642, "epoch": 0.7570492394637167, "flos": 20769369865440.0, "grad_norm": 1.8786263549340325, "language_loss": 0.6956144, "learning_rate": 5.877156075265166e-07, "loss": 0.71757555, "num_input_tokens_seen": 135359880, "step": 6296, "time_per_iteration": 2.4495296478271484 }, { "auxiliary_loss_clip": 0.01138835, "auxiliary_loss_mlp": 0.01025265, "balance_loss_clip": 1.04261172, "balance_loss_mlp": 1.01783252, "epoch": 0.7571694823543558, "flos": 15664093694880.0, "grad_norm": 3.2590404834840347, "language_loss": 0.69614589, "learning_rate": 5.871641481667715e-07, "loss": 0.71778691, "num_input_tokens_seen": 135374325, "step": 6297, "time_per_iteration": 2.481686592102051 }, { "auxiliary_loss_clip": 0.01116191, "auxiliary_loss_mlp": 0.01024993, "balance_loss_clip": 1.04202986, "balance_loss_mlp": 1.01772976, "epoch": 0.7572897252449949, "flos": 25409245282560.0, "grad_norm": 1.668618758688523, "language_loss": 0.84625602, "learning_rate": 5.866129031331011e-07, "loss": 0.86766779, "num_input_tokens_seen": 135393980, "step": 6298, "time_per_iteration": 2.63659405708313 }, { "auxiliary_loss_clip": 0.01144114, "auxiliary_loss_mlp": 0.01025715, "balance_loss_clip": 1.04522443, "balance_loss_mlp": 1.01870215, "epoch": 0.757409968135634, "flos": 24279352091040.0, "grad_norm": 2.0332413722424896, "language_loss": 0.83315909, "learning_rate": 5.8606187250913e-07, "loss": 0.85485744, "num_input_tokens_seen": 135412030, "step": 6299, "time_per_iteration": 2.5362300872802734 }, { "auxiliary_loss_clip": 0.01156244, "auxiliary_loss_mlp": 0.00762474, "balance_loss_clip": 1.04846787, "balance_loss_mlp": 1.00064516, "epoch": 0.757530211026273, "flos": 24133770288480.0, "grad_norm": 1.7068712979337786, "language_loss": 0.840698, "learning_rate": 5.855110563784482e-07, "loss": 0.85988522, "num_input_tokens_seen": 135430565, "step": 6300, "time_per_iteration": 3.299657106399536 }, { "auxiliary_loss_clip": 0.01149073, "auxiliary_loss_mlp": 0.0076165, "balance_loss_clip": 1.04528391, "balance_loss_mlp": 1.00061226, "epoch": 0.7576504539169122, "flos": 23951810993760.0, "grad_norm": 1.5818691091433055, "language_loss": 0.64110428, "learning_rate": 5.849604548246156e-07, "loss": 0.6602115, "num_input_tokens_seen": 135451675, "step": 6301, "time_per_iteration": 2.5279910564422607 }, { "auxiliary_loss_clip": 0.01150073, "auxiliary_loss_mlp": 0.00762238, "balance_loss_clip": 1.04965758, "balance_loss_mlp": 1.00055611, "epoch": 0.7577706968075513, "flos": 21251364254400.0, "grad_norm": 2.0543244979820763, "language_loss": 0.80083287, "learning_rate": 5.844100679311565e-07, "loss": 0.81995595, "num_input_tokens_seen": 135470635, "step": 6302, "time_per_iteration": 2.529287576675415 }, { "auxiliary_loss_clip": 0.01142355, "auxiliary_loss_mlp": 0.01023894, "balance_loss_clip": 1.04669428, "balance_loss_mlp": 1.01695919, "epoch": 0.7578909396981903, "flos": 18296598322560.0, "grad_norm": 2.2998973130620866, "language_loss": 0.75866359, "learning_rate": 5.838598957815637e-07, "loss": 0.78032607, "num_input_tokens_seen": 135487865, "step": 6303, "time_per_iteration": 3.9575767517089844 }, { "auxiliary_loss_clip": 0.01133991, "auxiliary_loss_mlp": 0.01024864, "balance_loss_clip": 1.04372716, "balance_loss_mlp": 1.01796436, "epoch": 0.7580111825888295, "flos": 25373873465280.0, "grad_norm": 1.4941884903177223, "language_loss": 0.85538054, "learning_rate": 5.833099384592996e-07, "loss": 0.87696916, "num_input_tokens_seen": 135508440, "step": 6304, "time_per_iteration": 2.5262303352355957 }, { "auxiliary_loss_clip": 0.01136128, "auxiliary_loss_mlp": 0.0102541, "balance_loss_clip": 1.04378831, "balance_loss_mlp": 1.01848435, "epoch": 0.7581314254794685, "flos": 23768666439360.0, "grad_norm": 2.0172725707553933, "language_loss": 0.71513683, "learning_rate": 5.827601960477913e-07, "loss": 0.73675215, "num_input_tokens_seen": 135526365, "step": 6305, "time_per_iteration": 2.533128499984741 }, { "auxiliary_loss_clip": 0.0115198, "auxiliary_loss_mlp": 0.01023502, "balance_loss_clip": 1.04442692, "balance_loss_mlp": 1.01713061, "epoch": 0.7582516683701076, "flos": 22054614272640.0, "grad_norm": 1.7593207290770458, "language_loss": 0.70359325, "learning_rate": 5.822106686304344e-07, "loss": 0.72534811, "num_input_tokens_seen": 135545655, "step": 6306, "time_per_iteration": 2.46751070022583 }, { "auxiliary_loss_clip": 0.01133181, "auxiliary_loss_mlp": 0.01028625, "balance_loss_clip": 1.04349601, "balance_loss_mlp": 1.0218544, "epoch": 0.7583719112607467, "flos": 31649730354720.0, "grad_norm": 1.752432743841298, "language_loss": 0.57970732, "learning_rate": 5.816613562905919e-07, "loss": 0.60132539, "num_input_tokens_seen": 135566840, "step": 6307, "time_per_iteration": 2.622584581375122 }, { "auxiliary_loss_clip": 0.01124115, "auxiliary_loss_mlp": 0.01025196, "balance_loss_clip": 1.04725659, "balance_loss_mlp": 1.01809728, "epoch": 0.7584921541513858, "flos": 33068380715040.0, "grad_norm": 1.5720458772092158, "language_loss": 0.6993041, "learning_rate": 5.811122591115933e-07, "loss": 0.7207973, "num_input_tokens_seen": 135587825, "step": 6308, "time_per_iteration": 2.6388845443725586 }, { "auxiliary_loss_clip": 0.0112895, "auxiliary_loss_mlp": 0.01027756, "balance_loss_clip": 1.04844403, "balance_loss_mlp": 1.02020407, "epoch": 0.7586123970420249, "flos": 23326353902880.0, "grad_norm": 2.5652372958083958, "language_loss": 0.71618032, "learning_rate": 5.805633771767376e-07, "loss": 0.73774743, "num_input_tokens_seen": 135605220, "step": 6309, "time_per_iteration": 2.572148561477661 }, { "auxiliary_loss_clip": 0.01135715, "auxiliary_loss_mlp": 0.01026045, "balance_loss_clip": 1.04654622, "balance_loss_mlp": 1.01964331, "epoch": 0.7587326399326639, "flos": 18334232908320.0, "grad_norm": 1.6590539830985567, "language_loss": 0.77774858, "learning_rate": 5.800147105692888e-07, "loss": 0.79936624, "num_input_tokens_seen": 135624795, "step": 6310, "time_per_iteration": 2.518991470336914 }, { "auxiliary_loss_clip": 0.01156024, "auxiliary_loss_mlp": 0.01026428, "balance_loss_clip": 1.04488897, "balance_loss_mlp": 1.01945138, "epoch": 0.7588528828233031, "flos": 17275083351360.0, "grad_norm": 1.910721094002749, "language_loss": 0.79215598, "learning_rate": 5.794662593724795e-07, "loss": 0.81398046, "num_input_tokens_seen": 135643800, "step": 6311, "time_per_iteration": 2.484218120574951 }, { "auxiliary_loss_clip": 0.01171751, "auxiliary_loss_mlp": 0.01029145, "balance_loss_clip": 1.05050278, "balance_loss_mlp": 1.02126575, "epoch": 0.7589731257139422, "flos": 17713624607040.0, "grad_norm": 2.0642355600615794, "language_loss": 0.75077772, "learning_rate": 5.789180236695091e-07, "loss": 0.77278668, "num_input_tokens_seen": 135660655, "step": 6312, "time_per_iteration": 2.442631244659424 }, { "auxiliary_loss_clip": 0.0114993, "auxiliary_loss_mlp": 0.01025335, "balance_loss_clip": 1.04645717, "balance_loss_mlp": 1.01909482, "epoch": 0.7590933686045812, "flos": 15961075847520.0, "grad_norm": 1.826028300126172, "language_loss": 0.84994698, "learning_rate": 5.78370003543544e-07, "loss": 0.87169963, "num_input_tokens_seen": 135679410, "step": 6313, "time_per_iteration": 3.209164619445801 }, { "auxiliary_loss_clip": 0.01155871, "auxiliary_loss_mlp": 0.00761911, "balance_loss_clip": 1.04602242, "balance_loss_mlp": 1.0005517, "epoch": 0.7592136114952204, "flos": 21068076032160.0, "grad_norm": 1.928631471488619, "language_loss": 0.83715093, "learning_rate": 5.778221990777203e-07, "loss": 0.85632873, "num_input_tokens_seen": 135697150, "step": 6314, "time_per_iteration": 2.470984935760498 }, { "auxiliary_loss_clip": 0.01146263, "auxiliary_loss_mlp": 0.0102628, "balance_loss_clip": 1.04835963, "balance_loss_mlp": 1.01916599, "epoch": 0.7593338543858594, "flos": 25297670452800.0, "grad_norm": 1.974480749137742, "language_loss": 0.8272711, "learning_rate": 5.772746103551372e-07, "loss": 0.84899652, "num_input_tokens_seen": 135712545, "step": 6315, "time_per_iteration": 2.5228943824768066 }, { "auxiliary_loss_clip": 0.01137922, "auxiliary_loss_mlp": 0.01023387, "balance_loss_clip": 1.04522681, "balance_loss_mlp": 1.01646066, "epoch": 0.7594540972764985, "flos": 31832372071680.0, "grad_norm": 3.6331750584001714, "language_loss": 0.72279304, "learning_rate": 5.767272374588648e-07, "loss": 0.74440616, "num_input_tokens_seen": 135733950, "step": 6316, "time_per_iteration": 2.5807156562805176 }, { "auxiliary_loss_clip": 0.01156051, "auxiliary_loss_mlp": 0.01029357, "balance_loss_clip": 1.04928946, "balance_loss_mlp": 1.02270818, "epoch": 0.7595743401671377, "flos": 37597255973760.0, "grad_norm": 1.576393416161843, "language_loss": 0.77826864, "learning_rate": 5.76180080471939e-07, "loss": 0.80012274, "num_input_tokens_seen": 135757120, "step": 6317, "time_per_iteration": 2.6345794200897217 }, { "auxiliary_loss_clip": 0.0117447, "auxiliary_loss_mlp": 0.01029517, "balance_loss_clip": 1.04923534, "balance_loss_mlp": 1.02161932, "epoch": 0.7596945830577767, "flos": 18287726833440.0, "grad_norm": 1.9281849305760075, "language_loss": 0.72147691, "learning_rate": 5.756331394773631e-07, "loss": 0.74351674, "num_input_tokens_seen": 135773335, "step": 6318, "time_per_iteration": 2.411121368408203 }, { "auxiliary_loss_clip": 0.01100479, "auxiliary_loss_mlp": 0.00762247, "balance_loss_clip": 1.04149985, "balance_loss_mlp": 1.00053906, "epoch": 0.7598148259484158, "flos": 22233125539200.0, "grad_norm": 1.9467875977733278, "language_loss": 0.76138604, "learning_rate": 5.750864145581071e-07, "loss": 0.78001332, "num_input_tokens_seen": 135792555, "step": 6319, "time_per_iteration": 2.61428165435791 }, { "auxiliary_loss_clip": 0.01170927, "auxiliary_loss_mlp": 0.01025453, "balance_loss_clip": 1.04937482, "balance_loss_mlp": 1.01859236, "epoch": 0.7599350688390549, "flos": 27161722208160.0, "grad_norm": 1.7734894979324871, "language_loss": 0.86232913, "learning_rate": 5.745399057971085e-07, "loss": 0.88429296, "num_input_tokens_seen": 135813690, "step": 6320, "time_per_iteration": 2.4927899837493896 }, { "auxiliary_loss_clip": 0.01160159, "auxiliary_loss_mlp": 0.01024255, "balance_loss_clip": 1.0476861, "balance_loss_mlp": 1.01672673, "epoch": 0.760055311729694, "flos": 15560707932000.0, "grad_norm": 2.1288129560677422, "language_loss": 0.75520307, "learning_rate": 5.739936132772738e-07, "loss": 0.77704716, "num_input_tokens_seen": 135832255, "step": 6321, "time_per_iteration": 2.488921642303467 }, { "auxiliary_loss_clip": 0.01169061, "auxiliary_loss_mlp": 0.01026922, "balance_loss_clip": 1.04768896, "balance_loss_mlp": 1.01951337, "epoch": 0.760175554620333, "flos": 25155500761440.0, "grad_norm": 2.3938655288324244, "language_loss": 0.74459326, "learning_rate": 5.734475370814733e-07, "loss": 0.7665531, "num_input_tokens_seen": 135851935, "step": 6322, "time_per_iteration": 2.4774699211120605 }, { "auxiliary_loss_clip": 0.01158311, "auxiliary_loss_mlp": 0.01023638, "balance_loss_clip": 1.0461781, "balance_loss_mlp": 1.01677406, "epoch": 0.7602957975109722, "flos": 24353795172480.0, "grad_norm": 1.6497846952806032, "language_loss": 0.78601021, "learning_rate": 5.729016772925483e-07, "loss": 0.80782974, "num_input_tokens_seen": 135873510, "step": 6323, "time_per_iteration": 2.5304737091064453 }, { "auxiliary_loss_clip": 0.01111061, "auxiliary_loss_mlp": 0.01025652, "balance_loss_clip": 1.04416931, "balance_loss_mlp": 1.0182966, "epoch": 0.7604160404016113, "flos": 25192668426720.0, "grad_norm": 2.0831447979228845, "language_loss": 0.70645332, "learning_rate": 5.723560339933038e-07, "loss": 0.72782046, "num_input_tokens_seen": 135893845, "step": 6324, "time_per_iteration": 2.6084342002868652 }, { "auxiliary_loss_clip": 0.01152755, "auxiliary_loss_mlp": 0.00762167, "balance_loss_clip": 1.04455578, "balance_loss_mlp": 1.00057173, "epoch": 0.7605362832922503, "flos": 29861845694880.0, "grad_norm": 2.1312861460361603, "language_loss": 0.65135372, "learning_rate": 5.71810607266513e-07, "loss": 0.67050296, "num_input_tokens_seen": 135912430, "step": 6325, "time_per_iteration": 2.5399417877197266 }, { "auxiliary_loss_clip": 0.0115707, "auxiliary_loss_mlp": 0.01029663, "balance_loss_clip": 1.04619169, "balance_loss_mlp": 1.02251661, "epoch": 0.7606565261828895, "flos": 13917938154240.0, "grad_norm": 1.9202855852547875, "language_loss": 0.60257065, "learning_rate": 5.712653971949184e-07, "loss": 0.62443793, "num_input_tokens_seen": 135930550, "step": 6326, "time_per_iteration": 3.2134487628936768 }, { "auxiliary_loss_clip": 0.01149822, "auxiliary_loss_mlp": 0.0102676, "balance_loss_clip": 1.04461098, "balance_loss_mlp": 1.01933312, "epoch": 0.7607767690735285, "flos": 18551276684640.0, "grad_norm": 2.308730399557653, "language_loss": 0.75010681, "learning_rate": 5.707204038612268e-07, "loss": 0.77187264, "num_input_tokens_seen": 135947980, "step": 6327, "time_per_iteration": 2.4357075691223145 }, { "auxiliary_loss_clip": 0.01152345, "auxiliary_loss_mlp": 0.01029999, "balance_loss_clip": 1.05236673, "balance_loss_mlp": 1.02146959, "epoch": 0.7608970119641676, "flos": 20922997067040.0, "grad_norm": 2.1823939645840547, "language_loss": 0.73704535, "learning_rate": 5.701756273481138e-07, "loss": 0.75886881, "num_input_tokens_seen": 135965400, "step": 6328, "time_per_iteration": 2.511087656021118 }, { "auxiliary_loss_clip": 0.01143806, "auxiliary_loss_mlp": 0.01021464, "balance_loss_clip": 1.04641533, "balance_loss_mlp": 1.01484537, "epoch": 0.7610172548548068, "flos": 23807306700000.0, "grad_norm": 1.5209417051885414, "language_loss": 0.73980027, "learning_rate": 5.696310677382212e-07, "loss": 0.76145291, "num_input_tokens_seen": 135986795, "step": 6329, "time_per_iteration": 3.332432270050049 }, { "auxiliary_loss_clip": 0.01036142, "auxiliary_loss_mlp": 0.01002175, "balance_loss_clip": 1.01233292, "balance_loss_mlp": 1.00101304, "epoch": 0.7611374977454458, "flos": 66496591874400.0, "grad_norm": 0.8587984398159536, "language_loss": 0.6178683, "learning_rate": 5.690867251141576e-07, "loss": 0.63825154, "num_input_tokens_seen": 136053450, "step": 6330, "time_per_iteration": 3.2795119285583496 }, { "auxiliary_loss_clip": 0.01159403, "auxiliary_loss_mlp": 0.01024861, "balance_loss_clip": 1.04699171, "balance_loss_mlp": 1.01808441, "epoch": 0.7612577406360849, "flos": 15633139663680.0, "grad_norm": 2.575538714696992, "language_loss": 0.91319263, "learning_rate": 5.685425995585013e-07, "loss": 0.93503523, "num_input_tokens_seen": 136071375, "step": 6331, "time_per_iteration": 2.460057497024536 }, { "auxiliary_loss_clip": 0.01045593, "auxiliary_loss_mlp": 0.01001382, "balance_loss_clip": 1.00715792, "balance_loss_mlp": 1.00005233, "epoch": 0.761377983526724, "flos": 60526265374080.0, "grad_norm": 0.818459618824208, "language_loss": 0.59036291, "learning_rate": 5.679986911537935e-07, "loss": 0.61083263, "num_input_tokens_seen": 136138905, "step": 6332, "time_per_iteration": 3.3091659545898438 }, { "auxiliary_loss_clip": 0.01101773, "auxiliary_loss_mlp": 0.01021304, "balance_loss_clip": 1.04219866, "balance_loss_mlp": 1.01439619, "epoch": 0.7614982264173631, "flos": 35772526901280.0, "grad_norm": 1.7816569420339472, "language_loss": 0.67257237, "learning_rate": 5.674549999825462e-07, "loss": 0.69380319, "num_input_tokens_seen": 136161720, "step": 6333, "time_per_iteration": 2.6882383823394775 }, { "auxiliary_loss_clip": 0.01055976, "auxiliary_loss_mlp": 0.0100176, "balance_loss_clip": 1.00838923, "balance_loss_mlp": 1.00041926, "epoch": 0.7616184693080021, "flos": 67925514485280.0, "grad_norm": 0.9288968132126226, "language_loss": 0.71417093, "learning_rate": 5.669115261272363e-07, "loss": 0.7347483, "num_input_tokens_seen": 136222040, "step": 6334, "time_per_iteration": 3.0938961505889893 }, { "auxiliary_loss_clip": 0.0115699, "auxiliary_loss_mlp": 0.01019411, "balance_loss_clip": 1.0470258, "balance_loss_mlp": 1.01160324, "epoch": 0.7617387121986413, "flos": 20521982646240.0, "grad_norm": 8.267312362419155, "language_loss": 0.72611415, "learning_rate": 5.663682696703081e-07, "loss": 0.74787819, "num_input_tokens_seen": 136240305, "step": 6335, "time_per_iteration": 2.4819562435150146 }, { "auxiliary_loss_clip": 0.01169095, "auxiliary_loss_mlp": 0.01027535, "balance_loss_clip": 1.04777765, "balance_loss_mlp": 1.02087116, "epoch": 0.7618589550892804, "flos": 18624498589440.0, "grad_norm": 1.9957865241938857, "language_loss": 0.81753051, "learning_rate": 5.658252306941746e-07, "loss": 0.83949673, "num_input_tokens_seen": 136259625, "step": 6336, "time_per_iteration": 2.428985118865967 }, { "auxiliary_loss_clip": 0.01118417, "auxiliary_loss_mlp": 0.01026768, "balance_loss_clip": 1.04704404, "balance_loss_mlp": 1.01921618, "epoch": 0.7619791979799194, "flos": 17453738285760.0, "grad_norm": 3.1880433801435215, "language_loss": 0.75463581, "learning_rate": 5.65282409281212e-07, "loss": 0.7760877, "num_input_tokens_seen": 136277090, "step": 6337, "time_per_iteration": 2.5871469974517822 }, { "auxiliary_loss_clip": 0.01137819, "auxiliary_loss_mlp": 0.01030341, "balance_loss_clip": 1.04456306, "balance_loss_mlp": 1.02302134, "epoch": 0.7620994408705585, "flos": 14137424283840.0, "grad_norm": 2.2202867888454922, "language_loss": 0.70069265, "learning_rate": 5.64739805513768e-07, "loss": 0.7223742, "num_input_tokens_seen": 136294635, "step": 6338, "time_per_iteration": 2.4974539279937744 }, { "auxiliary_loss_clip": 0.01051735, "auxiliary_loss_mlp": 0.00752821, "balance_loss_clip": 1.00848317, "balance_loss_mlp": 1.00009394, "epoch": 0.7622196837611976, "flos": 70708802486400.0, "grad_norm": 0.7859786996720781, "language_loss": 0.55716127, "learning_rate": 5.641974194741541e-07, "loss": 0.57520682, "num_input_tokens_seen": 136350320, "step": 6339, "time_per_iteration": 3.728660821914673 }, { "auxiliary_loss_clip": 0.01062655, "auxiliary_loss_mlp": 0.01003447, "balance_loss_clip": 1.03839409, "balance_loss_mlp": 1.00214756, "epoch": 0.7623399266518367, "flos": 60684160305600.0, "grad_norm": 0.7851030490051804, "language_loss": 0.63794148, "learning_rate": 5.636552512446502e-07, "loss": 0.65860248, "num_input_tokens_seen": 136411375, "step": 6340, "time_per_iteration": 3.0697550773620605 }, { "auxiliary_loss_clip": 0.011517, "auxiliary_loss_mlp": 0.01024355, "balance_loss_clip": 1.04722953, "balance_loss_mlp": 1.01781046, "epoch": 0.7624601695424758, "flos": 26468897676960.0, "grad_norm": 1.6560303808363759, "language_loss": 0.78039169, "learning_rate": 5.631133009075027e-07, "loss": 0.80215228, "num_input_tokens_seen": 136430560, "step": 6341, "time_per_iteration": 2.5295610427856445 }, { "auxiliary_loss_clip": 0.01156527, "auxiliary_loss_mlp": 0.00761541, "balance_loss_clip": 1.04740524, "balance_loss_mlp": 1.00053239, "epoch": 0.7625804124331149, "flos": 19135758912480.0, "grad_norm": 1.8254446221980947, "language_loss": 0.68393242, "learning_rate": 5.625715685449242e-07, "loss": 0.70311308, "num_input_tokens_seen": 136448665, "step": 6342, "time_per_iteration": 2.512758493423462 }, { "auxiliary_loss_clip": 0.01124858, "auxiliary_loss_mlp": 0.01024934, "balance_loss_clip": 1.04888558, "balance_loss_mlp": 1.01778424, "epoch": 0.762700655323754, "flos": 26213105889120.0, "grad_norm": 1.57470372257277, "language_loss": 0.71895432, "learning_rate": 5.620300542390966e-07, "loss": 0.74045229, "num_input_tokens_seen": 136469710, "step": 6343, "time_per_iteration": 2.5956056118011475 }, { "auxiliary_loss_clip": 0.01132553, "auxiliary_loss_mlp": 0.01025058, "balance_loss_clip": 1.04245055, "balance_loss_mlp": 1.01850152, "epoch": 0.762820898214393, "flos": 22382586373440.0, "grad_norm": 1.7052663517151625, "language_loss": 0.85061562, "learning_rate": 5.614887580721659e-07, "loss": 0.87219179, "num_input_tokens_seen": 136489855, "step": 6344, "time_per_iteration": 2.533419609069824 }, { "auxiliary_loss_clip": 0.01123402, "auxiliary_loss_mlp": 0.01026089, "balance_loss_clip": 1.04722381, "balance_loss_mlp": 1.018713, "epoch": 0.7629411411050322, "flos": 15700507104000.0, "grad_norm": 1.9890296527573228, "language_loss": 0.73688483, "learning_rate": 5.609476801262481e-07, "loss": 0.75837976, "num_input_tokens_seen": 136504715, "step": 6345, "time_per_iteration": 2.4959073066711426 }, { "auxiliary_loss_clip": 0.01125484, "auxiliary_loss_mlp": 0.01022159, "balance_loss_clip": 1.04455328, "balance_loss_mlp": 1.01470232, "epoch": 0.7630613839956712, "flos": 13770345001920.0, "grad_norm": 2.3176578749098886, "language_loss": 0.63803744, "learning_rate": 5.604068204834223e-07, "loss": 0.65951389, "num_input_tokens_seen": 136521610, "step": 6346, "time_per_iteration": 2.5166263580322266 }, { "auxiliary_loss_clip": 0.01111934, "auxiliary_loss_mlp": 0.00762764, "balance_loss_clip": 1.04191709, "balance_loss_mlp": 1.00054467, "epoch": 0.7631816268863103, "flos": 14569572320640.0, "grad_norm": 2.263622997833177, "language_loss": 0.76572204, "learning_rate": 5.598661792257367e-07, "loss": 0.78446901, "num_input_tokens_seen": 136538655, "step": 6347, "time_per_iteration": 2.5434277057647705 }, { "auxiliary_loss_clip": 0.01153833, "auxiliary_loss_mlp": 0.01028494, "balance_loss_clip": 1.04526758, "balance_loss_mlp": 1.02172613, "epoch": 0.7633018697769495, "flos": 19062213755040.0, "grad_norm": 3.3745846867869087, "language_loss": 0.76172256, "learning_rate": 5.593257564352071e-07, "loss": 0.78354585, "num_input_tokens_seen": 136557095, "step": 6348, "time_per_iteration": 2.4815750122070312 }, { "auxiliary_loss_clip": 0.01153199, "auxiliary_loss_mlp": 0.01020178, "balance_loss_clip": 1.04650187, "balance_loss_mlp": 1.01349366, "epoch": 0.7634221126675885, "flos": 22052962092480.0, "grad_norm": 1.5230385766559424, "language_loss": 0.75570047, "learning_rate": 5.58785552193815e-07, "loss": 0.77743423, "num_input_tokens_seen": 136577340, "step": 6349, "time_per_iteration": 2.4929816722869873 }, { "auxiliary_loss_clip": 0.01168781, "auxiliary_loss_mlp": 0.01026042, "balance_loss_clip": 1.04728222, "balance_loss_mlp": 1.01944351, "epoch": 0.7635423555582276, "flos": 29382724662720.0, "grad_norm": 1.7870051883751885, "language_loss": 0.75415879, "learning_rate": 5.582455665835086e-07, "loss": 0.77610707, "num_input_tokens_seen": 136597635, "step": 6350, "time_per_iteration": 2.5198121070861816 }, { "auxiliary_loss_clip": 0.01149307, "auxiliary_loss_mlp": 0.01025649, "balance_loss_clip": 1.04425526, "balance_loss_mlp": 1.0179745, "epoch": 0.7636625984488667, "flos": 17784906996000.0, "grad_norm": 4.493900909997018, "language_loss": 0.7291075, "learning_rate": 5.577057996862036e-07, "loss": 0.75085711, "num_input_tokens_seen": 136615260, "step": 6351, "time_per_iteration": 2.470323085784912 }, { "auxiliary_loss_clip": 0.01166575, "auxiliary_loss_mlp": 0.01023004, "balance_loss_clip": 1.04849076, "balance_loss_mlp": 1.01639056, "epoch": 0.7637828413395058, "flos": 23734587632640.0, "grad_norm": 1.5294703782826995, "language_loss": 0.75891531, "learning_rate": 5.571662515837814e-07, "loss": 0.78081107, "num_input_tokens_seen": 136637220, "step": 6352, "time_per_iteration": 3.2300963401794434 }, { "auxiliary_loss_clip": 0.01142589, "auxiliary_loss_mlp": 0.0102322, "balance_loss_clip": 1.04769266, "balance_loss_mlp": 1.01641655, "epoch": 0.7639030842301449, "flos": 36283284386880.0, "grad_norm": 1.659125466042343, "language_loss": 0.83709937, "learning_rate": 5.566269223580926e-07, "loss": 0.8587575, "num_input_tokens_seen": 136658930, "step": 6353, "time_per_iteration": 2.658238410949707 }, { "auxiliary_loss_clip": 0.01156364, "auxiliary_loss_mlp": 0.0102409, "balance_loss_clip": 1.0469383, "balance_loss_mlp": 1.01741743, "epoch": 0.764023327120784, "flos": 28878108977280.0, "grad_norm": 1.7737669572458297, "language_loss": 0.75353169, "learning_rate": 5.560878120909511e-07, "loss": 0.77533621, "num_input_tokens_seen": 136681530, "step": 6354, "time_per_iteration": 2.55483341217041 }, { "auxiliary_loss_clip": 0.01054814, "auxiliary_loss_mlp": 0.0100408, "balance_loss_clip": 1.00708258, "balance_loss_mlp": 1.00274456, "epoch": 0.7641435700114231, "flos": 64789723099680.0, "grad_norm": 0.8498855458040481, "language_loss": 0.58615541, "learning_rate": 5.55548920864141e-07, "loss": 0.60674435, "num_input_tokens_seen": 136742185, "step": 6355, "time_per_iteration": 4.60629415512085 }, { "auxiliary_loss_clip": 0.01155034, "auxiliary_loss_mlp": 0.0103228, "balance_loss_clip": 1.04811585, "balance_loss_mlp": 1.02552056, "epoch": 0.7642638129020621, "flos": 16835787839520.0, "grad_norm": 1.6917620555399921, "language_loss": 0.77879775, "learning_rate": 5.550102487594113e-07, "loss": 0.80067086, "num_input_tokens_seen": 136760855, "step": 6356, "time_per_iteration": 2.4543871879577637 }, { "auxiliary_loss_clip": 0.01117372, "auxiliary_loss_mlp": 0.00761959, "balance_loss_clip": 1.04038656, "balance_loss_mlp": 1.00061321, "epoch": 0.7643840557927013, "flos": 30408944755680.0, "grad_norm": 1.501869018081072, "language_loss": 0.71641451, "learning_rate": 5.54471795858477e-07, "loss": 0.7352078, "num_input_tokens_seen": 136780925, "step": 6357, "time_per_iteration": 2.662152051925659 }, { "auxiliary_loss_clip": 0.01124958, "auxiliary_loss_mlp": 0.01022941, "balance_loss_clip": 1.03985119, "balance_loss_mlp": 1.01589227, "epoch": 0.7645042986833404, "flos": 16983237324000.0, "grad_norm": 2.186899106596666, "language_loss": 0.82982528, "learning_rate": 5.539335622430235e-07, "loss": 0.85130423, "num_input_tokens_seen": 136799545, "step": 6358, "time_per_iteration": 2.5263662338256836 }, { "auxiliary_loss_clip": 0.01146886, "auxiliary_loss_mlp": 0.01024108, "balance_loss_clip": 1.04185402, "balance_loss_mlp": 1.01731324, "epoch": 0.7646245415739794, "flos": 17311496760480.0, "grad_norm": 2.1113940438080157, "language_loss": 0.74635559, "learning_rate": 5.533955479946975e-07, "loss": 0.76806551, "num_input_tokens_seen": 136818325, "step": 6359, "time_per_iteration": 2.4669084548950195 }, { "auxiliary_loss_clip": 0.01047168, "auxiliary_loss_mlp": 0.00753307, "balance_loss_clip": 1.03757286, "balance_loss_mlp": 1.00001645, "epoch": 0.7647447844646186, "flos": 70402344644640.0, "grad_norm": 0.8552642255194107, "language_loss": 0.65816039, "learning_rate": 5.528577531951173e-07, "loss": 0.67616516, "num_input_tokens_seen": 136878730, "step": 6360, "time_per_iteration": 3.1433470249176025 }, { "auxiliary_loss_clip": 0.01143313, "auxiliary_loss_mlp": 0.01026094, "balance_loss_clip": 1.04479492, "balance_loss_mlp": 1.01929283, "epoch": 0.7648650273552576, "flos": 17675918187360.0, "grad_norm": 2.185551631973145, "language_loss": 0.73882318, "learning_rate": 5.523201779258653e-07, "loss": 0.76051736, "num_input_tokens_seen": 136897705, "step": 6361, "time_per_iteration": 2.5048720836639404 }, { "auxiliary_loss_clip": 0.0116818, "auxiliary_loss_mlp": 0.01023529, "balance_loss_clip": 1.04583061, "balance_loss_mlp": 1.01620066, "epoch": 0.7649852702458967, "flos": 22162022735040.0, "grad_norm": 1.7863752383090272, "language_loss": 0.83763927, "learning_rate": 5.517828222684912e-07, "loss": 0.85955644, "num_input_tokens_seen": 136918360, "step": 6362, "time_per_iteration": 2.457092046737671 }, { "auxiliary_loss_clip": 0.01042271, "auxiliary_loss_mlp": 0.01001575, "balance_loss_clip": 1.00828099, "balance_loss_mlp": 1.00023961, "epoch": 0.7651055131365359, "flos": 69848349527520.0, "grad_norm": 0.7691067729732084, "language_loss": 0.59123701, "learning_rate": 5.512456863045117e-07, "loss": 0.61167544, "num_input_tokens_seen": 136979050, "step": 6363, "time_per_iteration": 3.110253095626831 }, { "auxiliary_loss_clip": 0.01168816, "auxiliary_loss_mlp": 0.01026263, "balance_loss_clip": 1.04616237, "balance_loss_mlp": 1.01894331, "epoch": 0.7652257560271749, "flos": 19464018348960.0, "grad_norm": 1.6594203890393913, "language_loss": 0.74122286, "learning_rate": 5.507087701154089e-07, "loss": 0.76317358, "num_input_tokens_seen": 136998970, "step": 6364, "time_per_iteration": 2.456113338470459 }, { "auxiliary_loss_clip": 0.01116568, "auxiliary_loss_mlp": 0.0102328, "balance_loss_clip": 1.04250896, "balance_loss_mlp": 1.01645231, "epoch": 0.765345998917814, "flos": 15961111764480.0, "grad_norm": 2.108173496872997, "language_loss": 0.75425059, "learning_rate": 5.50172073782634e-07, "loss": 0.77564907, "num_input_tokens_seen": 137016950, "step": 6365, "time_per_iteration": 3.273815870285034 }, { "auxiliary_loss_clip": 0.01125711, "auxiliary_loss_mlp": 0.01025606, "balance_loss_clip": 1.04669166, "balance_loss_mlp": 1.01903439, "epoch": 0.7654662418084531, "flos": 23659857215520.0, "grad_norm": 1.708278985189029, "language_loss": 0.87634718, "learning_rate": 5.496355973876023e-07, "loss": 0.89786029, "num_input_tokens_seen": 137036205, "step": 6366, "time_per_iteration": 2.558565855026245 }, { "auxiliary_loss_clip": 0.01124634, "auxiliary_loss_mlp": 0.00762587, "balance_loss_clip": 1.04261661, "balance_loss_mlp": 1.00050306, "epoch": 0.7655864846990922, "flos": 41463614227200.0, "grad_norm": 1.7499652762813667, "language_loss": 0.70988452, "learning_rate": 5.490993410116984e-07, "loss": 0.72875679, "num_input_tokens_seen": 137059195, "step": 6367, "time_per_iteration": 2.7098748683929443 }, { "auxiliary_loss_clip": 0.01123076, "auxiliary_loss_mlp": 0.01030604, "balance_loss_clip": 1.04555821, "balance_loss_mlp": 1.02390432, "epoch": 0.7657067275897312, "flos": 43142689663200.0, "grad_norm": 1.59665391478437, "language_loss": 0.69390368, "learning_rate": 5.485633047362704e-07, "loss": 0.71544051, "num_input_tokens_seen": 137081200, "step": 6368, "time_per_iteration": 2.7452659606933594 }, { "auxiliary_loss_clip": 0.01170881, "auxiliary_loss_mlp": 0.01028276, "balance_loss_clip": 1.04855657, "balance_loss_mlp": 1.02085805, "epoch": 0.7658269704803703, "flos": 17311784096160.0, "grad_norm": 3.097595930819368, "language_loss": 0.78348482, "learning_rate": 5.480274886426341e-07, "loss": 0.80547631, "num_input_tokens_seen": 137097840, "step": 6369, "time_per_iteration": 2.4238784313201904 }, { "auxiliary_loss_clip": 0.01154623, "auxiliary_loss_mlp": 0.010243, "balance_loss_clip": 1.05093265, "balance_loss_mlp": 1.0176425, "epoch": 0.7659472133710095, "flos": 12568163746560.0, "grad_norm": 3.1830848396691596, "language_loss": 0.77869743, "learning_rate": 5.474918928120744e-07, "loss": 0.80048668, "num_input_tokens_seen": 137114335, "step": 6370, "time_per_iteration": 2.455451726913452 }, { "auxiliary_loss_clip": 0.01152786, "auxiliary_loss_mlp": 0.01026665, "balance_loss_clip": 1.04668581, "balance_loss_mlp": 1.02020121, "epoch": 0.7660674562616485, "flos": 22707433698720.0, "grad_norm": 1.849997233502717, "language_loss": 0.87351739, "learning_rate": 5.469565173258392e-07, "loss": 0.89531189, "num_input_tokens_seen": 137132850, "step": 6371, "time_per_iteration": 2.48492693901062 }, { "auxiliary_loss_clip": 0.01174123, "auxiliary_loss_mlp": 0.01025148, "balance_loss_clip": 1.04917336, "balance_loss_mlp": 1.01763463, "epoch": 0.7661876991522876, "flos": 17056459228800.0, "grad_norm": 1.7637531887896243, "language_loss": 0.63970208, "learning_rate": 5.464213622651454e-07, "loss": 0.66169477, "num_input_tokens_seen": 137150665, "step": 6372, "time_per_iteration": 2.4334359169006348 }, { "auxiliary_loss_clip": 0.01134509, "auxiliary_loss_mlp": 0.01031513, "balance_loss_clip": 1.04551363, "balance_loss_mlp": 1.02427673, "epoch": 0.7663079420429267, "flos": 20084231563680.0, "grad_norm": 3.9073362840136765, "language_loss": 0.84097767, "learning_rate": 5.458864277111753e-07, "loss": 0.86263794, "num_input_tokens_seen": 137168500, "step": 6373, "time_per_iteration": 2.534193277359009 }, { "auxiliary_loss_clip": 0.01134243, "auxiliary_loss_mlp": 0.00761713, "balance_loss_clip": 1.04315567, "balance_loss_mlp": 1.00050735, "epoch": 0.7664281849335658, "flos": 12677475807840.0, "grad_norm": 2.7565378178247197, "language_loss": 0.69631994, "learning_rate": 5.453517137450769e-07, "loss": 0.71527958, "num_input_tokens_seen": 137185075, "step": 6374, "time_per_iteration": 2.5409274101257324 }, { "auxiliary_loss_clip": 0.01155962, "auxiliary_loss_mlp": 0.01026678, "balance_loss_clip": 1.04860163, "balance_loss_mlp": 1.01904309, "epoch": 0.7665484278242048, "flos": 22345275040320.0, "grad_norm": 1.7584229519652776, "language_loss": 0.76012909, "learning_rate": 5.448172204479684e-07, "loss": 0.78195548, "num_input_tokens_seen": 137204355, "step": 6375, "time_per_iteration": 2.481663465499878 }, { "auxiliary_loss_clip": 0.01166228, "auxiliary_loss_mlp": 0.01024977, "balance_loss_clip": 1.04611588, "balance_loss_mlp": 1.0180068, "epoch": 0.766668670714844, "flos": 23617912594560.0, "grad_norm": 1.6726145484367239, "language_loss": 0.74562764, "learning_rate": 5.442829479009294e-07, "loss": 0.76753974, "num_input_tokens_seen": 137223135, "step": 6376, "time_per_iteration": 2.4713504314422607 }, { "auxiliary_loss_clip": 0.01160504, "auxiliary_loss_mlp": 0.01026144, "balance_loss_clip": 1.04824388, "balance_loss_mlp": 1.01856852, "epoch": 0.7667889136054831, "flos": 19427138019360.0, "grad_norm": 2.0600494692314695, "language_loss": 0.71377081, "learning_rate": 5.437488961850103e-07, "loss": 0.73563731, "num_input_tokens_seen": 137242935, "step": 6377, "time_per_iteration": 2.467406988143921 }, { "auxiliary_loss_clip": 0.01111663, "auxiliary_loss_mlp": 0.01024358, "balance_loss_clip": 1.04332173, "balance_loss_mlp": 1.01788473, "epoch": 0.7669091564961221, "flos": 26866356318720.0, "grad_norm": 1.8616553863272345, "language_loss": 0.75582176, "learning_rate": 5.432150653812258e-07, "loss": 0.77718198, "num_input_tokens_seen": 137262970, "step": 6378, "time_per_iteration": 3.4546682834625244 }, { "auxiliary_loss_clip": 0.01150066, "auxiliary_loss_mlp": 0.01027975, "balance_loss_clip": 1.04470682, "balance_loss_mlp": 1.02072144, "epoch": 0.7670293993867613, "flos": 12385306527840.0, "grad_norm": 2.479192302148769, "language_loss": 0.82527745, "learning_rate": 5.42681455570557e-07, "loss": 0.84705788, "num_input_tokens_seen": 137279500, "step": 6379, "time_per_iteration": 2.441063165664673 }, { "auxiliary_loss_clip": 0.01165785, "auxiliary_loss_mlp": 0.01023827, "balance_loss_clip": 1.04591417, "balance_loss_mlp": 1.01733339, "epoch": 0.7671496422774003, "flos": 21762947830080.0, "grad_norm": 1.8077974365896365, "language_loss": 0.64829874, "learning_rate": 5.42148066833954e-07, "loss": 0.67019486, "num_input_tokens_seen": 137298745, "step": 6380, "time_per_iteration": 2.4714176654815674 }, { "auxiliary_loss_clip": 0.01167534, "auxiliary_loss_mlp": 0.01025974, "balance_loss_clip": 1.04817295, "balance_loss_mlp": 1.0186044, "epoch": 0.7672698851680394, "flos": 21069225374880.0, "grad_norm": 2.627354368026011, "language_loss": 0.75501537, "learning_rate": 5.416148992523289e-07, "loss": 0.77695048, "num_input_tokens_seen": 137317320, "step": 6381, "time_per_iteration": 3.9874651432037354 }, { "auxiliary_loss_clip": 0.01086392, "auxiliary_loss_mlp": 0.01026083, "balance_loss_clip": 1.04150081, "balance_loss_mlp": 1.01907027, "epoch": 0.7673901280586786, "flos": 16976700437280.0, "grad_norm": 2.0383213860993536, "language_loss": 0.78727967, "learning_rate": 5.410819529065644e-07, "loss": 0.80840439, "num_input_tokens_seen": 137335275, "step": 6382, "time_per_iteration": 2.6176817417144775 }, { "auxiliary_loss_clip": 0.01111097, "auxiliary_loss_mlp": 0.01030073, "balance_loss_clip": 1.04102921, "balance_loss_mlp": 1.02295339, "epoch": 0.7675103709493176, "flos": 29242674072000.0, "grad_norm": 2.421774709092011, "language_loss": 0.65382719, "learning_rate": 5.405492278775079e-07, "loss": 0.67523891, "num_input_tokens_seen": 137355055, "step": 6383, "time_per_iteration": 2.647764205932617 }, { "auxiliary_loss_clip": 0.0114225, "auxiliary_loss_mlp": 0.0102342, "balance_loss_clip": 1.04388356, "balance_loss_mlp": 1.01598692, "epoch": 0.7676306138399567, "flos": 29023008357600.0, "grad_norm": 2.225498908817865, "language_loss": 0.79918528, "learning_rate": 5.400167242459732e-07, "loss": 0.82084197, "num_input_tokens_seen": 137374015, "step": 6384, "time_per_iteration": 2.5759973526000977 }, { "auxiliary_loss_clip": 0.011525, "auxiliary_loss_mlp": 0.01023929, "balance_loss_clip": 1.0462178, "balance_loss_mlp": 1.01678205, "epoch": 0.7677508567305958, "flos": 22565120339520.0, "grad_norm": 1.677824596396961, "language_loss": 0.80635875, "learning_rate": 5.394844420927405e-07, "loss": 0.82812303, "num_input_tokens_seen": 137393625, "step": 6385, "time_per_iteration": 2.4845690727233887 }, { "auxiliary_loss_clip": 0.01167892, "auxiliary_loss_mlp": 0.0102767, "balance_loss_clip": 1.04845834, "balance_loss_mlp": 1.02044034, "epoch": 0.7678710996212349, "flos": 25411436217120.0, "grad_norm": 2.9270156082624106, "language_loss": 0.73431081, "learning_rate": 5.389523814985562e-07, "loss": 0.75626642, "num_input_tokens_seen": 137413045, "step": 6386, "time_per_iteration": 2.4855473041534424 }, { "auxiliary_loss_clip": 0.01112066, "auxiliary_loss_mlp": 0.01023009, "balance_loss_clip": 1.04248118, "balance_loss_mlp": 1.01572251, "epoch": 0.767991342511874, "flos": 26756828755680.0, "grad_norm": 5.756255958128166, "language_loss": 0.76042145, "learning_rate": 5.384205425441344e-07, "loss": 0.7817722, "num_input_tokens_seen": 137433955, "step": 6387, "time_per_iteration": 2.6572160720825195 }, { "auxiliary_loss_clip": 0.01140898, "auxiliary_loss_mlp": 0.01024264, "balance_loss_clip": 1.04395628, "balance_loss_mlp": 1.01757312, "epoch": 0.7681115854025131, "flos": 26359513781760.0, "grad_norm": 1.648110923753789, "language_loss": 0.84180087, "learning_rate": 5.378889253101537e-07, "loss": 0.86345243, "num_input_tokens_seen": 137454510, "step": 6388, "time_per_iteration": 2.5563650131225586 }, { "auxiliary_loss_clip": 0.01152944, "auxiliary_loss_mlp": 0.01024605, "balance_loss_clip": 1.04343128, "balance_loss_mlp": 1.01785505, "epoch": 0.7682318282931522, "flos": 23257046946720.0, "grad_norm": 1.6101205155862228, "language_loss": 0.81097847, "learning_rate": 5.373575298772617e-07, "loss": 0.83275396, "num_input_tokens_seen": 137473630, "step": 6389, "time_per_iteration": 2.504166841506958 }, { "auxiliary_loss_clip": 0.01053169, "auxiliary_loss_mlp": 0.01003361, "balance_loss_clip": 1.00613523, "balance_loss_mlp": 1.00199056, "epoch": 0.7683520711837912, "flos": 70072468944960.0, "grad_norm": 0.7543962334034744, "language_loss": 0.61369038, "learning_rate": 5.368263563260689e-07, "loss": 0.63425565, "num_input_tokens_seen": 137538765, "step": 6390, "time_per_iteration": 3.2111501693725586 }, { "auxiliary_loss_clip": 0.01154755, "auxiliary_loss_mlp": 0.01027879, "balance_loss_clip": 1.04451632, "balance_loss_mlp": 1.02002025, "epoch": 0.7684723140744304, "flos": 18624893676000.0, "grad_norm": 1.7606115277733545, "language_loss": 0.63909739, "learning_rate": 5.362954047371537e-07, "loss": 0.66092372, "num_input_tokens_seen": 137557875, "step": 6391, "time_per_iteration": 2.4730663299560547 }, { "auxiliary_loss_clip": 0.01131925, "auxiliary_loss_mlp": 0.01034992, "balance_loss_clip": 1.05023336, "balance_loss_mlp": 1.0278213, "epoch": 0.7685925569650695, "flos": 27452993564160.0, "grad_norm": 1.9152364520767002, "language_loss": 0.71902549, "learning_rate": 5.357646751910627e-07, "loss": 0.74069464, "num_input_tokens_seen": 137579055, "step": 6392, "time_per_iteration": 3.336181640625 }, { "auxiliary_loss_clip": 0.01137314, "auxiliary_loss_mlp": 0.01030529, "balance_loss_clip": 1.04334211, "balance_loss_mlp": 1.02233005, "epoch": 0.7687127998557085, "flos": 24535718550240.0, "grad_norm": 2.08939614033483, "language_loss": 0.79909539, "learning_rate": 5.352341677683061e-07, "loss": 0.82077384, "num_input_tokens_seen": 137600355, "step": 6393, "time_per_iteration": 2.5458734035491943 }, { "auxiliary_loss_clip": 0.01137063, "auxiliary_loss_mlp": 0.01022749, "balance_loss_clip": 1.04620147, "balance_loss_mlp": 1.01597166, "epoch": 0.7688330427463477, "flos": 25155967681920.0, "grad_norm": 1.838945339965047, "language_loss": 0.78954297, "learning_rate": 5.347038825493617e-07, "loss": 0.81114107, "num_input_tokens_seen": 137621885, "step": 6394, "time_per_iteration": 2.60630464553833 }, { "auxiliary_loss_clip": 0.01137564, "auxiliary_loss_mlp": 0.01024817, "balance_loss_clip": 1.04733682, "balance_loss_mlp": 1.01883841, "epoch": 0.7689532856369867, "flos": 21211287315360.0, "grad_norm": 1.879171271519603, "language_loss": 0.6882714, "learning_rate": 5.341738196146732e-07, "loss": 0.70989519, "num_input_tokens_seen": 137640230, "step": 6395, "time_per_iteration": 2.492441415786743 }, { "auxiliary_loss_clip": 0.01149352, "auxiliary_loss_mlp": 0.01031673, "balance_loss_clip": 1.04305649, "balance_loss_mlp": 1.02403224, "epoch": 0.7690735285276258, "flos": 25119087352320.0, "grad_norm": 2.0814910612843236, "language_loss": 0.73515171, "learning_rate": 5.336439790446503e-07, "loss": 0.75696194, "num_input_tokens_seen": 137659330, "step": 6396, "time_per_iteration": 2.515350103378296 }, { "auxiliary_loss_clip": 0.01123003, "auxiliary_loss_mlp": 0.01025691, "balance_loss_clip": 1.04072714, "balance_loss_mlp": 1.01849675, "epoch": 0.769193771418265, "flos": 54744027192000.0, "grad_norm": 2.2274528464970134, "language_loss": 0.62297809, "learning_rate": 5.331143609196711e-07, "loss": 0.64446509, "num_input_tokens_seen": 137683145, "step": 6397, "time_per_iteration": 2.849738597869873 }, { "auxiliary_loss_clip": 0.01154175, "auxiliary_loss_mlp": 0.01027037, "balance_loss_clip": 1.04718995, "balance_loss_mlp": 1.01958656, "epoch": 0.769314014308904, "flos": 37341895189440.0, "grad_norm": 1.7747735650451468, "language_loss": 0.76941574, "learning_rate": 5.325849653200758e-07, "loss": 0.79122782, "num_input_tokens_seen": 137707095, "step": 6398, "time_per_iteration": 2.6138718128204346 }, { "auxiliary_loss_clip": 0.01168032, "auxiliary_loss_mlp": 0.01026294, "balance_loss_clip": 1.04758835, "balance_loss_mlp": 1.01899242, "epoch": 0.7694342571995431, "flos": 20631689794080.0, "grad_norm": 1.7099161722078384, "language_loss": 0.76100028, "learning_rate": 5.32055792326175e-07, "loss": 0.78294355, "num_input_tokens_seen": 137725520, "step": 6399, "time_per_iteration": 2.4334053993225098 }, { "auxiliary_loss_clip": 0.01144596, "auxiliary_loss_mlp": 0.01030372, "balance_loss_clip": 1.04722774, "balance_loss_mlp": 1.02356851, "epoch": 0.7695545000901821, "flos": 24207710532480.0, "grad_norm": 1.9311629315090912, "language_loss": 0.72711551, "learning_rate": 5.315268420182437e-07, "loss": 0.74886513, "num_input_tokens_seen": 137744195, "step": 6400, "time_per_iteration": 2.5598978996276855 }, { "auxiliary_loss_clip": 0.01133881, "auxiliary_loss_mlp": 0.00761287, "balance_loss_clip": 1.04570282, "balance_loss_mlp": 1.00049353, "epoch": 0.7696747429808213, "flos": 28001277884640.0, "grad_norm": 1.782561650193979, "language_loss": 0.76482224, "learning_rate": 5.309981144765221e-07, "loss": 0.78377396, "num_input_tokens_seen": 137764340, "step": 6401, "time_per_iteration": 2.594856023788452 }, { "auxiliary_loss_clip": 0.01119289, "auxiliary_loss_mlp": 0.01021471, "balance_loss_clip": 1.04204857, "balance_loss_mlp": 1.01425934, "epoch": 0.7697949858714603, "flos": 11509552944000.0, "grad_norm": 5.625835639303789, "language_loss": 0.75817192, "learning_rate": 5.304696097812196e-07, "loss": 0.77957946, "num_input_tokens_seen": 137780940, "step": 6402, "time_per_iteration": 2.556408166885376 }, { "auxiliary_loss_clip": 0.01138627, "auxiliary_loss_mlp": 0.01031167, "balance_loss_clip": 1.04480338, "balance_loss_mlp": 1.02351367, "epoch": 0.7699152287620994, "flos": 26688276055680.0, "grad_norm": 3.132156932012968, "language_loss": 0.59960222, "learning_rate": 5.299413280125078e-07, "loss": 0.62130022, "num_input_tokens_seen": 137799250, "step": 6403, "time_per_iteration": 3.327568769454956 }, { "auxiliary_loss_clip": 0.01139933, "auxiliary_loss_mlp": 0.01024931, "balance_loss_clip": 1.04380202, "balance_loss_mlp": 1.01830888, "epoch": 0.7700354716527386, "flos": 16544947487040.0, "grad_norm": 2.0205828922366593, "language_loss": 0.72918415, "learning_rate": 5.294132692505284e-07, "loss": 0.75083274, "num_input_tokens_seen": 137817660, "step": 6404, "time_per_iteration": 2.503167152404785 }, { "auxiliary_loss_clip": 0.0110324, "auxiliary_loss_mlp": 0.01021273, "balance_loss_clip": 1.03985107, "balance_loss_mlp": 1.01441526, "epoch": 0.7701557145433776, "flos": 19242736371360.0, "grad_norm": 2.238574869824141, "language_loss": 0.79063559, "learning_rate": 5.288854335753861e-07, "loss": 0.81188071, "num_input_tokens_seen": 137835920, "step": 6405, "time_per_iteration": 2.5578408241271973 }, { "auxiliary_loss_clip": 0.01156233, "auxiliary_loss_mlp": 0.01023843, "balance_loss_clip": 1.04591286, "balance_loss_mlp": 1.01716125, "epoch": 0.7702759574340167, "flos": 31685748677280.0, "grad_norm": 2.027157031313732, "language_loss": 0.75603229, "learning_rate": 5.283578210671551e-07, "loss": 0.77783298, "num_input_tokens_seen": 137858160, "step": 6406, "time_per_iteration": 2.5615689754486084 }, { "auxiliary_loss_clip": 0.01144086, "auxiliary_loss_mlp": 0.01027451, "balance_loss_clip": 1.04489195, "balance_loss_mlp": 1.0205009, "epoch": 0.7703962003246558, "flos": 16800092769600.0, "grad_norm": 1.9970012575255078, "language_loss": 0.76614535, "learning_rate": 5.278304318058719e-07, "loss": 0.78786075, "num_input_tokens_seen": 137876015, "step": 6407, "time_per_iteration": 4.0179123878479 }, { "auxiliary_loss_clip": 0.01097263, "auxiliary_loss_mlp": 0.01021999, "balance_loss_clip": 1.04075599, "balance_loss_mlp": 1.01468825, "epoch": 0.7705164432152949, "flos": 35736077575200.0, "grad_norm": 2.219467324882242, "language_loss": 0.78943485, "learning_rate": 5.273032658715411e-07, "loss": 0.81062746, "num_input_tokens_seen": 137898825, "step": 6408, "time_per_iteration": 2.70906925201416 }, { "auxiliary_loss_clip": 0.01108823, "auxiliary_loss_mlp": 0.01022902, "balance_loss_clip": 1.04125094, "balance_loss_mlp": 1.0158031, "epoch": 0.7706366861059339, "flos": 23365963921440.0, "grad_norm": 2.0331633299461886, "language_loss": 0.7654109, "learning_rate": 5.267763233441347e-07, "loss": 0.78672814, "num_input_tokens_seen": 137919455, "step": 6409, "time_per_iteration": 2.6160593032836914 }, { "auxiliary_loss_clip": 0.01160372, "auxiliary_loss_mlp": 0.01024233, "balance_loss_clip": 1.04832339, "balance_loss_mlp": 1.01643062, "epoch": 0.7707569289965731, "flos": 22929900936000.0, "grad_norm": 2.939982860192641, "language_loss": 0.69354886, "learning_rate": 5.26249604303588e-07, "loss": 0.71539491, "num_input_tokens_seen": 137937960, "step": 6410, "time_per_iteration": 2.4876620769500732 }, { "auxiliary_loss_clip": 0.01169866, "auxiliary_loss_mlp": 0.01023468, "balance_loss_clip": 1.04857612, "balance_loss_mlp": 1.01661646, "epoch": 0.7708771718872122, "flos": 17420665153920.0, "grad_norm": 2.517013066746099, "language_loss": 0.78655219, "learning_rate": 5.257231088298057e-07, "loss": 0.80848551, "num_input_tokens_seen": 137956370, "step": 6411, "time_per_iteration": 2.4435184001922607 }, { "auxiliary_loss_clip": 0.01029713, "auxiliary_loss_mlp": 0.01003107, "balance_loss_clip": 1.00799298, "balance_loss_mlp": 1.00161731, "epoch": 0.7709974147778512, "flos": 72241326361920.0, "grad_norm": 0.7922121271055548, "language_loss": 0.53935945, "learning_rate": 5.25196837002655e-07, "loss": 0.55968767, "num_input_tokens_seen": 138016080, "step": 6412, "time_per_iteration": 3.178584098815918 }, { "auxiliary_loss_clip": 0.01137248, "auxiliary_loss_mlp": 0.01029049, "balance_loss_clip": 1.04484916, "balance_loss_mlp": 1.02180409, "epoch": 0.7711176576684904, "flos": 39859700211840.0, "grad_norm": 4.179034479779039, "language_loss": 0.68186039, "learning_rate": 5.24670788901971e-07, "loss": 0.7035234, "num_input_tokens_seen": 138039170, "step": 6413, "time_per_iteration": 2.6710004806518555 }, { "auxiliary_loss_clip": 0.01141728, "auxiliary_loss_mlp": 0.01025009, "balance_loss_clip": 1.04657793, "balance_loss_mlp": 1.01693869, "epoch": 0.7712379005591294, "flos": 36976396253760.0, "grad_norm": 2.198840033003096, "language_loss": 0.68580574, "learning_rate": 5.241449646075557e-07, "loss": 0.7074731, "num_input_tokens_seen": 138062395, "step": 6414, "time_per_iteration": 2.642096996307373 }, { "auxiliary_loss_clip": 0.01163153, "auxiliary_loss_mlp": 0.01020863, "balance_loss_clip": 1.04742539, "balance_loss_mlp": 1.01348984, "epoch": 0.7713581434497685, "flos": 22776776571840.0, "grad_norm": 1.9666983531934674, "language_loss": 0.72222835, "learning_rate": 5.236193641991762e-07, "loss": 0.7440685, "num_input_tokens_seen": 138080325, "step": 6415, "time_per_iteration": 2.4893345832824707 }, { "auxiliary_loss_clip": 0.0113934, "auxiliary_loss_mlp": 0.01023872, "balance_loss_clip": 1.04537439, "balance_loss_mlp": 1.01681161, "epoch": 0.7714783863404077, "flos": 24097464630240.0, "grad_norm": 2.284347499892092, "language_loss": 0.69781882, "learning_rate": 5.23093987756565e-07, "loss": 0.71945095, "num_input_tokens_seen": 138099020, "step": 6416, "time_per_iteration": 2.535057306289673 }, { "auxiliary_loss_clip": 0.0113327, "auxiliary_loss_mlp": 0.01026271, "balance_loss_clip": 1.04093468, "balance_loss_mlp": 1.01879668, "epoch": 0.7715986292310467, "flos": 21063658246080.0, "grad_norm": 1.7397349605743226, "language_loss": 0.75528932, "learning_rate": 5.225688353594217e-07, "loss": 0.77688479, "num_input_tokens_seen": 138118650, "step": 6417, "time_per_iteration": 2.6717841625213623 }, { "auxiliary_loss_clip": 0.01145454, "auxiliary_loss_mlp": 0.00762033, "balance_loss_clip": 1.04581618, "balance_loss_mlp": 1.00059366, "epoch": 0.7717188721216858, "flos": 20594881298400.0, "grad_norm": 2.234210463999468, "language_loss": 0.77769637, "learning_rate": 5.220439070874108e-07, "loss": 0.79677117, "num_input_tokens_seen": 138137890, "step": 6418, "time_per_iteration": 3.2658894062042236 }, { "auxiliary_loss_clip": 0.01153718, "auxiliary_loss_mlp": 0.01027081, "balance_loss_clip": 1.04662514, "balance_loss_mlp": 1.02022088, "epoch": 0.7718391150123249, "flos": 26250955976640.0, "grad_norm": 1.5922582983094575, "language_loss": 0.71108234, "learning_rate": 5.215192030201652e-07, "loss": 0.73289037, "num_input_tokens_seen": 138158880, "step": 6419, "time_per_iteration": 2.504760265350342 }, { "auxiliary_loss_clip": 0.0111218, "auxiliary_loss_mlp": 0.01029504, "balance_loss_clip": 1.03839648, "balance_loss_mlp": 1.02251196, "epoch": 0.771959357902964, "flos": 22049765483040.0, "grad_norm": 2.069399463991938, "language_loss": 0.86273843, "learning_rate": 5.209947232372798e-07, "loss": 0.88415527, "num_input_tokens_seen": 138176370, "step": 6420, "time_per_iteration": 2.5369679927825928 }, { "auxiliary_loss_clip": 0.01158162, "auxiliary_loss_mlp": 0.00762257, "balance_loss_clip": 1.0454073, "balance_loss_mlp": 1.00062287, "epoch": 0.772079600793603, "flos": 30446004670080.0, "grad_norm": 1.8919673680987412, "language_loss": 0.81281722, "learning_rate": 5.204704678183196e-07, "loss": 0.83202147, "num_input_tokens_seen": 138195105, "step": 6421, "time_per_iteration": 2.5580854415893555 }, { "auxiliary_loss_clip": 0.01169751, "auxiliary_loss_mlp": 0.01027132, "balance_loss_clip": 1.04883158, "balance_loss_mlp": 1.01937771, "epoch": 0.7721998436842422, "flos": 12969860589600.0, "grad_norm": 1.875847022781266, "language_loss": 0.85063255, "learning_rate": 5.19946436842813e-07, "loss": 0.87260139, "num_input_tokens_seen": 138212235, "step": 6422, "time_per_iteration": 2.445582151412964 }, { "auxiliary_loss_clip": 0.01128341, "auxiliary_loss_mlp": 0.01023191, "balance_loss_clip": 1.04524565, "balance_loss_mlp": 1.01650655, "epoch": 0.7723200865748813, "flos": 32635514339040.0, "grad_norm": 1.6258830707621208, "language_loss": 0.68428802, "learning_rate": 5.194226303902546e-07, "loss": 0.70580333, "num_input_tokens_seen": 138231970, "step": 6423, "time_per_iteration": 2.6657726764678955 }, { "auxiliary_loss_clip": 0.01138229, "auxiliary_loss_mlp": 0.01022895, "balance_loss_clip": 1.04421723, "balance_loss_mlp": 1.0157423, "epoch": 0.7724403294655203, "flos": 21105710617920.0, "grad_norm": 1.6687011490218309, "language_loss": 0.70539284, "learning_rate": 5.188990485401072e-07, "loss": 0.72700405, "num_input_tokens_seen": 138251175, "step": 6424, "time_per_iteration": 2.534992218017578 }, { "auxiliary_loss_clip": 0.01158452, "auxiliary_loss_mlp": 0.0102571, "balance_loss_clip": 1.04748201, "balance_loss_mlp": 1.01845312, "epoch": 0.7725605723561595, "flos": 22090740346080.0, "grad_norm": 1.9681819293903462, "language_loss": 0.86166525, "learning_rate": 5.183756913717954e-07, "loss": 0.88350689, "num_input_tokens_seen": 138270950, "step": 6425, "time_per_iteration": 2.49784779548645 }, { "auxiliary_loss_clip": 0.01135856, "auxiliary_loss_mlp": 0.01024084, "balance_loss_clip": 1.04399562, "balance_loss_mlp": 1.01660037, "epoch": 0.7726808152467985, "flos": 34495615228800.0, "grad_norm": 1.7312442217723827, "language_loss": 0.72948509, "learning_rate": 5.178525589647136e-07, "loss": 0.75108451, "num_input_tokens_seen": 138292590, "step": 6426, "time_per_iteration": 2.6194112300872803 }, { "auxiliary_loss_clip": 0.0114342, "auxiliary_loss_mlp": 0.0102355, "balance_loss_clip": 1.04195344, "balance_loss_mlp": 1.01741624, "epoch": 0.7728010581374376, "flos": 22306347444000.0, "grad_norm": 1.8197108043414496, "language_loss": 0.78912306, "learning_rate": 5.173296513982197e-07, "loss": 0.8107928, "num_input_tokens_seen": 138311115, "step": 6427, "time_per_iteration": 2.505903959274292 }, { "auxiliary_loss_clip": 0.01140607, "auxiliary_loss_mlp": 0.01028881, "balance_loss_clip": 1.04880416, "balance_loss_mlp": 1.02181768, "epoch": 0.7729213010280768, "flos": 27126458141760.0, "grad_norm": 2.2546777797540423, "language_loss": 0.64917284, "learning_rate": 5.168069687516398e-07, "loss": 0.67086768, "num_input_tokens_seen": 138330885, "step": 6428, "time_per_iteration": 2.603076934814453 }, { "auxiliary_loss_clip": 0.01141768, "auxiliary_loss_mlp": 0.0103313, "balance_loss_clip": 1.0479064, "balance_loss_mlp": 1.02582872, "epoch": 0.7730415439187158, "flos": 18150226346880.0, "grad_norm": 1.9994252940286252, "language_loss": 0.71521842, "learning_rate": 5.16284511104263e-07, "loss": 0.73696744, "num_input_tokens_seen": 138350020, "step": 6429, "time_per_iteration": 2.5076136589050293 }, { "auxiliary_loss_clip": 0.01142918, "auxiliary_loss_mlp": 0.01025536, "balance_loss_clip": 1.04841113, "balance_loss_mlp": 1.01838934, "epoch": 0.7731617868093549, "flos": 11947483611360.0, "grad_norm": 2.770265981116078, "language_loss": 0.80549091, "learning_rate": 5.157622785353457e-07, "loss": 0.82717544, "num_input_tokens_seen": 138368135, "step": 6430, "time_per_iteration": 3.2650837898254395 }, { "auxiliary_loss_clip": 0.01054353, "auxiliary_loss_mlp": 0.01000392, "balance_loss_clip": 1.00750351, "balance_loss_mlp": 0.99908656, "epoch": 0.7732820296999939, "flos": 64201038587520.0, "grad_norm": 0.648656809808443, "language_loss": 0.60401285, "learning_rate": 5.152402711241113e-07, "loss": 0.62456024, "num_input_tokens_seen": 138436040, "step": 6431, "time_per_iteration": 3.1628975868225098 }, { "auxiliary_loss_clip": 0.01124493, "auxiliary_loss_mlp": 0.01028072, "balance_loss_clip": 1.04188418, "balance_loss_mlp": 1.02045798, "epoch": 0.7734022725906331, "flos": 25302195989760.0, "grad_norm": 1.7807572184299088, "language_loss": 0.82954586, "learning_rate": 5.147184889497465e-07, "loss": 0.85107148, "num_input_tokens_seen": 138455510, "step": 6432, "time_per_iteration": 3.3730697631835938 }, { "auxiliary_loss_clip": 0.01119061, "auxiliary_loss_mlp": 0.01030047, "balance_loss_clip": 1.04374695, "balance_loss_mlp": 1.02245903, "epoch": 0.7735225154812722, "flos": 17347443249120.0, "grad_norm": 2.4866996233936827, "language_loss": 0.79743743, "learning_rate": 5.141969320914072e-07, "loss": 0.81892848, "num_input_tokens_seen": 138473015, "step": 6433, "time_per_iteration": 3.203433036804199 }, { "auxiliary_loss_clip": 0.01173242, "auxiliary_loss_mlp": 0.01022944, "balance_loss_clip": 1.04912961, "balance_loss_mlp": 1.0153625, "epoch": 0.7736427583719112, "flos": 32630090878080.0, "grad_norm": 3.539209182131396, "language_loss": 0.62273729, "learning_rate": 5.136756006282113e-07, "loss": 0.64469922, "num_input_tokens_seen": 138491680, "step": 6434, "time_per_iteration": 2.5592129230499268 }, { "auxiliary_loss_clip": 0.01171304, "auxiliary_loss_mlp": 0.01027768, "balance_loss_clip": 1.04823709, "balance_loss_mlp": 1.02060318, "epoch": 0.7737630012625504, "flos": 19860076229280.0, "grad_norm": 2.465276753887125, "language_loss": 0.85133576, "learning_rate": 5.131544946392446e-07, "loss": 0.87332642, "num_input_tokens_seen": 138506960, "step": 6435, "time_per_iteration": 2.466621160507202 }, { "auxiliary_loss_clip": 0.01144433, "auxiliary_loss_mlp": 0.01028423, "balance_loss_clip": 1.04973936, "balance_loss_mlp": 1.02099299, "epoch": 0.7738832441531894, "flos": 36022643809440.0, "grad_norm": 2.1360125750376255, "language_loss": 0.63870168, "learning_rate": 5.126336142035592e-07, "loss": 0.66043019, "num_input_tokens_seen": 138526995, "step": 6436, "time_per_iteration": 2.6264588832855225 }, { "auxiliary_loss_clip": 0.01140635, "auxiliary_loss_mlp": 0.01024974, "balance_loss_clip": 1.04364944, "balance_loss_mlp": 1.01762462, "epoch": 0.7740034870438285, "flos": 13405277069760.0, "grad_norm": 2.553537892178908, "language_loss": 0.71707273, "learning_rate": 5.121129594001721e-07, "loss": 0.73872882, "num_input_tokens_seen": 138541260, "step": 6437, "time_per_iteration": 2.4845032691955566 }, { "auxiliary_loss_clip": 0.01156635, "auxiliary_loss_mlp": 0.01031739, "balance_loss_clip": 1.04844618, "balance_loss_mlp": 1.02433872, "epoch": 0.7741237299344677, "flos": 22086717646560.0, "grad_norm": 1.5631418050807935, "language_loss": 0.8138293, "learning_rate": 5.115925303080661e-07, "loss": 0.83571303, "num_input_tokens_seen": 138560970, "step": 6438, "time_per_iteration": 2.4648473262786865 }, { "auxiliary_loss_clip": 0.01138592, "auxiliary_loss_mlp": 0.01023502, "balance_loss_clip": 1.04448342, "balance_loss_mlp": 1.01653743, "epoch": 0.7742439728251067, "flos": 19864781351040.0, "grad_norm": 1.991321927707907, "language_loss": 0.78698218, "learning_rate": 5.110723270061899e-07, "loss": 0.80860317, "num_input_tokens_seen": 138577460, "step": 6439, "time_per_iteration": 2.479794979095459 }, { "auxiliary_loss_clip": 0.01167361, "auxiliary_loss_mlp": 0.01026196, "balance_loss_clip": 1.04728842, "balance_loss_mlp": 1.01940072, "epoch": 0.7743642157157458, "flos": 16690170120000.0, "grad_norm": 1.694293664322886, "language_loss": 0.79337585, "learning_rate": 5.105523495734572e-07, "loss": 0.81531137, "num_input_tokens_seen": 138594860, "step": 6440, "time_per_iteration": 2.410759687423706 }, { "auxiliary_loss_clip": 0.01168721, "auxiliary_loss_mlp": 0.01026427, "balance_loss_clip": 1.04724288, "balance_loss_mlp": 1.01965594, "epoch": 0.7744844586063849, "flos": 20304364198560.0, "grad_norm": 1.766298247452254, "language_loss": 0.75184035, "learning_rate": 5.100325980887499e-07, "loss": 0.77379191, "num_input_tokens_seen": 138614785, "step": 6441, "time_per_iteration": 2.4693617820739746 }, { "auxiliary_loss_clip": 0.01147767, "auxiliary_loss_mlp": 0.01027611, "balance_loss_clip": 1.04689765, "balance_loss_mlp": 1.02100933, "epoch": 0.774604701497024, "flos": 22966709431680.0, "grad_norm": 1.8766860255010396, "language_loss": 0.83104593, "learning_rate": 5.095130726309116e-07, "loss": 0.85279965, "num_input_tokens_seen": 138634960, "step": 6442, "time_per_iteration": 2.503903388977051 }, { "auxiliary_loss_clip": 0.0106212, "auxiliary_loss_mlp": 0.01000764, "balance_loss_clip": 1.00645494, "balance_loss_mlp": 0.99948829, "epoch": 0.774724944387663, "flos": 60288533428800.0, "grad_norm": 0.7883504134450972, "language_loss": 0.59097064, "learning_rate": 5.089937732787559e-07, "loss": 0.61159945, "num_input_tokens_seen": 138699520, "step": 6443, "time_per_iteration": 3.109330415725708 }, { "auxiliary_loss_clip": 0.01128102, "auxiliary_loss_mlp": 0.01023627, "balance_loss_clip": 1.04303694, "balance_loss_mlp": 1.01595616, "epoch": 0.7748451872783022, "flos": 26761031040000.0, "grad_norm": 2.1086994157501073, "language_loss": 0.66322982, "learning_rate": 5.084747001110592e-07, "loss": 0.6847471, "num_input_tokens_seen": 138719145, "step": 6444, "time_per_iteration": 3.3263022899627686 }, { "auxiliary_loss_clip": 0.01152007, "auxiliary_loss_mlp": 0.0076202, "balance_loss_clip": 1.04900002, "balance_loss_mlp": 1.00057304, "epoch": 0.7749654301689413, "flos": 30338632124640.0, "grad_norm": 1.9925529283988992, "language_loss": 0.70303035, "learning_rate": 5.07955853206564e-07, "loss": 0.72217059, "num_input_tokens_seen": 138743850, "step": 6445, "time_per_iteration": 2.576298952102661 }, { "auxiliary_loss_clip": 0.01159965, "auxiliary_loss_mlp": 0.01026185, "balance_loss_clip": 1.04733455, "balance_loss_mlp": 1.01906812, "epoch": 0.7750856730595803, "flos": 43179857328480.0, "grad_norm": 1.5722914470540297, "language_loss": 0.71123147, "learning_rate": 5.074372326439807e-07, "loss": 0.73309296, "num_input_tokens_seen": 138766860, "step": 6446, "time_per_iteration": 2.686169147491455 }, { "auxiliary_loss_clip": 0.01127769, "auxiliary_loss_mlp": 0.01027881, "balance_loss_clip": 1.04353178, "balance_loss_mlp": 1.02087724, "epoch": 0.7752059159502195, "flos": 17640043532640.0, "grad_norm": 3.6426606695947235, "language_loss": 0.73464012, "learning_rate": 5.069188385019814e-07, "loss": 0.75619662, "num_input_tokens_seen": 138784560, "step": 6447, "time_per_iteration": 2.5014495849609375 }, { "auxiliary_loss_clip": 0.01118571, "auxiliary_loss_mlp": 0.01023845, "balance_loss_clip": 1.04111409, "balance_loss_mlp": 1.01625109, "epoch": 0.7753261588408585, "flos": 12677691309600.0, "grad_norm": 14.003717993102482, "language_loss": 0.60686588, "learning_rate": 5.064006708592077e-07, "loss": 0.62829, "num_input_tokens_seen": 138800805, "step": 6448, "time_per_iteration": 2.5567851066589355 }, { "auxiliary_loss_clip": 0.01136603, "auxiliary_loss_mlp": 0.01023457, "balance_loss_clip": 1.046924, "balance_loss_mlp": 1.01657569, "epoch": 0.7754464017314976, "flos": 16690744791360.0, "grad_norm": 2.5058015787288026, "language_loss": 0.75465423, "learning_rate": 5.058827297942641e-07, "loss": 0.77625483, "num_input_tokens_seen": 138815910, "step": 6449, "time_per_iteration": 2.466454029083252 }, { "auxiliary_loss_clip": 0.01146383, "auxiliary_loss_mlp": 0.01027932, "balance_loss_clip": 1.04709792, "balance_loss_mlp": 1.02145338, "epoch": 0.7755666446221368, "flos": 19718948129760.0, "grad_norm": 2.7002309624208616, "language_loss": 0.75017756, "learning_rate": 5.053650153857237e-07, "loss": 0.77192068, "num_input_tokens_seen": 138834920, "step": 6450, "time_per_iteration": 2.5224967002868652 }, { "auxiliary_loss_clip": 0.01155779, "auxiliary_loss_mlp": 0.01025037, "balance_loss_clip": 1.04860401, "balance_loss_mlp": 1.01761079, "epoch": 0.7756868875127758, "flos": 18693625960800.0, "grad_norm": 1.6095105844882023, "language_loss": 0.70035326, "learning_rate": 5.048475277121214e-07, "loss": 0.72216141, "num_input_tokens_seen": 138852135, "step": 6451, "time_per_iteration": 2.4631259441375732 }, { "auxiliary_loss_clip": 0.01154888, "auxiliary_loss_mlp": 0.01021169, "balance_loss_clip": 1.04621077, "balance_loss_mlp": 1.01414418, "epoch": 0.7758071304034149, "flos": 28404195904320.0, "grad_norm": 1.9102571495923433, "language_loss": 0.77222347, "learning_rate": 5.043302668519598e-07, "loss": 0.79398406, "num_input_tokens_seen": 138871470, "step": 6452, "time_per_iteration": 2.5550639629364014 }, { "auxiliary_loss_clip": 0.01160358, "auxiliary_loss_mlp": 0.01025567, "balance_loss_clip": 1.0472765, "balance_loss_mlp": 1.01865888, "epoch": 0.775927373294054, "flos": 20595348218880.0, "grad_norm": 1.750834824088589, "language_loss": 0.72175741, "learning_rate": 5.038132328837079e-07, "loss": 0.7436167, "num_input_tokens_seen": 138889860, "step": 6453, "time_per_iteration": 2.4671688079833984 }, { "auxiliary_loss_clip": 0.01156853, "auxiliary_loss_mlp": 0.01028479, "balance_loss_clip": 1.04648685, "balance_loss_mlp": 1.02149928, "epoch": 0.7760476161846931, "flos": 22526372328000.0, "grad_norm": 1.8382815975150957, "language_loss": 0.73748255, "learning_rate": 5.032964258857993e-07, "loss": 0.75933588, "num_input_tokens_seen": 138909955, "step": 6454, "time_per_iteration": 2.4844558238983154 }, { "auxiliary_loss_clip": 0.01154065, "auxiliary_loss_mlp": 0.01023466, "balance_loss_clip": 1.04308963, "balance_loss_mlp": 1.01639128, "epoch": 0.7761678590753321, "flos": 48651494441760.0, "grad_norm": 1.5832653859781798, "language_loss": 0.68045884, "learning_rate": 5.027798459366329e-07, "loss": 0.70223415, "num_input_tokens_seen": 138935320, "step": 6455, "time_per_iteration": 3.474670171737671 }, { "auxiliary_loss_clip": 0.01158433, "auxiliary_loss_mlp": 0.0102702, "balance_loss_clip": 1.04492486, "balance_loss_mlp": 1.01986742, "epoch": 0.7762881019659713, "flos": 26177051649600.0, "grad_norm": 1.7017308456114244, "language_loss": 0.63786626, "learning_rate": 5.02263493114573e-07, "loss": 0.65972078, "num_input_tokens_seen": 138957115, "step": 6456, "time_per_iteration": 2.5510141849517822 }, { "auxiliary_loss_clip": 0.01168196, "auxiliary_loss_mlp": 0.0102405, "balance_loss_clip": 1.04730856, "balance_loss_mlp": 1.01730227, "epoch": 0.7764083448566104, "flos": 20588344411680.0, "grad_norm": 2.2489997745460997, "language_loss": 0.7697795, "learning_rate": 5.017473674979502e-07, "loss": 0.79170197, "num_input_tokens_seen": 138973140, "step": 6457, "time_per_iteration": 2.4117114543914795 }, { "auxiliary_loss_clip": 0.0102931, "auxiliary_loss_mlp": 0.01002383, "balance_loss_clip": 1.01128936, "balance_loss_mlp": 1.00121498, "epoch": 0.7765285877472494, "flos": 67293089504160.0, "grad_norm": 0.7421862136994987, "language_loss": 0.58317447, "learning_rate": 5.01231469165061e-07, "loss": 0.60349143, "num_input_tokens_seen": 139028965, "step": 6458, "time_per_iteration": 3.0279006958007812 }, { "auxiliary_loss_clip": 0.01052706, "auxiliary_loss_mlp": 0.01002081, "balance_loss_clip": 1.00602007, "balance_loss_mlp": 1.00073361, "epoch": 0.7766488306378886, "flos": 61344486376320.0, "grad_norm": 0.8324120045268987, "language_loss": 0.56902075, "learning_rate": 5.007157981941663e-07, "loss": 0.58956861, "num_input_tokens_seen": 139094325, "step": 6459, "time_per_iteration": 4.677664518356323 }, { "auxiliary_loss_clip": 0.01042568, "auxiliary_loss_mlp": 0.01002347, "balance_loss_clip": 1.0063374, "balance_loss_mlp": 1.00095868, "epoch": 0.7767690735285276, "flos": 62946209457120.0, "grad_norm": 0.884240015066597, "language_loss": 0.67442143, "learning_rate": 5.002003546634928e-07, "loss": 0.69487059, "num_input_tokens_seen": 139150425, "step": 6460, "time_per_iteration": 3.054136037826538 }, { "auxiliary_loss_clip": 0.01115284, "auxiliary_loss_mlp": 0.01027932, "balance_loss_clip": 1.04746926, "balance_loss_mlp": 1.02103877, "epoch": 0.7768893164191667, "flos": 20886404073120.0, "grad_norm": 2.110473951388548, "language_loss": 0.76231217, "learning_rate": 4.996851386512331e-07, "loss": 0.78374434, "num_input_tokens_seen": 139169130, "step": 6461, "time_per_iteration": 2.5746049880981445 }, { "auxiliary_loss_clip": 0.01142777, "auxiliary_loss_mlp": 0.01027757, "balance_loss_clip": 1.04744458, "balance_loss_mlp": 1.02000833, "epoch": 0.7770095593098058, "flos": 20704588446240.0, "grad_norm": 1.7692401942722809, "language_loss": 0.83041823, "learning_rate": 4.991701502355444e-07, "loss": 0.85212356, "num_input_tokens_seen": 139189595, "step": 6462, "time_per_iteration": 2.52923321723938 }, { "auxiliary_loss_clip": 0.01159339, "auxiliary_loss_mlp": 0.01024379, "balance_loss_clip": 1.04684544, "balance_loss_mlp": 1.01773, "epoch": 0.7771298022004449, "flos": 24717713761920.0, "grad_norm": 1.648994118758253, "language_loss": 0.76069903, "learning_rate": 4.986553894945518e-07, "loss": 0.78253621, "num_input_tokens_seen": 139210805, "step": 6463, "time_per_iteration": 2.5112545490264893 }, { "auxiliary_loss_clip": 0.01110341, "auxiliary_loss_mlp": 0.01024064, "balance_loss_clip": 1.0409441, "balance_loss_mlp": 1.01780868, "epoch": 0.777250045091084, "flos": 25009236536640.0, "grad_norm": 2.0535212261181806, "language_loss": 0.86077827, "learning_rate": 4.981408565063416e-07, "loss": 0.88212228, "num_input_tokens_seen": 139230750, "step": 6464, "time_per_iteration": 2.602144718170166 }, { "auxiliary_loss_clip": 0.01169297, "auxiliary_loss_mlp": 0.01030035, "balance_loss_clip": 1.04666781, "balance_loss_mlp": 1.02336538, "epoch": 0.777370287981723, "flos": 20119890716640.0, "grad_norm": 1.7969531536241394, "language_loss": 0.75932312, "learning_rate": 4.976265513489701e-07, "loss": 0.78131646, "num_input_tokens_seen": 139250720, "step": 6465, "time_per_iteration": 2.4361424446105957 }, { "auxiliary_loss_clip": 0.01152138, "auxiliary_loss_mlp": 0.01025261, "balance_loss_clip": 1.04395235, "balance_loss_mlp": 1.01794434, "epoch": 0.7774905308723622, "flos": 21718812274560.0, "grad_norm": 1.9812607729378517, "language_loss": 0.80629081, "learning_rate": 4.971124741004562e-07, "loss": 0.8280648, "num_input_tokens_seen": 139269720, "step": 6466, "time_per_iteration": 2.5046722888946533 }, { "auxiliary_loss_clip": 0.01155164, "auxiliary_loss_mlp": 0.01028418, "balance_loss_clip": 1.04623532, "balance_loss_mlp": 1.02113152, "epoch": 0.7776107737630013, "flos": 16034118167520.0, "grad_norm": 2.1074416796811706, "language_loss": 0.76210093, "learning_rate": 4.965986248387846e-07, "loss": 0.7839368, "num_input_tokens_seen": 139288035, "step": 6467, "time_per_iteration": 2.4457359313964844 }, { "auxiliary_loss_clip": 0.01142464, "auxiliary_loss_mlp": 0.01021332, "balance_loss_clip": 1.04374576, "balance_loss_mlp": 1.0140841, "epoch": 0.7777310166536403, "flos": 24790899749760.0, "grad_norm": 1.6958171658885985, "language_loss": 0.77336943, "learning_rate": 4.960850036419073e-07, "loss": 0.79500741, "num_input_tokens_seen": 139307135, "step": 6468, "time_per_iteration": 2.5543458461761475 }, { "auxiliary_loss_clip": 0.01139915, "auxiliary_loss_mlp": 0.01031589, "balance_loss_clip": 1.04625416, "balance_loss_mlp": 1.02417123, "epoch": 0.7778512595442795, "flos": 17272533247200.0, "grad_norm": 1.9312245317950634, "language_loss": 0.78561974, "learning_rate": 4.955716105877378e-07, "loss": 0.80733478, "num_input_tokens_seen": 139325905, "step": 6469, "time_per_iteration": 2.492739677429199 }, { "auxiliary_loss_clip": 0.01158343, "auxiliary_loss_mlp": 0.00761944, "balance_loss_clip": 1.04669094, "balance_loss_mlp": 1.00054932, "epoch": 0.7779715024349185, "flos": 17748421752960.0, "grad_norm": 1.625342570346492, "language_loss": 0.82948548, "learning_rate": 4.950584457541598e-07, "loss": 0.8486883, "num_input_tokens_seen": 139344370, "step": 6470, "time_per_iteration": 3.1733384132385254 }, { "auxiliary_loss_clip": 0.01155293, "auxiliary_loss_mlp": 0.01022325, "balance_loss_clip": 1.04380536, "balance_loss_mlp": 1.01583064, "epoch": 0.7780917453255576, "flos": 24316878925920.0, "grad_norm": 1.5056139426362343, "language_loss": 0.818811, "learning_rate": 4.945455092190183e-07, "loss": 0.84058714, "num_input_tokens_seen": 139365625, "step": 6471, "time_per_iteration": 2.5368294715881348 }, { "auxiliary_loss_clip": 0.0106221, "auxiliary_loss_mlp": 0.00999769, "balance_loss_clip": 1.00651741, "balance_loss_mlp": 0.99837416, "epoch": 0.7782119882161967, "flos": 56364613065120.0, "grad_norm": 0.6880804863724294, "language_loss": 0.56035352, "learning_rate": 4.940328010601271e-07, "loss": 0.58097327, "num_input_tokens_seen": 139430540, "step": 6472, "time_per_iteration": 3.086036205291748 }, { "auxiliary_loss_clip": 0.01150136, "auxiliary_loss_mlp": 0.01031367, "balance_loss_clip": 1.05217361, "balance_loss_mlp": 1.02318597, "epoch": 0.7783322311068358, "flos": 46789992790560.0, "grad_norm": 1.7075135349014985, "language_loss": 0.76299053, "learning_rate": 4.935203213552621e-07, "loss": 0.7848056, "num_input_tokens_seen": 139454280, "step": 6473, "time_per_iteration": 2.760037422180176 }, { "auxiliary_loss_clip": 0.0114486, "auxiliary_loss_mlp": 0.01026359, "balance_loss_clip": 1.0478332, "balance_loss_mlp": 1.01889014, "epoch": 0.7784524739974749, "flos": 19057867802880.0, "grad_norm": 2.2697977907135245, "language_loss": 0.66619527, "learning_rate": 4.930080701821662e-07, "loss": 0.68790752, "num_input_tokens_seen": 139471745, "step": 6474, "time_per_iteration": 2.4976606369018555 }, { "auxiliary_loss_clip": 0.01142919, "auxiliary_loss_mlp": 0.01027636, "balance_loss_clip": 1.04520369, "balance_loss_mlp": 1.02060556, "epoch": 0.778572716888114, "flos": 24791115251520.0, "grad_norm": 2.412961316390256, "language_loss": 0.77065051, "learning_rate": 4.92496047618548e-07, "loss": 0.79235601, "num_input_tokens_seen": 139491505, "step": 6475, "time_per_iteration": 2.52616024017334 }, { "auxiliary_loss_clip": 0.01158251, "auxiliary_loss_mlp": 0.0102254, "balance_loss_clip": 1.04889166, "balance_loss_mlp": 1.01569796, "epoch": 0.7786929597787531, "flos": 20078089763520.0, "grad_norm": 1.7599367598647095, "language_loss": 0.7765429, "learning_rate": 4.919842537420811e-07, "loss": 0.79835081, "num_input_tokens_seen": 139508620, "step": 6476, "time_per_iteration": 2.4607386589050293 }, { "auxiliary_loss_clip": 0.01144106, "auxiliary_loss_mlp": 0.01027953, "balance_loss_clip": 1.04808092, "balance_loss_mlp": 1.02088082, "epoch": 0.7788132026693921, "flos": 21872223974400.0, "grad_norm": 1.5797587757208282, "language_loss": 0.79703647, "learning_rate": 4.91472688630404e-07, "loss": 0.81875706, "num_input_tokens_seen": 139529360, "step": 6477, "time_per_iteration": 2.5149199962615967 }, { "auxiliary_loss_clip": 0.01169164, "auxiliary_loss_mlp": 0.01024959, "balance_loss_clip": 1.0503509, "balance_loss_mlp": 1.01806307, "epoch": 0.7789334455600313, "flos": 11181940012800.0, "grad_norm": 1.7075146167954078, "language_loss": 0.74107993, "learning_rate": 4.909613523611202e-07, "loss": 0.76302117, "num_input_tokens_seen": 139546240, "step": 6478, "time_per_iteration": 2.430480718612671 }, { "auxiliary_loss_clip": 0.01107767, "auxiliary_loss_mlp": 0.00762337, "balance_loss_clip": 1.04084635, "balance_loss_mlp": 1.00048351, "epoch": 0.7790536884506704, "flos": 28695431343360.0, "grad_norm": 2.197058454504592, "language_loss": 0.74199659, "learning_rate": 4.904502450117991e-07, "loss": 0.76069766, "num_input_tokens_seen": 139567200, "step": 6479, "time_per_iteration": 2.632925271987915 }, { "auxiliary_loss_clip": 0.01141704, "auxiliary_loss_mlp": 0.01025921, "balance_loss_clip": 1.0488956, "balance_loss_mlp": 1.01852739, "epoch": 0.7791739313413094, "flos": 11072304698880.0, "grad_norm": 2.95548761010181, "language_loss": 0.72397196, "learning_rate": 4.899393666599762e-07, "loss": 0.74564815, "num_input_tokens_seen": 139583775, "step": 6480, "time_per_iteration": 2.4856154918670654 }, { "auxiliary_loss_clip": 0.01167707, "auxiliary_loss_mlp": 0.01021415, "balance_loss_clip": 1.04627299, "balance_loss_mlp": 1.01470649, "epoch": 0.7792941742319486, "flos": 14679279468480.0, "grad_norm": 2.6250524167269496, "language_loss": 0.73042858, "learning_rate": 4.894287173831506e-07, "loss": 0.75231981, "num_input_tokens_seen": 139599735, "step": 6481, "time_per_iteration": 3.162997007369995 }, { "auxiliary_loss_clip": 0.01142068, "auxiliary_loss_mlp": 0.01024257, "balance_loss_clip": 1.04303932, "balance_loss_mlp": 1.01694655, "epoch": 0.7794144171225876, "flos": 23258878711680.0, "grad_norm": 2.973899859195846, "language_loss": 0.84193391, "learning_rate": 4.889182972587877e-07, "loss": 0.86359715, "num_input_tokens_seen": 139619030, "step": 6482, "time_per_iteration": 2.5216243267059326 }, { "auxiliary_loss_clip": 0.01137572, "auxiliary_loss_mlp": 0.01024631, "balance_loss_clip": 1.04622388, "balance_loss_mlp": 1.01809573, "epoch": 0.7795346600132267, "flos": 21507084208320.0, "grad_norm": 2.62511967332874, "language_loss": 0.66057104, "learning_rate": 4.884081063643177e-07, "loss": 0.68219304, "num_input_tokens_seen": 139637690, "step": 6483, "time_per_iteration": 2.527604579925537 }, { "auxiliary_loss_clip": 0.01039047, "auxiliary_loss_mlp": 0.01004502, "balance_loss_clip": 1.00889468, "balance_loss_mlp": 1.00326848, "epoch": 0.7796549029038659, "flos": 70052283613440.0, "grad_norm": 0.8563848809361404, "language_loss": 0.5250963, "learning_rate": 4.878981447771353e-07, "loss": 0.54553175, "num_input_tokens_seen": 139692070, "step": 6484, "time_per_iteration": 3.8330647945404053 }, { "auxiliary_loss_clip": 0.01121811, "auxiliary_loss_mlp": 0.01026706, "balance_loss_clip": 1.04325294, "balance_loss_mlp": 1.0191927, "epoch": 0.7797751457945049, "flos": 23989409662560.0, "grad_norm": 1.5347586400325035, "language_loss": 0.72836971, "learning_rate": 4.873884125746035e-07, "loss": 0.74985486, "num_input_tokens_seen": 139713745, "step": 6485, "time_per_iteration": 3.328014850616455 }, { "auxiliary_loss_clip": 0.01133815, "auxiliary_loss_mlp": 0.01023039, "balance_loss_clip": 1.04232788, "balance_loss_mlp": 1.0163095, "epoch": 0.779895388685144, "flos": 22674755653440.0, "grad_norm": 2.3247250400565767, "language_loss": 0.7212435, "learning_rate": 4.868789098340456e-07, "loss": 0.74281204, "num_input_tokens_seen": 139731650, "step": 6486, "time_per_iteration": 2.4965267181396484 }, { "auxiliary_loss_clip": 0.0112831, "auxiliary_loss_mlp": 0.01028009, "balance_loss_clip": 1.04449034, "balance_loss_mlp": 1.0213716, "epoch": 0.7800156315757831, "flos": 23768702356320.0, "grad_norm": 2.2600608731659086, "language_loss": 0.72934884, "learning_rate": 4.863696366327543e-07, "loss": 0.75091201, "num_input_tokens_seen": 139750820, "step": 6487, "time_per_iteration": 2.5546579360961914 }, { "auxiliary_loss_clip": 0.01154815, "auxiliary_loss_mlp": 0.0102011, "balance_loss_clip": 1.04425383, "balance_loss_mlp": 1.01292181, "epoch": 0.7801358744664222, "flos": 26429718661920.0, "grad_norm": 2.4385268221312297, "language_loss": 0.78031737, "learning_rate": 4.85860593047986e-07, "loss": 0.80206656, "num_input_tokens_seen": 139770885, "step": 6488, "time_per_iteration": 2.518564224243164 }, { "auxiliary_loss_clip": 0.01118256, "auxiliary_loss_mlp": 0.01026299, "balance_loss_clip": 1.03928113, "balance_loss_mlp": 1.01943886, "epoch": 0.7802561173570612, "flos": 26322166531680.0, "grad_norm": 1.6031989352917952, "language_loss": 0.74805427, "learning_rate": 4.853517791569613e-07, "loss": 0.76949978, "num_input_tokens_seen": 139793065, "step": 6489, "time_per_iteration": 2.5935707092285156 }, { "auxiliary_loss_clip": 0.01145905, "auxiliary_loss_mlp": 0.00762381, "balance_loss_clip": 1.04547143, "balance_loss_mlp": 1.00059628, "epoch": 0.7803763602477004, "flos": 40333756952640.0, "grad_norm": 1.916953462389462, "language_loss": 0.65811688, "learning_rate": 4.848431950368684e-07, "loss": 0.67719978, "num_input_tokens_seen": 139815625, "step": 6490, "time_per_iteration": 2.6750948429107666 }, { "auxiliary_loss_clip": 0.01062145, "auxiliary_loss_mlp": 0.00752883, "balance_loss_clip": 1.00657177, "balance_loss_mlp": 0.99991506, "epoch": 0.7804966031383395, "flos": 67001458978560.0, "grad_norm": 0.7158130936101674, "language_loss": 0.55814964, "learning_rate": 4.843348407648569e-07, "loss": 0.57629991, "num_input_tokens_seen": 139876905, "step": 6491, "time_per_iteration": 3.0298447608947754 }, { "auxiliary_loss_clip": 0.01156092, "auxiliary_loss_mlp": 0.01022212, "balance_loss_clip": 1.04206276, "balance_loss_mlp": 1.014925, "epoch": 0.7806168460289785, "flos": 17740735523520.0, "grad_norm": 2.0937783153095535, "language_loss": 0.83013999, "learning_rate": 4.838267164180457e-07, "loss": 0.85192305, "num_input_tokens_seen": 139892575, "step": 6492, "time_per_iteration": 2.461310863494873 }, { "auxiliary_loss_clip": 0.01169896, "auxiliary_loss_mlp": 0.01031245, "balance_loss_clip": 1.04662824, "balance_loss_mlp": 1.02251267, "epoch": 0.7807370889196176, "flos": 23946243864960.0, "grad_norm": 3.2001796638145685, "language_loss": 0.83348083, "learning_rate": 4.833188220735156e-07, "loss": 0.85549223, "num_input_tokens_seen": 139912245, "step": 6493, "time_per_iteration": 2.458620548248291 }, { "auxiliary_loss_clip": 0.01152318, "auxiliary_loss_mlp": 0.01022653, "balance_loss_clip": 1.04533982, "balance_loss_mlp": 1.01605785, "epoch": 0.7808573318102567, "flos": 18989027767200.0, "grad_norm": 2.2867185003834707, "language_loss": 0.74772257, "learning_rate": 4.828111578083152e-07, "loss": 0.76947224, "num_input_tokens_seen": 139929150, "step": 6494, "time_per_iteration": 2.4731967449188232 }, { "auxiliary_loss_clip": 0.0114087, "auxiliary_loss_mlp": 0.01023467, "balance_loss_clip": 1.0470084, "balance_loss_mlp": 1.01678491, "epoch": 0.7809775747008958, "flos": 23980753675200.0, "grad_norm": 2.0589855377877977, "language_loss": 0.81398892, "learning_rate": 4.823037236994556e-07, "loss": 0.83563232, "num_input_tokens_seen": 139947315, "step": 6495, "time_per_iteration": 2.5124831199645996 }, { "auxiliary_loss_clip": 0.01053686, "auxiliary_loss_mlp": 0.01001523, "balance_loss_clip": 1.00686789, "balance_loss_mlp": 1.0002358, "epoch": 0.7810978175915348, "flos": 68535886452960.0, "grad_norm": 0.7147760753220985, "language_loss": 0.56374252, "learning_rate": 4.817965198239136e-07, "loss": 0.58429462, "num_input_tokens_seen": 140013775, "step": 6496, "time_per_iteration": 3.8299925327301025 }, { "auxiliary_loss_clip": 0.01126188, "auxiliary_loss_mlp": 0.01025692, "balance_loss_clip": 1.04244018, "balance_loss_mlp": 1.01838136, "epoch": 0.781218060482174, "flos": 19642134528960.0, "grad_norm": 2.154888065884148, "language_loss": 0.75026119, "learning_rate": 4.812895462586331e-07, "loss": 0.77178001, "num_input_tokens_seen": 140031600, "step": 6497, "time_per_iteration": 2.621613025665283 }, { "auxiliary_loss_clip": 0.01131767, "auxiliary_loss_mlp": 0.01022372, "balance_loss_clip": 1.04522514, "balance_loss_mlp": 1.01568413, "epoch": 0.7813383033728131, "flos": 25627869405120.0, "grad_norm": 1.6781027265760464, "language_loss": 0.81634188, "learning_rate": 4.807828030805207e-07, "loss": 0.83788323, "num_input_tokens_seen": 140050590, "step": 6498, "time_per_iteration": 2.6441657543182373 }, { "auxiliary_loss_clip": 0.01154421, "auxiliary_loss_mlp": 0.01035183, "balance_loss_clip": 1.04738617, "balance_loss_mlp": 1.02771449, "epoch": 0.7814585462634521, "flos": 20485928406720.0, "grad_norm": 1.8690982528409872, "language_loss": 0.68021303, "learning_rate": 4.802762903664495e-07, "loss": 0.7021091, "num_input_tokens_seen": 140069770, "step": 6499, "time_per_iteration": 2.4842817783355713 }, { "auxiliary_loss_clip": 0.01147038, "auxiliary_loss_mlp": 0.01032568, "balance_loss_clip": 1.04703271, "balance_loss_mlp": 1.02530468, "epoch": 0.7815787891540913, "flos": 22304300177280.0, "grad_norm": 2.192894535284383, "language_loss": 0.73947316, "learning_rate": 4.797700081932565e-07, "loss": 0.76126921, "num_input_tokens_seen": 140087635, "step": 6500, "time_per_iteration": 2.5208113193511963 }, { "auxiliary_loss_clip": 0.01093131, "auxiliary_loss_mlp": 0.01033164, "balance_loss_clip": 1.03842127, "balance_loss_mlp": 1.02624106, "epoch": 0.7816990320447303, "flos": 22600671741600.0, "grad_norm": 2.1162756192961107, "language_loss": 0.82000041, "learning_rate": 4.792639566377442e-07, "loss": 0.84126329, "num_input_tokens_seen": 140105045, "step": 6501, "time_per_iteration": 2.607816219329834 }, { "auxiliary_loss_clip": 0.01149657, "auxiliary_loss_mlp": 0.01028837, "balance_loss_clip": 1.04434872, "balance_loss_mlp": 1.0220933, "epoch": 0.7818192749353694, "flos": 24935978714880.0, "grad_norm": 1.8592741419506398, "language_loss": 0.77793461, "learning_rate": 4.78758135776681e-07, "loss": 0.79971957, "num_input_tokens_seen": 140124900, "step": 6502, "time_per_iteration": 2.502066135406494 }, { "auxiliary_loss_clip": 0.0114322, "auxiliary_loss_mlp": 0.01027352, "balance_loss_clip": 1.04668725, "balance_loss_mlp": 1.02077174, "epoch": 0.7819395178260086, "flos": 23733043203360.0, "grad_norm": 1.9157908818199907, "language_loss": 0.7892065, "learning_rate": 4.782525456867989e-07, "loss": 0.81091225, "num_input_tokens_seen": 140143755, "step": 6503, "time_per_iteration": 2.518419027328491 }, { "auxiliary_loss_clip": 0.01127564, "auxiliary_loss_mlp": 0.01027125, "balance_loss_clip": 1.04506242, "balance_loss_mlp": 1.01993966, "epoch": 0.7820597607166476, "flos": 23221675129440.0, "grad_norm": 1.5906976463903018, "language_loss": 0.83302724, "learning_rate": 4.777471864447959e-07, "loss": 0.85457408, "num_input_tokens_seen": 140164495, "step": 6504, "time_per_iteration": 2.565498113632202 }, { "auxiliary_loss_clip": 0.01142323, "auxiliary_loss_mlp": 0.01033852, "balance_loss_clip": 1.04385662, "balance_loss_mlp": 1.02674413, "epoch": 0.7821800036072867, "flos": 22309544053440.0, "grad_norm": 2.133860616782824, "language_loss": 0.80517924, "learning_rate": 4.772420581273344e-07, "loss": 0.82694101, "num_input_tokens_seen": 140181980, "step": 6505, "time_per_iteration": 2.517754077911377 }, { "auxiliary_loss_clip": 0.01152354, "auxiliary_loss_mlp": 0.01025617, "balance_loss_clip": 1.04784417, "balance_loss_mlp": 1.01901591, "epoch": 0.7823002464979258, "flos": 21544180039680.0, "grad_norm": 1.8878099974734655, "language_loss": 0.76392484, "learning_rate": 4.7673716081104134e-07, "loss": 0.78570455, "num_input_tokens_seen": 140202155, "step": 6506, "time_per_iteration": 2.4987518787384033 }, { "auxiliary_loss_clip": 0.01154611, "auxiliary_loss_mlp": 0.01023651, "balance_loss_clip": 1.04736137, "balance_loss_mlp": 1.0171895, "epoch": 0.7824204893885649, "flos": 24535646716320.0, "grad_norm": 2.2102690903331084, "language_loss": 0.84487057, "learning_rate": 4.762324945725109e-07, "loss": 0.8666532, "num_input_tokens_seen": 140221600, "step": 6507, "time_per_iteration": 3.2786834239959717 }, { "auxiliary_loss_clip": 0.01140251, "auxiliary_loss_mlp": 0.01026129, "balance_loss_clip": 1.04889977, "balance_loss_mlp": 1.01933718, "epoch": 0.782540732279204, "flos": 27415215310560.0, "grad_norm": 1.706235882607515, "language_loss": 0.75475454, "learning_rate": 4.7572805948829844e-07, "loss": 0.77641833, "num_input_tokens_seen": 140241860, "step": 6508, "time_per_iteration": 2.542966365814209 }, { "auxiliary_loss_clip": 0.01115357, "auxiliary_loss_mlp": 0.01023944, "balance_loss_clip": 1.04098821, "balance_loss_mlp": 1.01672602, "epoch": 0.7826609751698431, "flos": 24353220501120.0, "grad_norm": 1.935104304461823, "language_loss": 0.7059837, "learning_rate": 4.7522385563492795e-07, "loss": 0.72737682, "num_input_tokens_seen": 140262160, "step": 6509, "time_per_iteration": 3.3789799213409424 }, { "auxiliary_loss_clip": 0.01131797, "auxiliary_loss_mlp": 0.01026448, "balance_loss_clip": 1.0453527, "balance_loss_mlp": 1.01956344, "epoch": 0.7827812180604822, "flos": 23988547655520.0, "grad_norm": 1.795212905454833, "language_loss": 0.70225108, "learning_rate": 4.747198830888863e-07, "loss": 0.72383344, "num_input_tokens_seen": 140282030, "step": 6510, "time_per_iteration": 2.563582181930542 }, { "auxiliary_loss_clip": 0.01137473, "auxiliary_loss_mlp": 0.01025611, "balance_loss_clip": 1.04531312, "balance_loss_mlp": 1.01866364, "epoch": 0.7829014609511212, "flos": 27454322491680.0, "grad_norm": 2.0139716214356316, "language_loss": 0.68559241, "learning_rate": 4.742161419266251e-07, "loss": 0.7072233, "num_input_tokens_seen": 140301190, "step": 6511, "time_per_iteration": 3.35191011428833 }, { "auxiliary_loss_clip": 0.01162384, "auxiliary_loss_mlp": 0.01026456, "balance_loss_clip": 1.04895258, "balance_loss_mlp": 1.01903558, "epoch": 0.7830217038417604, "flos": 29204536648800.0, "grad_norm": 3.193315671363297, "language_loss": 0.64935946, "learning_rate": 4.7371263222456304e-07, "loss": 0.6712479, "num_input_tokens_seen": 140318510, "step": 6512, "time_per_iteration": 2.539724349975586 }, { "auxiliary_loss_clip": 0.0105024, "auxiliary_loss_mlp": 0.01001886, "balance_loss_clip": 1.00768602, "balance_loss_mlp": 1.00054491, "epoch": 0.7831419467323995, "flos": 60950906766240.0, "grad_norm": 0.8005244193768173, "language_loss": 0.61517608, "learning_rate": 4.7320935405908004e-07, "loss": 0.63569736, "num_input_tokens_seen": 140379380, "step": 6513, "time_per_iteration": 3.0788493156433105 }, { "auxiliary_loss_clip": 0.01174016, "auxiliary_loss_mlp": 0.01031345, "balance_loss_clip": 1.04903948, "balance_loss_mlp": 1.02413583, "epoch": 0.7832621896230385, "flos": 19682534720640.0, "grad_norm": 2.0638205749031893, "language_loss": 0.8424654, "learning_rate": 4.7270630750652475e-07, "loss": 0.86451906, "num_input_tokens_seen": 140395335, "step": 6514, "time_per_iteration": 2.4523367881774902 }, { "auxiliary_loss_clip": 0.01153751, "auxiliary_loss_mlp": 0.01026536, "balance_loss_clip": 1.04571342, "balance_loss_mlp": 1.0198487, "epoch": 0.7833824325136777, "flos": 25009236536640.0, "grad_norm": 2.2349888277002554, "language_loss": 0.80261147, "learning_rate": 4.7220349264320746e-07, "loss": 0.82441431, "num_input_tokens_seen": 140414420, "step": 6515, "time_per_iteration": 2.4993069171905518 }, { "auxiliary_loss_clip": 0.01051184, "auxiliary_loss_mlp": 0.01002544, "balance_loss_clip": 1.0062654, "balance_loss_mlp": 1.00125647, "epoch": 0.7835026754043167, "flos": 68800154643360.0, "grad_norm": 0.7387661540828926, "language_loss": 0.54975563, "learning_rate": 4.71700909545407e-07, "loss": 0.57029289, "num_input_tokens_seen": 140477365, "step": 6516, "time_per_iteration": 3.083906650543213 }, { "auxiliary_loss_clip": 0.01154924, "auxiliary_loss_mlp": 0.0102434, "balance_loss_clip": 1.04493141, "balance_loss_mlp": 1.0174551, "epoch": 0.7836229182949558, "flos": 19864601766240.0, "grad_norm": 2.051510191867642, "language_loss": 0.76837111, "learning_rate": 4.711985582893627e-07, "loss": 0.79016376, "num_input_tokens_seen": 140495885, "step": 6517, "time_per_iteration": 2.4661383628845215 }, { "auxiliary_loss_clip": 0.01115437, "auxiliary_loss_mlp": 0.01028765, "balance_loss_clip": 1.04102015, "balance_loss_mlp": 1.02128434, "epoch": 0.783743161185595, "flos": 22965847424640.0, "grad_norm": 1.7133900736700636, "language_loss": 0.71468663, "learning_rate": 4.706964389512811e-07, "loss": 0.73612863, "num_input_tokens_seen": 140515920, "step": 6518, "time_per_iteration": 2.5944700241088867 }, { "auxiliary_loss_clip": 0.01165334, "auxiliary_loss_mlp": 0.01019397, "balance_loss_clip": 1.04675269, "balance_loss_mlp": 1.01279926, "epoch": 0.783863404076234, "flos": 12458492515680.0, "grad_norm": 2.8867399260120528, "language_loss": 0.87546456, "learning_rate": 4.701945516073345e-07, "loss": 0.89731187, "num_input_tokens_seen": 140533395, "step": 6519, "time_per_iteration": 2.42826771736145 }, { "auxiliary_loss_clip": 0.01123848, "auxiliary_loss_mlp": 0.01021763, "balance_loss_clip": 1.0435313, "balance_loss_mlp": 1.0149169, "epoch": 0.7839836469668731, "flos": 24243944356800.0, "grad_norm": 1.858230099506036, "language_loss": 0.74906999, "learning_rate": 4.696928963336577e-07, "loss": 0.77052617, "num_input_tokens_seen": 140552825, "step": 6520, "time_per_iteration": 2.577796697616577 }, { "auxiliary_loss_clip": 0.01050144, "auxiliary_loss_mlp": 0.0100198, "balance_loss_clip": 1.0074048, "balance_loss_mlp": 1.000615, "epoch": 0.7841038898575122, "flos": 62121990322560.0, "grad_norm": 0.9092929137356344, "language_loss": 0.61064118, "learning_rate": 4.6919147320635224e-07, "loss": 0.63116246, "num_input_tokens_seen": 140615535, "step": 6521, "time_per_iteration": 3.046462059020996 }, { "auxiliary_loss_clip": 0.01156279, "auxiliary_loss_mlp": 0.01024625, "balance_loss_clip": 1.04582024, "balance_loss_mlp": 1.01733804, "epoch": 0.7842241327481513, "flos": 20193902794560.0, "grad_norm": 1.9859146854338852, "language_loss": 0.7322678, "learning_rate": 4.6869028230148286e-07, "loss": 0.75407684, "num_input_tokens_seen": 140633330, "step": 6522, "time_per_iteration": 3.2383639812469482 }, { "auxiliary_loss_clip": 0.01118844, "auxiliary_loss_mlp": 0.01029383, "balance_loss_clip": 1.03866315, "balance_loss_mlp": 1.0224328, "epoch": 0.7843443756387903, "flos": 28074535706400.0, "grad_norm": 5.066197401809337, "language_loss": 0.59640682, "learning_rate": 4.6818932369507957e-07, "loss": 0.61788911, "num_input_tokens_seen": 140652830, "step": 6523, "time_per_iteration": 2.5959229469299316 }, { "auxiliary_loss_clip": 0.01153535, "auxiliary_loss_mlp": 0.01035169, "balance_loss_clip": 1.04622316, "balance_loss_mlp": 1.02813578, "epoch": 0.7844646185294295, "flos": 21323400899520.0, "grad_norm": 1.9834585854007678, "language_loss": 0.88958848, "learning_rate": 4.676885974631386e-07, "loss": 0.91147554, "num_input_tokens_seen": 140671190, "step": 6524, "time_per_iteration": 2.486891269683838 }, { "auxiliary_loss_clip": 0.01156962, "auxiliary_loss_mlp": 0.01028451, "balance_loss_clip": 1.04786444, "balance_loss_mlp": 1.021433, "epoch": 0.7845848614200686, "flos": 23656588772160.0, "grad_norm": 2.241004890345847, "language_loss": 0.81252575, "learning_rate": 4.67188103681619e-07, "loss": 0.83437991, "num_input_tokens_seen": 140690975, "step": 6525, "time_per_iteration": 2.54384446144104 }, { "auxiliary_loss_clip": 0.01153195, "auxiliary_loss_mlp": 0.00762056, "balance_loss_clip": 1.04798734, "balance_loss_mlp": 1.0005095, "epoch": 0.7847051043107076, "flos": 23402197745760.0, "grad_norm": 2.358481556469374, "language_loss": 0.69435263, "learning_rate": 4.666878424264453e-07, "loss": 0.71350515, "num_input_tokens_seen": 140710930, "step": 6526, "time_per_iteration": 2.5118257999420166 }, { "auxiliary_loss_clip": 0.01130899, "auxiliary_loss_mlp": 0.01022141, "balance_loss_clip": 1.04368544, "balance_loss_mlp": 1.01588583, "epoch": 0.7848253472013467, "flos": 19022280483840.0, "grad_norm": 1.5539717030494282, "language_loss": 0.73842478, "learning_rate": 4.661878137735069e-07, "loss": 0.75995517, "num_input_tokens_seen": 140729120, "step": 6527, "time_per_iteration": 2.516260862350464 }, { "auxiliary_loss_clip": 0.01141352, "auxiliary_loss_mlp": 0.01029326, "balance_loss_clip": 1.04617357, "balance_loss_mlp": 1.02262044, "epoch": 0.7849455900919858, "flos": 21179183941440.0, "grad_norm": 1.9694220412659034, "language_loss": 0.74613672, "learning_rate": 4.656880177986571e-07, "loss": 0.76784348, "num_input_tokens_seen": 140747665, "step": 6528, "time_per_iteration": 2.5097408294677734 }, { "auxiliary_loss_clip": 0.01145005, "auxiliary_loss_mlp": 0.01023226, "balance_loss_clip": 1.04399467, "balance_loss_mlp": 1.01638627, "epoch": 0.7850658329826249, "flos": 19536485997600.0, "grad_norm": 1.9282722321424473, "language_loss": 0.81489068, "learning_rate": 4.6518845457771607e-07, "loss": 0.836573, "num_input_tokens_seen": 140766525, "step": 6529, "time_per_iteration": 2.517674207687378 }, { "auxiliary_loss_clip": 0.01149165, "auxiliary_loss_mlp": 0.00761997, "balance_loss_clip": 1.04597449, "balance_loss_mlp": 1.00047076, "epoch": 0.7851860758732639, "flos": 12495336928320.0, "grad_norm": 1.868777113848222, "language_loss": 0.79444981, "learning_rate": 4.646891241864652e-07, "loss": 0.81356144, "num_input_tokens_seen": 140785090, "step": 6530, "time_per_iteration": 2.4891529083251953 }, { "auxiliary_loss_clip": 0.01153677, "auxiliary_loss_mlp": 0.01028536, "balance_loss_clip": 1.04562736, "balance_loss_mlp": 1.0213505, "epoch": 0.7853063187639031, "flos": 22960962718080.0, "grad_norm": 1.9435244121923776, "language_loss": 0.73086512, "learning_rate": 4.6419002670065397e-07, "loss": 0.75268722, "num_input_tokens_seen": 140804670, "step": 6531, "time_per_iteration": 2.4874627590179443 }, { "auxiliary_loss_clip": 0.01132587, "auxiliary_loss_mlp": 0.01021743, "balance_loss_clip": 1.04617906, "balance_loss_mlp": 1.01483822, "epoch": 0.7854265616545422, "flos": 17347263664320.0, "grad_norm": 2.1216118713886565, "language_loss": 0.86314386, "learning_rate": 4.6369116219599445e-07, "loss": 0.88468719, "num_input_tokens_seen": 140820655, "step": 6532, "time_per_iteration": 2.5061450004577637 }, { "auxiliary_loss_clip": 0.01125592, "auxiliary_loss_mlp": 0.01023729, "balance_loss_clip": 1.04199004, "balance_loss_mlp": 1.01689231, "epoch": 0.7855468045451812, "flos": 23838296648160.0, "grad_norm": 1.5408765384522907, "language_loss": 0.79179525, "learning_rate": 4.631925307481637e-07, "loss": 0.81328857, "num_input_tokens_seen": 140840470, "step": 6533, "time_per_iteration": 3.3265738487243652 }, { "auxiliary_loss_clip": 0.01141936, "auxiliary_loss_mlp": 0.01020275, "balance_loss_clip": 1.04793763, "balance_loss_mlp": 1.01407039, "epoch": 0.7856670474358204, "flos": 25666797001440.0, "grad_norm": 2.0122502891115652, "language_loss": 0.75705731, "learning_rate": 4.6269413243280533e-07, "loss": 0.77867949, "num_input_tokens_seen": 140859890, "step": 6534, "time_per_iteration": 2.5475850105285645 }, { "auxiliary_loss_clip": 0.01146992, "auxiliary_loss_mlp": 0.01027032, "balance_loss_clip": 1.04777336, "balance_loss_mlp": 1.01957595, "epoch": 0.7857872903264594, "flos": 18144659218080.0, "grad_norm": 2.4327191941975497, "language_loss": 0.73521602, "learning_rate": 4.621959673255236e-07, "loss": 0.75695628, "num_input_tokens_seen": 140876190, "step": 6535, "time_per_iteration": 3.2473554611206055 }, { "auxiliary_loss_clip": 0.01111053, "auxiliary_loss_mlp": 0.0102193, "balance_loss_clip": 1.04233384, "balance_loss_mlp": 1.01513481, "epoch": 0.7859075332170985, "flos": 14386140430560.0, "grad_norm": 2.3259133638326825, "language_loss": 0.9043628, "learning_rate": 4.6169803550189135e-07, "loss": 0.92569256, "num_input_tokens_seen": 140891885, "step": 6536, "time_per_iteration": 3.298038959503174 }, { "auxiliary_loss_clip": 0.01112225, "auxiliary_loss_mlp": 0.01025248, "balance_loss_clip": 1.04635739, "balance_loss_mlp": 1.01715946, "epoch": 0.7860277761077377, "flos": 19864063011840.0, "grad_norm": 2.189389655519155, "language_loss": 0.77268708, "learning_rate": 4.6120033703744355e-07, "loss": 0.79406178, "num_input_tokens_seen": 140910780, "step": 6537, "time_per_iteration": 2.5778846740722656 }, { "auxiliary_loss_clip": 0.01130614, "auxiliary_loss_mlp": 0.01018906, "balance_loss_clip": 1.04264486, "balance_loss_mlp": 1.01208735, "epoch": 0.7861480189983767, "flos": 26396178609600.0, "grad_norm": 1.7167570371727778, "language_loss": 0.78440702, "learning_rate": 4.607028720076822e-07, "loss": 0.80590218, "num_input_tokens_seen": 140927460, "step": 6538, "time_per_iteration": 2.5609629154205322 }, { "auxiliary_loss_clip": 0.01155156, "auxiliary_loss_mlp": 0.01022777, "balance_loss_clip": 1.04691362, "balance_loss_mlp": 1.01557064, "epoch": 0.7862682618890158, "flos": 24236581380000.0, "grad_norm": 1.9325905403109795, "language_loss": 0.739259, "learning_rate": 4.6020564048807074e-07, "loss": 0.76103836, "num_input_tokens_seen": 140945135, "step": 6539, "time_per_iteration": 2.512512445449829 }, { "auxiliary_loss_clip": 0.01156243, "auxiliary_loss_mlp": 0.01028911, "balance_loss_clip": 1.04724026, "balance_loss_mlp": 1.02147806, "epoch": 0.7863885047796549, "flos": 47551513689600.0, "grad_norm": 2.1032184328336876, "language_loss": 0.71863556, "learning_rate": 4.5970864255403883e-07, "loss": 0.7404871, "num_input_tokens_seen": 140966660, "step": 6540, "time_per_iteration": 2.7182884216308594 }, { "auxiliary_loss_clip": 0.01144734, "auxiliary_loss_mlp": 0.0102281, "balance_loss_clip": 1.04461145, "balance_loss_mlp": 1.01587248, "epoch": 0.786508747670294, "flos": 24389238823680.0, "grad_norm": 1.915474764097066, "language_loss": 0.82248527, "learning_rate": 4.59211878280982e-07, "loss": 0.84416068, "num_input_tokens_seen": 140986175, "step": 6541, "time_per_iteration": 2.4968552589416504 }, { "auxiliary_loss_clip": 0.01141846, "auxiliary_loss_mlp": 0.0102277, "balance_loss_clip": 1.04529238, "balance_loss_mlp": 1.01493764, "epoch": 0.786628990560933, "flos": 18041237538240.0, "grad_norm": 3.6335485037844912, "language_loss": 0.69887811, "learning_rate": 4.587153477442578e-07, "loss": 0.72052419, "num_input_tokens_seen": 141002490, "step": 6542, "time_per_iteration": 2.5125033855438232 }, { "auxiliary_loss_clip": 0.01174398, "auxiliary_loss_mlp": 0.01027294, "balance_loss_clip": 1.04961872, "balance_loss_mlp": 1.0193218, "epoch": 0.7867492334515722, "flos": 25848864047040.0, "grad_norm": 2.1840299196312185, "language_loss": 0.80994445, "learning_rate": 4.582190510191899e-07, "loss": 0.83196139, "num_input_tokens_seen": 141021150, "step": 6543, "time_per_iteration": 2.4687626361846924 }, { "auxiliary_loss_clip": 0.01124326, "auxiliary_loss_mlp": 0.01021079, "balance_loss_clip": 1.04528737, "balance_loss_mlp": 1.01397371, "epoch": 0.7868694763422113, "flos": 16580822141760.0, "grad_norm": 2.6755976222455304, "language_loss": 0.87115705, "learning_rate": 4.5772298818106625e-07, "loss": 0.89261115, "num_input_tokens_seen": 141036940, "step": 6544, "time_per_iteration": 2.5259969234466553 }, { "auxiliary_loss_clip": 0.01131816, "auxiliary_loss_mlp": 0.01031761, "balance_loss_clip": 1.04670119, "balance_loss_mlp": 1.0241195, "epoch": 0.7869897192328503, "flos": 29386280441760.0, "grad_norm": 2.3039145602428226, "language_loss": 0.71956646, "learning_rate": 4.572271593051384e-07, "loss": 0.74120224, "num_input_tokens_seen": 141054295, "step": 6545, "time_per_iteration": 2.5722382068634033 }, { "auxiliary_loss_clip": 0.01105344, "auxiliary_loss_mlp": 0.01027817, "balance_loss_clip": 1.04289567, "balance_loss_mlp": 1.0207541, "epoch": 0.7871099621234895, "flos": 17128926877440.0, "grad_norm": 1.6122312577080264, "language_loss": 0.78290522, "learning_rate": 4.567315644666245e-07, "loss": 0.80423689, "num_input_tokens_seen": 141073090, "step": 6546, "time_per_iteration": 2.5552783012390137 }, { "auxiliary_loss_clip": 0.01121378, "auxiliary_loss_mlp": 0.01026067, "balance_loss_clip": 1.0438509, "balance_loss_mlp": 1.0195787, "epoch": 0.7872302050141285, "flos": 23440191501120.0, "grad_norm": 2.1527171831712844, "language_loss": 0.84511554, "learning_rate": 4.5623620374070507e-07, "loss": 0.86659002, "num_input_tokens_seen": 141092405, "step": 6547, "time_per_iteration": 2.5709331035614014 }, { "auxiliary_loss_clip": 0.01028941, "auxiliary_loss_mlp": 0.01002817, "balance_loss_clip": 1.00455689, "balance_loss_mlp": 1.00144029, "epoch": 0.7873504479047676, "flos": 65959765064160.0, "grad_norm": 0.7662863100941862, "language_loss": 0.58470422, "learning_rate": 4.557410772025263e-07, "loss": 0.60502183, "num_input_tokens_seen": 141154355, "step": 6548, "time_per_iteration": 3.9976999759674072 }, { "auxiliary_loss_clip": 0.01136121, "auxiliary_loss_mlp": 0.01030126, "balance_loss_clip": 1.04405832, "balance_loss_mlp": 1.02256536, "epoch": 0.7874706907954068, "flos": 23258339957280.0, "grad_norm": 1.8529978385414736, "language_loss": 0.66287142, "learning_rate": 4.5524618492719803e-07, "loss": 0.68453389, "num_input_tokens_seen": 141173575, "step": 6549, "time_per_iteration": 2.5321052074432373 }, { "auxiliary_loss_clip": 0.0115372, "auxiliary_loss_mlp": 0.01026242, "balance_loss_clip": 1.04503536, "balance_loss_mlp": 1.01995039, "epoch": 0.7875909336860458, "flos": 28767791241120.0, "grad_norm": 1.4821095477764157, "language_loss": 0.78783149, "learning_rate": 4.54751526989795e-07, "loss": 0.80963105, "num_input_tokens_seen": 141195415, "step": 6550, "time_per_iteration": 2.5624566078186035 }, { "auxiliary_loss_clip": 0.01155791, "auxiliary_loss_mlp": 0.01024633, "balance_loss_clip": 1.04509079, "balance_loss_mlp": 1.01829982, "epoch": 0.7877111765766849, "flos": 18697289490720.0, "grad_norm": 2.007376518090703, "language_loss": 0.79085511, "learning_rate": 4.5425710346535775e-07, "loss": 0.81265938, "num_input_tokens_seen": 141213360, "step": 6551, "time_per_iteration": 2.480583906173706 }, { "auxiliary_loss_clip": 0.01158162, "auxiliary_loss_mlp": 0.01023162, "balance_loss_clip": 1.0479629, "balance_loss_mlp": 1.0160718, "epoch": 0.787831419467324, "flos": 27592972320960.0, "grad_norm": 2.014576749272105, "language_loss": 0.81468546, "learning_rate": 4.537629144288877e-07, "loss": 0.83649874, "num_input_tokens_seen": 141230815, "step": 6552, "time_per_iteration": 2.519695997238159 }, { "auxiliary_loss_clip": 0.01115571, "auxiliary_loss_mlp": 0.01031807, "balance_loss_clip": 1.04078662, "balance_loss_mlp": 1.02498567, "epoch": 0.7879516623579631, "flos": 18150190429920.0, "grad_norm": 4.591887101266382, "language_loss": 0.74740738, "learning_rate": 4.5326895995535477e-07, "loss": 0.7688812, "num_input_tokens_seen": 141249715, "step": 6553, "time_per_iteration": 2.5724096298217773 }, { "auxiliary_loss_clip": 0.0115206, "auxiliary_loss_mlp": 0.0102604, "balance_loss_clip": 1.04576099, "balance_loss_mlp": 1.01893544, "epoch": 0.7880719052486022, "flos": 20339197261440.0, "grad_norm": 2.3499687098797564, "language_loss": 0.84256142, "learning_rate": 4.527752401196907e-07, "loss": 0.86434245, "num_input_tokens_seen": 141267730, "step": 6554, "time_per_iteration": 2.485252857208252 }, { "auxiliary_loss_clip": 0.01137449, "auxiliary_loss_mlp": 0.01021232, "balance_loss_clip": 1.04449141, "balance_loss_mlp": 1.01453865, "epoch": 0.7881921481392413, "flos": 21653240682240.0, "grad_norm": 1.7891856188156987, "language_loss": 0.66348875, "learning_rate": 4.5228175499679254e-07, "loss": 0.68507558, "num_input_tokens_seen": 141287315, "step": 6555, "time_per_iteration": 2.5125088691711426 }, { "auxiliary_loss_clip": 0.01051721, "auxiliary_loss_mlp": 0.01000647, "balance_loss_clip": 1.0062058, "balance_loss_mlp": 0.99932355, "epoch": 0.7883123910298804, "flos": 68565870726240.0, "grad_norm": 0.8426158908664211, "language_loss": 0.54580688, "learning_rate": 4.5178850466152174e-07, "loss": 0.56633061, "num_input_tokens_seen": 141346145, "step": 6556, "time_per_iteration": 3.147134304046631 }, { "auxiliary_loss_clip": 0.01134738, "auxiliary_loss_mlp": 0.01023353, "balance_loss_clip": 1.04220235, "balance_loss_mlp": 1.01682663, "epoch": 0.7884326339205194, "flos": 19318221044640.0, "grad_norm": 1.7784009516346595, "language_loss": 0.81542277, "learning_rate": 4.512954891887031e-07, "loss": 0.83700365, "num_input_tokens_seen": 141364445, "step": 6557, "time_per_iteration": 2.5236480236053467 }, { "auxiliary_loss_clip": 0.01135324, "auxiliary_loss_mlp": 0.01026234, "balance_loss_clip": 1.04565859, "balance_loss_mlp": 1.01932585, "epoch": 0.7885528768111585, "flos": 17784906996000.0, "grad_norm": 2.0827101930933165, "language_loss": 0.83299661, "learning_rate": 4.5080270865312806e-07, "loss": 0.85461223, "num_input_tokens_seen": 141381640, "step": 6558, "time_per_iteration": 2.4775795936584473 }, { "auxiliary_loss_clip": 0.01154851, "auxiliary_loss_mlp": 0.01024901, "balance_loss_clip": 1.04663348, "balance_loss_mlp": 1.01804352, "epoch": 0.7886731197017977, "flos": 18807643143840.0, "grad_norm": 2.464229150107205, "language_loss": 0.71042228, "learning_rate": 4.5031016312954985e-07, "loss": 0.73221987, "num_input_tokens_seen": 141399955, "step": 6559, "time_per_iteration": 3.2331347465515137 }, { "auxiliary_loss_clip": 0.01162746, "auxiliary_loss_mlp": 0.01024211, "balance_loss_clip": 1.04875684, "balance_loss_mlp": 1.01717508, "epoch": 0.7887933625924367, "flos": 33365362867680.0, "grad_norm": 2.3395882648856894, "language_loss": 0.74140954, "learning_rate": 4.498178526926886e-07, "loss": 0.76327914, "num_input_tokens_seen": 141420820, "step": 6560, "time_per_iteration": 2.594324827194214 }, { "auxiliary_loss_clip": 0.01170279, "auxiliary_loss_mlp": 0.01029763, "balance_loss_clip": 1.04949391, "balance_loss_mlp": 1.02345657, "epoch": 0.7889136054830758, "flos": 17019363397440.0, "grad_norm": 2.430574624330802, "language_loss": 0.7233302, "learning_rate": 4.4932577741722635e-07, "loss": 0.74533057, "num_input_tokens_seen": 141439350, "step": 6561, "time_per_iteration": 3.2121787071228027 }, { "auxiliary_loss_clip": 0.01139764, "auxiliary_loss_mlp": 0.01028513, "balance_loss_clip": 1.04583991, "balance_loss_mlp": 1.02181315, "epoch": 0.7890338483737149, "flos": 29424633366720.0, "grad_norm": 1.6781054759741267, "language_loss": 0.74098825, "learning_rate": 4.4883393737780985e-07, "loss": 0.76267099, "num_input_tokens_seen": 141460300, "step": 6562, "time_per_iteration": 3.3478901386260986 }, { "auxiliary_loss_clip": 0.01147154, "auxiliary_loss_mlp": 0.01028114, "balance_loss_clip": 1.04364705, "balance_loss_mlp": 1.02107787, "epoch": 0.789154091264354, "flos": 19971579225120.0, "grad_norm": 2.1346342756418832, "language_loss": 0.7810601, "learning_rate": 4.4834233264905254e-07, "loss": 0.80281276, "num_input_tokens_seen": 141477315, "step": 6563, "time_per_iteration": 2.4785006046295166 }, { "auxiliary_loss_clip": 0.01118575, "auxiliary_loss_mlp": 0.01026205, "balance_loss_clip": 1.04057431, "balance_loss_mlp": 1.01919508, "epoch": 0.789274334154993, "flos": 14537828116320.0, "grad_norm": 2.3537093651604386, "language_loss": 0.71361804, "learning_rate": 4.478509633055294e-07, "loss": 0.73506588, "num_input_tokens_seen": 141495025, "step": 6564, "time_per_iteration": 2.5317025184631348 }, { "auxiliary_loss_clip": 0.01144474, "auxiliary_loss_mlp": 0.01031803, "balance_loss_clip": 1.04575515, "balance_loss_mlp": 1.02364945, "epoch": 0.7893945770456322, "flos": 21827405996640.0, "grad_norm": 2.9263022667125083, "language_loss": 0.79980731, "learning_rate": 4.473598294217813e-07, "loss": 0.8215701, "num_input_tokens_seen": 141510450, "step": 6565, "time_per_iteration": 2.502563714981079 }, { "auxiliary_loss_clip": 0.01154967, "auxiliary_loss_mlp": 0.01027032, "balance_loss_clip": 1.0481503, "balance_loss_mlp": 1.02060914, "epoch": 0.7895148199362713, "flos": 20740642685760.0, "grad_norm": 2.165168221263626, "language_loss": 0.71451861, "learning_rate": 4.468689310723124e-07, "loss": 0.73633856, "num_input_tokens_seen": 141528265, "step": 6566, "time_per_iteration": 2.4988913536071777 }, { "auxiliary_loss_clip": 0.01131646, "auxiliary_loss_mlp": 0.01030282, "balance_loss_clip": 1.04418325, "balance_loss_mlp": 1.0230906, "epoch": 0.7896350628269103, "flos": 16690672957440.0, "grad_norm": 1.7019349624269366, "language_loss": 0.78423071, "learning_rate": 4.463782683315913e-07, "loss": 0.80585003, "num_input_tokens_seen": 141547270, "step": 6567, "time_per_iteration": 2.561553478240967 }, { "auxiliary_loss_clip": 0.01165382, "auxiliary_loss_mlp": 0.01027927, "balance_loss_clip": 1.04664719, "balance_loss_mlp": 1.02117944, "epoch": 0.7897553057175495, "flos": 22638378161280.0, "grad_norm": 1.5889127283479714, "language_loss": 0.73369515, "learning_rate": 4.458878412740523e-07, "loss": 0.75562823, "num_input_tokens_seen": 141566050, "step": 6568, "time_per_iteration": 2.4967024326324463 }, { "auxiliary_loss_clip": 0.01153409, "auxiliary_loss_mlp": 0.01022261, "balance_loss_clip": 1.04836714, "balance_loss_mlp": 1.01512337, "epoch": 0.7898755486081885, "flos": 14537576697600.0, "grad_norm": 3.7979412898053533, "language_loss": 0.78275502, "learning_rate": 4.453976499740919e-07, "loss": 0.80451173, "num_input_tokens_seen": 141583695, "step": 6569, "time_per_iteration": 2.4312431812286377 }, { "auxiliary_loss_clip": 0.01153463, "auxiliary_loss_mlp": 0.01020175, "balance_loss_clip": 1.04872704, "balance_loss_mlp": 1.01378298, "epoch": 0.7899957914988276, "flos": 17238490357440.0, "grad_norm": 1.8209976617347599, "language_loss": 0.7775352, "learning_rate": 4.4490769450607215e-07, "loss": 0.79927158, "num_input_tokens_seen": 141601320, "step": 6570, "time_per_iteration": 2.4604997634887695 }, { "auxiliary_loss_clip": 0.01124084, "auxiliary_loss_mlp": 0.01029203, "balance_loss_clip": 1.04055977, "balance_loss_mlp": 1.0218122, "epoch": 0.7901160343894668, "flos": 41279356247040.0, "grad_norm": 1.7968621898913244, "language_loss": 0.72967297, "learning_rate": 4.4441797494431845e-07, "loss": 0.7512058, "num_input_tokens_seen": 141623125, "step": 6571, "time_per_iteration": 2.7199013233184814 }, { "auxiliary_loss_clip": 0.01152812, "auxiliary_loss_mlp": 0.01022631, "balance_loss_clip": 1.04691255, "balance_loss_mlp": 1.01526356, "epoch": 0.7902362772801058, "flos": 16837009016160.0, "grad_norm": 6.5451395260906295, "language_loss": 0.77618182, "learning_rate": 4.439284913631207e-07, "loss": 0.7979362, "num_input_tokens_seen": 141640335, "step": 6572, "time_per_iteration": 2.4678549766540527 }, { "auxiliary_loss_clip": 0.01128629, "auxiliary_loss_mlp": 0.01026179, "balance_loss_clip": 1.04696941, "balance_loss_mlp": 1.01934528, "epoch": 0.7903565201707449, "flos": 27125991221280.0, "grad_norm": 2.421630575263859, "language_loss": 0.83505356, "learning_rate": 4.434392438367347e-07, "loss": 0.85660166, "num_input_tokens_seen": 141659760, "step": 6573, "time_per_iteration": 2.5682544708251953 }, { "auxiliary_loss_clip": 0.01156067, "auxiliary_loss_mlp": 0.01023063, "balance_loss_clip": 1.04698944, "balance_loss_mlp": 1.0155592, "epoch": 0.790476763061384, "flos": 31025171187840.0, "grad_norm": 1.931511171508619, "language_loss": 0.73824662, "learning_rate": 4.4295023243937677e-07, "loss": 0.76003796, "num_input_tokens_seen": 141679965, "step": 6574, "time_per_iteration": 3.318059206008911 }, { "auxiliary_loss_clip": 0.01159943, "auxiliary_loss_mlp": 0.0102875, "balance_loss_clip": 1.05042338, "balance_loss_mlp": 1.02070975, "epoch": 0.7905970059520231, "flos": 22089088165920.0, "grad_norm": 1.8103809230502441, "language_loss": 0.80366111, "learning_rate": 4.4246145724523123e-07, "loss": 0.82554805, "num_input_tokens_seen": 141697710, "step": 6575, "time_per_iteration": 2.500302314758301 }, { "auxiliary_loss_clip": 0.01128582, "auxiliary_loss_mlp": 0.01020751, "balance_loss_clip": 1.04597151, "balance_loss_mlp": 1.01434636, "epoch": 0.7907172488426621, "flos": 20558144636640.0, "grad_norm": 2.1381895008121194, "language_loss": 0.77427202, "learning_rate": 4.41972918328444e-07, "loss": 0.7957654, "num_input_tokens_seen": 141715145, "step": 6576, "time_per_iteration": 2.5450448989868164 }, { "auxiliary_loss_clip": 0.01152244, "auxiliary_loss_mlp": 0.01025599, "balance_loss_clip": 1.04721117, "balance_loss_mlp": 1.01850057, "epoch": 0.7908374917333013, "flos": 30081547326240.0, "grad_norm": 2.124638109449116, "language_loss": 0.7735225, "learning_rate": 4.4148461576312646e-07, "loss": 0.79530096, "num_input_tokens_seen": 141734810, "step": 6577, "time_per_iteration": 2.537379026412964 }, { "auxiliary_loss_clip": 0.01155758, "auxiliary_loss_mlp": 0.01025886, "balance_loss_clip": 1.04840326, "balance_loss_mlp": 1.01963091, "epoch": 0.7909577346239404, "flos": 20996362639680.0, "grad_norm": 1.690816142716623, "language_loss": 0.74442315, "learning_rate": 4.4099654962335343e-07, "loss": 0.76623964, "num_input_tokens_seen": 141755260, "step": 6578, "time_per_iteration": 2.5181591510772705 }, { "auxiliary_loss_clip": 0.01146272, "auxiliary_loss_mlp": 0.01022722, "balance_loss_clip": 1.04598999, "balance_loss_mlp": 1.01522648, "epoch": 0.7910779775145794, "flos": 26247938952000.0, "grad_norm": 1.7537953858501019, "language_loss": 0.74970692, "learning_rate": 4.405087199831636e-07, "loss": 0.77139688, "num_input_tokens_seen": 141775500, "step": 6579, "time_per_iteration": 2.5551722049713135 }, { "auxiliary_loss_clip": 0.01141124, "auxiliary_loss_mlp": 0.00762464, "balance_loss_clip": 1.04303694, "balance_loss_mlp": 1.00063634, "epoch": 0.7911982204052186, "flos": 22564437917280.0, "grad_norm": 2.0098101206513705, "language_loss": 0.67266345, "learning_rate": 4.400211269165619e-07, "loss": 0.69169939, "num_input_tokens_seen": 141791955, "step": 6580, "time_per_iteration": 2.531879425048828 }, { "auxiliary_loss_clip": 0.01172, "auxiliary_loss_mlp": 0.01024784, "balance_loss_clip": 1.0518117, "balance_loss_mlp": 1.0182395, "epoch": 0.7913184632958576, "flos": 23112542652960.0, "grad_norm": 1.5365840463743574, "language_loss": 0.76920128, "learning_rate": 4.3953377049751416e-07, "loss": 0.79116911, "num_input_tokens_seen": 141812380, "step": 6581, "time_per_iteration": 2.459907293319702 }, { "auxiliary_loss_clip": 0.01147714, "auxiliary_loss_mlp": 0.01026336, "balance_loss_clip": 1.04765272, "balance_loss_mlp": 1.01936817, "epoch": 0.7914387061864967, "flos": 12311761370400.0, "grad_norm": 2.802533471173947, "language_loss": 0.77653122, "learning_rate": 4.390466507999537e-07, "loss": 0.79827178, "num_input_tokens_seen": 141828130, "step": 6582, "time_per_iteration": 2.493041515350342 }, { "auxiliary_loss_clip": 0.01127, "auxiliary_loss_mlp": 0.01027712, "balance_loss_clip": 1.04504013, "balance_loss_mlp": 1.02064848, "epoch": 0.7915589490771359, "flos": 17603270953920.0, "grad_norm": 2.53402865248158, "language_loss": 0.75478578, "learning_rate": 4.385597678977748e-07, "loss": 0.77633286, "num_input_tokens_seen": 141846965, "step": 6583, "time_per_iteration": 2.5323567390441895 }, { "auxiliary_loss_clip": 0.01141267, "auxiliary_loss_mlp": 0.01025091, "balance_loss_clip": 1.04466033, "balance_loss_mlp": 1.01757455, "epoch": 0.7916791919677749, "flos": 25591276411200.0, "grad_norm": 2.1534136288368804, "language_loss": 0.75572753, "learning_rate": 4.3807312186483726e-07, "loss": 0.77739114, "num_input_tokens_seen": 141867685, "step": 6584, "time_per_iteration": 2.571521759033203 }, { "auxiliary_loss_clip": 0.01153452, "auxiliary_loss_mlp": 0.01026695, "balance_loss_clip": 1.05033422, "balance_loss_mlp": 1.01982236, "epoch": 0.791799434858414, "flos": 18844343888640.0, "grad_norm": 1.8976957262520993, "language_loss": 0.78056335, "learning_rate": 4.375867127749655e-07, "loss": 0.80236483, "num_input_tokens_seen": 141885960, "step": 6585, "time_per_iteration": 3.266669511795044 }, { "auxiliary_loss_clip": 0.01127567, "auxiliary_loss_mlp": 0.01026466, "balance_loss_clip": 1.04471767, "balance_loss_mlp": 1.01983845, "epoch": 0.7919196777490531, "flos": 25812019634400.0, "grad_norm": 1.7851285434856392, "language_loss": 0.67349315, "learning_rate": 4.3710054070194744e-07, "loss": 0.69503343, "num_input_tokens_seen": 141905655, "step": 6586, "time_per_iteration": 2.5741987228393555 }, { "auxiliary_loss_clip": 0.01168381, "auxiliary_loss_mlp": 0.00762729, "balance_loss_clip": 1.04621792, "balance_loss_mlp": 1.00065088, "epoch": 0.7920399206396922, "flos": 11947627279200.0, "grad_norm": 2.647652081188097, "language_loss": 0.66948795, "learning_rate": 4.3661460571953455e-07, "loss": 0.68879914, "num_input_tokens_seen": 141922390, "step": 6587, "time_per_iteration": 3.2199325561523438 }, { "auxiliary_loss_clip": 0.01151631, "auxiliary_loss_mlp": 0.01025998, "balance_loss_clip": 1.04240906, "balance_loss_mlp": 1.01931977, "epoch": 0.7921601635303313, "flos": 21579911026560.0, "grad_norm": 1.5342542046836791, "language_loss": 0.68368351, "learning_rate": 4.36128907901443e-07, "loss": 0.70545971, "num_input_tokens_seen": 141941985, "step": 6588, "time_per_iteration": 3.188037633895874 }, { "auxiliary_loss_clip": 0.01128862, "auxiliary_loss_mlp": 0.01024328, "balance_loss_clip": 1.04248381, "balance_loss_mlp": 1.01779282, "epoch": 0.7922804064209703, "flos": 18113992522560.0, "grad_norm": 2.31458891915623, "language_loss": 0.72903669, "learning_rate": 4.356434473213519e-07, "loss": 0.75056863, "num_input_tokens_seen": 141959435, "step": 6589, "time_per_iteration": 2.5282230377197266 }, { "auxiliary_loss_clip": 0.01141513, "auxiliary_loss_mlp": 0.01026629, "balance_loss_clip": 1.04848754, "balance_loss_mlp": 1.01941407, "epoch": 0.7924006493116095, "flos": 21652809678720.0, "grad_norm": 1.557468196015323, "language_loss": 0.79718274, "learning_rate": 4.351582240529068e-07, "loss": 0.81886423, "num_input_tokens_seen": 141980265, "step": 6590, "time_per_iteration": 2.517542600631714 }, { "auxiliary_loss_clip": 0.01044326, "auxiliary_loss_mlp": 0.01001335, "balance_loss_clip": 1.00656772, "balance_loss_mlp": 1.00004184, "epoch": 0.7925208922022485, "flos": 64242767706720.0, "grad_norm": 0.743515577345931, "language_loss": 0.58227813, "learning_rate": 4.346732381697149e-07, "loss": 0.60273468, "num_input_tokens_seen": 142044395, "step": 6591, "time_per_iteration": 3.1860973834991455 }, { "auxiliary_loss_clip": 0.01134622, "auxiliary_loss_mlp": 0.01025129, "balance_loss_clip": 1.04529047, "balance_loss_mlp": 1.01820016, "epoch": 0.7926411350928876, "flos": 16941544121760.0, "grad_norm": 1.9096629196718848, "language_loss": 0.81128931, "learning_rate": 4.3418848974534825e-07, "loss": 0.83288682, "num_input_tokens_seen": 142061335, "step": 6592, "time_per_iteration": 2.4990131855010986 }, { "auxiliary_loss_clip": 0.01130333, "auxiliary_loss_mlp": 0.01027703, "balance_loss_clip": 1.04426038, "balance_loss_mlp": 1.02128363, "epoch": 0.7927613779835267, "flos": 34459991992800.0, "grad_norm": 1.5762795354693175, "language_loss": 0.689466, "learning_rate": 4.3370397885334276e-07, "loss": 0.71104634, "num_input_tokens_seen": 142081965, "step": 6593, "time_per_iteration": 2.660487651824951 }, { "auxiliary_loss_clip": 0.01147868, "auxiliary_loss_mlp": 0.01025118, "balance_loss_clip": 1.04577971, "balance_loss_mlp": 1.01819527, "epoch": 0.7928816208741658, "flos": 18951177679680.0, "grad_norm": 1.7916289241347734, "language_loss": 0.7545203, "learning_rate": 4.3321970556719777e-07, "loss": 0.77625024, "num_input_tokens_seen": 142100260, "step": 6594, "time_per_iteration": 2.4859848022460938 }, { "auxiliary_loss_clip": 0.0117174, "auxiliary_loss_mlp": 0.0102763, "balance_loss_clip": 1.04921675, "balance_loss_mlp": 1.02068925, "epoch": 0.7930018637648049, "flos": 18623025994080.0, "grad_norm": 2.516812478613065, "language_loss": 0.71939564, "learning_rate": 4.3273566996037856e-07, "loss": 0.74138933, "num_input_tokens_seen": 142116955, "step": 6595, "time_per_iteration": 2.432241439819336 }, { "auxiliary_loss_clip": 0.01140112, "auxiliary_loss_mlp": 0.01023464, "balance_loss_clip": 1.04527235, "balance_loss_mlp": 1.01716971, "epoch": 0.793122106655444, "flos": 24530654258880.0, "grad_norm": 1.91516200468355, "language_loss": 0.80545771, "learning_rate": 4.322518721063113e-07, "loss": 0.82709348, "num_input_tokens_seen": 142135505, "step": 6596, "time_per_iteration": 2.5509133338928223 }, { "auxiliary_loss_clip": 0.01155906, "auxiliary_loss_mlp": 0.01020014, "balance_loss_clip": 1.04834044, "balance_loss_mlp": 1.012954, "epoch": 0.7932423495460831, "flos": 34421208064320.0, "grad_norm": 2.999726834443032, "language_loss": 0.70636797, "learning_rate": 4.3176831207838906e-07, "loss": 0.72812712, "num_input_tokens_seen": 142158915, "step": 6597, "time_per_iteration": 2.6053497791290283 }, { "auxiliary_loss_clip": 0.01156671, "auxiliary_loss_mlp": 0.01025072, "balance_loss_clip": 1.05149877, "balance_loss_mlp": 1.01760316, "epoch": 0.7933625924367221, "flos": 26980337584800.0, "grad_norm": 2.299652737526059, "language_loss": 0.74274945, "learning_rate": 4.3128498994996685e-07, "loss": 0.7645669, "num_input_tokens_seen": 142178390, "step": 6598, "time_per_iteration": 2.529139280319214 }, { "auxiliary_loss_clip": 0.0115909, "auxiliary_loss_mlp": 0.01027922, "balance_loss_clip": 1.04739177, "balance_loss_mlp": 1.02066839, "epoch": 0.7934828353273613, "flos": 29568634823040.0, "grad_norm": 1.9642301978615246, "language_loss": 0.71227258, "learning_rate": 4.308019057943646e-07, "loss": 0.73414272, "num_input_tokens_seen": 142200115, "step": 6599, "time_per_iteration": 2.559767961502075 }, { "auxiliary_loss_clip": 0.01117819, "auxiliary_loss_mlp": 0.01023999, "balance_loss_clip": 1.04385304, "balance_loss_mlp": 1.01689458, "epoch": 0.7936030782180004, "flos": 28615385216160.0, "grad_norm": 1.583190656920848, "language_loss": 0.74484926, "learning_rate": 4.3031905968486535e-07, "loss": 0.76626742, "num_input_tokens_seen": 142220945, "step": 6600, "time_per_iteration": 3.3500008583068848 }, { "auxiliary_loss_clip": 0.0111132, "auxiliary_loss_mlp": 0.01023522, "balance_loss_clip": 1.04515648, "balance_loss_mlp": 1.01619065, "epoch": 0.7937233211086394, "flos": 16392577379040.0, "grad_norm": 2.075709073686296, "language_loss": 0.6894713, "learning_rate": 4.298364516947162e-07, "loss": 0.71081972, "num_input_tokens_seen": 142238175, "step": 6601, "time_per_iteration": 2.535573720932007 }, { "auxiliary_loss_clip": 0.01109342, "auxiliary_loss_mlp": 0.01024134, "balance_loss_clip": 1.04205549, "balance_loss_mlp": 1.01696086, "epoch": 0.7938435639992786, "flos": 22013423907840.0, "grad_norm": 2.52242349276309, "language_loss": 0.65539515, "learning_rate": 4.293540818971295e-07, "loss": 0.67672986, "num_input_tokens_seen": 142255980, "step": 6602, "time_per_iteration": 2.598132610321045 }, { "auxiliary_loss_clip": 0.01162209, "auxiliary_loss_mlp": 0.01025322, "balance_loss_clip": 1.04820371, "balance_loss_mlp": 1.01857758, "epoch": 0.7939638068899176, "flos": 22197035382720.0, "grad_norm": 1.897210752474476, "language_loss": 0.76426864, "learning_rate": 4.2887195036527934e-07, "loss": 0.78614396, "num_input_tokens_seen": 142274785, "step": 6603, "time_per_iteration": 2.482746124267578 }, { "auxiliary_loss_clip": 0.01146254, "auxiliary_loss_mlp": 0.01019977, "balance_loss_clip": 1.04244959, "balance_loss_mlp": 1.01252079, "epoch": 0.7940840497805567, "flos": 17745189226560.0, "grad_norm": 2.3948172935955427, "language_loss": 0.7311902, "learning_rate": 4.28390057172306e-07, "loss": 0.7528525, "num_input_tokens_seen": 142291290, "step": 6604, "time_per_iteration": 2.4597463607788086 }, { "auxiliary_loss_clip": 0.01119519, "auxiliary_loss_mlp": 0.01029164, "balance_loss_clip": 1.03964174, "balance_loss_mlp": 1.02170718, "epoch": 0.7942042926711959, "flos": 23805439018080.0, "grad_norm": 2.147514698343996, "language_loss": 0.71894282, "learning_rate": 4.279084023913111e-07, "loss": 0.74042964, "num_input_tokens_seen": 142309165, "step": 6605, "time_per_iteration": 2.5748162269592285 }, { "auxiliary_loss_clip": 0.01153727, "auxiliary_loss_mlp": 0.01021487, "balance_loss_clip": 1.04743385, "balance_loss_mlp": 1.01495421, "epoch": 0.7943245355618349, "flos": 19244963222880.0, "grad_norm": 2.375781782265412, "language_loss": 0.6929121, "learning_rate": 4.2742698609536096e-07, "loss": 0.71466428, "num_input_tokens_seen": 142327475, "step": 6606, "time_per_iteration": 2.497730255126953 }, { "auxiliary_loss_clip": 0.01144454, "auxiliary_loss_mlp": 0.01022777, "balance_loss_clip": 1.04608655, "balance_loss_mlp": 1.01566029, "epoch": 0.794444778452474, "flos": 25007620273440.0, "grad_norm": 2.1648833587572733, "language_loss": 0.78757781, "learning_rate": 4.2694580835748706e-07, "loss": 0.80925012, "num_input_tokens_seen": 142347335, "step": 6607, "time_per_iteration": 2.5393712520599365 }, { "auxiliary_loss_clip": 0.01135991, "auxiliary_loss_mlp": 0.01024328, "balance_loss_clip": 1.04502845, "balance_loss_mlp": 1.0177269, "epoch": 0.7945650213431131, "flos": 23221495544640.0, "grad_norm": 1.8869300194127703, "language_loss": 0.74523735, "learning_rate": 4.264648692506836e-07, "loss": 0.76684052, "num_input_tokens_seen": 142366125, "step": 6608, "time_per_iteration": 2.518904685974121 }, { "auxiliary_loss_clip": 0.01135739, "auxiliary_loss_mlp": 0.01026951, "balance_loss_clip": 1.04476452, "balance_loss_mlp": 1.01929164, "epoch": 0.7946852642337522, "flos": 26062890798720.0, "grad_norm": 3.340470264522786, "language_loss": 0.72168493, "learning_rate": 4.2598416884790824e-07, "loss": 0.74331176, "num_input_tokens_seen": 142385175, "step": 6609, "time_per_iteration": 2.5299298763275146 }, { "auxiliary_loss_clip": 0.01150493, "auxiliary_loss_mlp": 0.01029785, "balance_loss_clip": 1.04436123, "balance_loss_mlp": 1.02198243, "epoch": 0.7948055071243912, "flos": 23769708031200.0, "grad_norm": 2.3050061396569683, "language_loss": 0.81080908, "learning_rate": 4.255037072220828e-07, "loss": 0.83261192, "num_input_tokens_seen": 142406545, "step": 6610, "time_per_iteration": 2.534012794494629 }, { "auxiliary_loss_clip": 0.01166155, "auxiliary_loss_mlp": 0.01021635, "balance_loss_clip": 1.0472275, "balance_loss_mlp": 1.01464939, "epoch": 0.7949257500150304, "flos": 21980817696480.0, "grad_norm": 1.7812671537636824, "language_loss": 0.71767604, "learning_rate": 4.2502348444609293e-07, "loss": 0.73955393, "num_input_tokens_seen": 142426165, "step": 6611, "time_per_iteration": 3.218834638595581 }, { "auxiliary_loss_clip": 0.01107289, "auxiliary_loss_mlp": 0.01022146, "balance_loss_clip": 1.03807664, "balance_loss_mlp": 1.01551485, "epoch": 0.7950459929056695, "flos": 25774133629920.0, "grad_norm": 2.705054182933261, "language_loss": 0.69322032, "learning_rate": 4.2454350059278844e-07, "loss": 0.71451461, "num_input_tokens_seen": 142447225, "step": 6612, "time_per_iteration": 2.6152865886688232 }, { "auxiliary_loss_clip": 0.0113086, "auxiliary_loss_mlp": 0.01023755, "balance_loss_clip": 1.03794837, "balance_loss_mlp": 1.01687336, "epoch": 0.7951662357963085, "flos": 22158071869440.0, "grad_norm": 1.7508328503059674, "language_loss": 0.84114611, "learning_rate": 4.240637557349824e-07, "loss": 0.8626923, "num_input_tokens_seen": 142464440, "step": 6613, "time_per_iteration": 3.313257932662964 }, { "auxiliary_loss_clip": 0.01129066, "auxiliary_loss_mlp": 0.01028608, "balance_loss_clip": 1.04426825, "balance_loss_mlp": 1.02142596, "epoch": 0.7952864786869477, "flos": 24641941752960.0, "grad_norm": 2.304410666874899, "language_loss": 0.66665518, "learning_rate": 4.235842499454516e-07, "loss": 0.68823195, "num_input_tokens_seen": 142484355, "step": 6614, "time_per_iteration": 3.230853796005249 }, { "auxiliary_loss_clip": 0.01141733, "auxiliary_loss_mlp": 0.01030184, "balance_loss_clip": 1.04523015, "balance_loss_mlp": 1.02323151, "epoch": 0.7954067215775867, "flos": 21830925858720.0, "grad_norm": 1.6836056034093212, "language_loss": 0.83001417, "learning_rate": 4.2310498329693687e-07, "loss": 0.85173333, "num_input_tokens_seen": 142505255, "step": 6615, "time_per_iteration": 2.5704238414764404 }, { "auxiliary_loss_clip": 0.0115641, "auxiliary_loss_mlp": 0.01027983, "balance_loss_clip": 1.04624796, "balance_loss_mlp": 1.02084565, "epoch": 0.7955269644682258, "flos": 24060656134560.0, "grad_norm": 1.4695819766139209, "language_loss": 0.80820513, "learning_rate": 4.2262595586214164e-07, "loss": 0.83004904, "num_input_tokens_seen": 142526350, "step": 6616, "time_per_iteration": 2.5319087505340576 }, { "auxiliary_loss_clip": 0.01159585, "auxiliary_loss_mlp": 0.01023741, "balance_loss_clip": 1.04790306, "balance_loss_mlp": 1.01680326, "epoch": 0.795647207358865, "flos": 25010745048960.0, "grad_norm": 1.657528524396304, "language_loss": 0.77086699, "learning_rate": 4.221471677137358e-07, "loss": 0.79270017, "num_input_tokens_seen": 142547165, "step": 6617, "time_per_iteration": 2.5735220909118652 }, { "auxiliary_loss_clip": 0.01130171, "auxiliary_loss_mlp": 0.01027941, "balance_loss_clip": 1.04516077, "balance_loss_mlp": 1.02106559, "epoch": 0.795767450249504, "flos": 14648361354240.0, "grad_norm": 1.618527956050415, "language_loss": 0.70180637, "learning_rate": 4.216686189243492e-07, "loss": 0.72338748, "num_input_tokens_seen": 142565955, "step": 6618, "time_per_iteration": 2.575103521347046 }, { "auxiliary_loss_clip": 0.01121885, "auxiliary_loss_mlp": 0.0102294, "balance_loss_clip": 1.04289758, "balance_loss_mlp": 1.01584697, "epoch": 0.7958876931401431, "flos": 18547900490400.0, "grad_norm": 1.699235303745789, "language_loss": 0.72971648, "learning_rate": 4.211903095665785e-07, "loss": 0.75116467, "num_input_tokens_seen": 142585340, "step": 6619, "time_per_iteration": 2.5674221515655518 }, { "auxiliary_loss_clip": 0.01149034, "auxiliary_loss_mlp": 0.01026344, "balance_loss_clip": 1.04548883, "balance_loss_mlp": 1.0199182, "epoch": 0.7960079360307821, "flos": 21543964537920.0, "grad_norm": 1.8192199981583304, "language_loss": 0.75224054, "learning_rate": 4.2071223971298277e-07, "loss": 0.77399427, "num_input_tokens_seen": 142602525, "step": 6620, "time_per_iteration": 2.4919369220733643 }, { "auxiliary_loss_clip": 0.01155919, "auxiliary_loss_mlp": 0.01024502, "balance_loss_clip": 1.04603577, "balance_loss_mlp": 1.01691425, "epoch": 0.7961281789214213, "flos": 25481748848160.0, "grad_norm": 1.8367885979642917, "language_loss": 0.60996616, "learning_rate": 4.2023440943608433e-07, "loss": 0.63177037, "num_input_tokens_seen": 142622490, "step": 6621, "time_per_iteration": 2.5378592014312744 }, { "auxiliary_loss_clip": 0.01155346, "auxiliary_loss_mlp": 0.0102032, "balance_loss_clip": 1.04531693, "balance_loss_mlp": 1.01393926, "epoch": 0.7962484218120603, "flos": 21944440204320.0, "grad_norm": 1.6379231071796834, "language_loss": 0.77946532, "learning_rate": 4.1975681880837023e-07, "loss": 0.80122197, "num_input_tokens_seen": 142642495, "step": 6622, "time_per_iteration": 2.479890823364258 }, { "auxiliary_loss_clip": 0.01121843, "auxiliary_loss_mlp": 0.01031483, "balance_loss_clip": 1.04125679, "balance_loss_mlp": 1.02469397, "epoch": 0.7963686647026994, "flos": 18876267677760.0, "grad_norm": 2.0222899391087275, "language_loss": 0.823681, "learning_rate": 4.192794679022895e-07, "loss": 0.84521425, "num_input_tokens_seen": 142660820, "step": 6623, "time_per_iteration": 2.544642210006714 }, { "auxiliary_loss_clip": 0.01156369, "auxiliary_loss_mlp": 0.01025885, "balance_loss_clip": 1.04671597, "balance_loss_mlp": 1.01866055, "epoch": 0.7964889075933386, "flos": 29716587144960.0, "grad_norm": 1.7549085334843268, "language_loss": 0.71902889, "learning_rate": 4.1880235679025743e-07, "loss": 0.7408514, "num_input_tokens_seen": 142680915, "step": 6624, "time_per_iteration": 2.5489542484283447 }, { "auxiliary_loss_clip": 0.01100018, "auxiliary_loss_mlp": 0.01027109, "balance_loss_clip": 1.03916633, "balance_loss_mlp": 1.01949728, "epoch": 0.7966091504839776, "flos": 29491462052640.0, "grad_norm": 1.762630400253436, "language_loss": 0.63734078, "learning_rate": 4.1832548554464986e-07, "loss": 0.65861207, "num_input_tokens_seen": 142699210, "step": 6625, "time_per_iteration": 3.4319965839385986 }, { "auxiliary_loss_clip": 0.01046246, "auxiliary_loss_mlp": 0.01000745, "balance_loss_clip": 1.00506318, "balance_loss_mlp": 0.9994877, "epoch": 0.7967293933746167, "flos": 67288707635040.0, "grad_norm": 0.7410045120469166, "language_loss": 0.58769554, "learning_rate": 4.178488542378098e-07, "loss": 0.60816544, "num_input_tokens_seen": 142756790, "step": 6626, "time_per_iteration": 3.0116028785705566 }, { "auxiliary_loss_clip": 0.01172135, "auxiliary_loss_mlp": 0.01028596, "balance_loss_clip": 1.04893935, "balance_loss_mlp": 1.0208776, "epoch": 0.7968496362652558, "flos": 25554683417280.0, "grad_norm": 1.6462613731796067, "language_loss": 0.88927162, "learning_rate": 4.173724629420401e-07, "loss": 0.9112789, "num_input_tokens_seen": 142778150, "step": 6627, "time_per_iteration": 2.4805378913879395 }, { "auxiliary_loss_clip": 0.01144822, "auxiliary_loss_mlp": 0.01021286, "balance_loss_clip": 1.04599667, "balance_loss_mlp": 1.01391578, "epoch": 0.7969698791558949, "flos": 14501091454560.0, "grad_norm": 3.0748149321614706, "language_loss": 0.6800555, "learning_rate": 4.168963117296087e-07, "loss": 0.70171654, "num_input_tokens_seen": 142795485, "step": 6628, "time_per_iteration": 2.481127977371216 }, { "auxiliary_loss_clip": 0.01170591, "auxiliary_loss_mlp": 0.01022247, "balance_loss_clip": 1.04949486, "balance_loss_mlp": 1.01511836, "epoch": 0.797090122046534, "flos": 22127548841760.0, "grad_norm": 2.188458867017634, "language_loss": 0.76151693, "learning_rate": 4.1642040067274876e-07, "loss": 0.7834453, "num_input_tokens_seen": 142815155, "step": 6629, "time_per_iteration": 2.4521372318267822 }, { "auxiliary_loss_clip": 0.01144214, "auxiliary_loss_mlp": 0.01024671, "balance_loss_clip": 1.0451405, "balance_loss_mlp": 1.01812077, "epoch": 0.7972103649371731, "flos": 19897674898080.0, "grad_norm": 1.8539362806946844, "language_loss": 0.72490072, "learning_rate": 4.1594472984365493e-07, "loss": 0.74658954, "num_input_tokens_seen": 142833840, "step": 6630, "time_per_iteration": 2.5085713863372803 }, { "auxiliary_loss_clip": 0.0115139, "auxiliary_loss_mlp": 0.01025161, "balance_loss_clip": 1.04698718, "balance_loss_mlp": 1.01818419, "epoch": 0.7973306078278122, "flos": 36058626215040.0, "grad_norm": 2.0890306744995613, "language_loss": 0.77537048, "learning_rate": 4.154692993144862e-07, "loss": 0.79713595, "num_input_tokens_seen": 142853610, "step": 6631, "time_per_iteration": 2.587460517883301 }, { "auxiliary_loss_clip": 0.01166558, "auxiliary_loss_mlp": 0.00762183, "balance_loss_clip": 1.04618299, "balance_loss_mlp": 1.00057685, "epoch": 0.7974508507184512, "flos": 21360604481760.0, "grad_norm": 2.08833107127869, "language_loss": 0.71454096, "learning_rate": 4.1499410915736476e-07, "loss": 0.73382831, "num_input_tokens_seen": 142872540, "step": 6632, "time_per_iteration": 2.458312749862671 }, { "auxiliary_loss_clip": 0.01052898, "auxiliary_loss_mlp": 0.01000865, "balance_loss_clip": 1.00665545, "balance_loss_mlp": 0.99967289, "epoch": 0.7975710936090904, "flos": 68253127416480.0, "grad_norm": 0.7696544283192696, "language_loss": 0.64280677, "learning_rate": 4.145191594443762e-07, "loss": 0.66334438, "num_input_tokens_seen": 142936895, "step": 6633, "time_per_iteration": 3.228360176086426 }, { "auxiliary_loss_clip": 0.01121673, "auxiliary_loss_mlp": 0.01028024, "balance_loss_clip": 1.04335058, "balance_loss_mlp": 1.02029288, "epoch": 0.7976913364997295, "flos": 22492437189120.0, "grad_norm": 2.4356925078845095, "language_loss": 0.70492256, "learning_rate": 4.140444502475713e-07, "loss": 0.72641957, "num_input_tokens_seen": 142956445, "step": 6634, "time_per_iteration": 2.5696582794189453 }, { "auxiliary_loss_clip": 0.01150918, "auxiliary_loss_mlp": 0.01024397, "balance_loss_clip": 1.04390025, "balance_loss_mlp": 1.01744735, "epoch": 0.7978115793903685, "flos": 15263222941920.0, "grad_norm": 1.8014692313203824, "language_loss": 0.69961673, "learning_rate": 4.1356998163896216e-07, "loss": 0.72136986, "num_input_tokens_seen": 142973495, "step": 6635, "time_per_iteration": 2.442455291748047 }, { "auxiliary_loss_clip": 0.01132048, "auxiliary_loss_mlp": 0.01024888, "balance_loss_clip": 1.04534543, "balance_loss_mlp": 1.01787877, "epoch": 0.7979318222810077, "flos": 19719235465440.0, "grad_norm": 2.024472281076241, "language_loss": 0.74996769, "learning_rate": 4.130957536905255e-07, "loss": 0.77153707, "num_input_tokens_seen": 142991510, "step": 6636, "time_per_iteration": 2.547762632369995 }, { "auxiliary_loss_clip": 0.01145962, "auxiliary_loss_mlp": 0.01028764, "balance_loss_clip": 1.04582596, "balance_loss_mlp": 1.02157617, "epoch": 0.7980520651716467, "flos": 15560276928480.0, "grad_norm": 2.553826342770606, "language_loss": 0.71254796, "learning_rate": 4.1262176647420134e-07, "loss": 0.73429525, "num_input_tokens_seen": 143009675, "step": 6637, "time_per_iteration": 3.244904041290283 }, { "auxiliary_loss_clip": 0.01145556, "auxiliary_loss_mlp": 0.01028999, "balance_loss_clip": 1.04473341, "balance_loss_mlp": 1.02237082, "epoch": 0.7981723080622858, "flos": 22309436302560.0, "grad_norm": 1.860673950904996, "language_loss": 0.79526854, "learning_rate": 4.121480200618923e-07, "loss": 0.81701404, "num_input_tokens_seen": 143029330, "step": 6638, "time_per_iteration": 2.5123231410980225 }, { "auxiliary_loss_clip": 0.01136509, "auxiliary_loss_mlp": 0.01029478, "balance_loss_clip": 1.04501498, "balance_loss_mlp": 1.02224779, "epoch": 0.798292550952925, "flos": 22929577683360.0, "grad_norm": 2.0640129100198203, "language_loss": 0.80129218, "learning_rate": 4.116745145254674e-07, "loss": 0.82295203, "num_input_tokens_seen": 143048865, "step": 6639, "time_per_iteration": 3.3613672256469727 }, { "auxiliary_loss_clip": 0.01038138, "auxiliary_loss_mlp": 0.01000388, "balance_loss_clip": 1.00648117, "balance_loss_mlp": 0.99909419, "epoch": 0.798412793843564, "flos": 64497948906240.0, "grad_norm": 0.7687507224819606, "language_loss": 0.58031148, "learning_rate": 4.1120124993675476e-07, "loss": 0.60069674, "num_input_tokens_seen": 143113295, "step": 6640, "time_per_iteration": 3.1372272968292236 }, { "auxiliary_loss_clip": 0.01147853, "auxiliary_loss_mlp": 0.01026704, "balance_loss_clip": 1.04470968, "balance_loss_mlp": 1.01979303, "epoch": 0.7985330367342031, "flos": 13586913111840.0, "grad_norm": 2.1815354554789597, "language_loss": 0.61957383, "learning_rate": 4.107282263675498e-07, "loss": 0.64131945, "num_input_tokens_seen": 143130965, "step": 6641, "time_per_iteration": 3.254380226135254 }, { "auxiliary_loss_clip": 0.01041444, "auxiliary_loss_mlp": 0.00752633, "balance_loss_clip": 1.00995159, "balance_loss_mlp": 1.00029683, "epoch": 0.7986532796248422, "flos": 67698809046720.0, "grad_norm": 0.7659500580628628, "language_loss": 0.52480936, "learning_rate": 4.1025544388960907e-07, "loss": 0.54275012, "num_input_tokens_seen": 143192005, "step": 6642, "time_per_iteration": 3.1311392784118652 }, { "auxiliary_loss_clip": 0.01154446, "auxiliary_loss_mlp": 0.01027914, "balance_loss_clip": 1.04726386, "balance_loss_mlp": 1.02087808, "epoch": 0.7987735225154813, "flos": 22455377274720.0, "grad_norm": 2.389319961521174, "language_loss": 0.71828353, "learning_rate": 4.097829025746538e-07, "loss": 0.74010718, "num_input_tokens_seen": 143213550, "step": 6643, "time_per_iteration": 2.5617311000823975 }, { "auxiliary_loss_clip": 0.01050524, "auxiliary_loss_mlp": 0.0100121, "balance_loss_clip": 1.00682652, "balance_loss_mlp": 1.0, "epoch": 0.7988937654061203, "flos": 68864109972480.0, "grad_norm": 0.6588622254202351, "language_loss": 0.61063313, "learning_rate": 4.0931060249436757e-07, "loss": 0.63115048, "num_input_tokens_seen": 143277390, "step": 6644, "time_per_iteration": 3.130772113800049 }, { "auxiliary_loss_clip": 0.01156215, "auxiliary_loss_mlp": 0.0102793, "balance_loss_clip": 1.05003667, "balance_loss_mlp": 1.01968122, "epoch": 0.7990140082967595, "flos": 20806896700320.0, "grad_norm": 2.2620394647144626, "language_loss": 0.6972611, "learning_rate": 4.088385437203978e-07, "loss": 0.71910256, "num_input_tokens_seen": 143294400, "step": 6645, "time_per_iteration": 2.468806028366089 }, { "auxiliary_loss_clip": 0.01169051, "auxiliary_loss_mlp": 0.01027208, "balance_loss_clip": 1.04611254, "balance_loss_mlp": 1.02006781, "epoch": 0.7991342511873986, "flos": 18985292403360.0, "grad_norm": 2.653765829626635, "language_loss": 0.77583015, "learning_rate": 4.083667263243564e-07, "loss": 0.79779267, "num_input_tokens_seen": 143312745, "step": 6646, "time_per_iteration": 2.4473068714141846 }, { "auxiliary_loss_clip": 0.01154426, "auxiliary_loss_mlp": 0.01025946, "balance_loss_clip": 1.04956722, "balance_loss_mlp": 1.01900792, "epoch": 0.7992544940780376, "flos": 20816809781280.0, "grad_norm": 2.6827987982196797, "language_loss": 0.71609068, "learning_rate": 4.0789515037781653e-07, "loss": 0.73789442, "num_input_tokens_seen": 143333470, "step": 6647, "time_per_iteration": 2.4937238693237305 }, { "auxiliary_loss_clip": 0.0116041, "auxiliary_loss_mlp": 0.01027844, "balance_loss_clip": 1.04784536, "balance_loss_mlp": 1.02101684, "epoch": 0.7993747369686768, "flos": 12640775063040.0, "grad_norm": 1.9524639739934972, "language_loss": 0.8241483, "learning_rate": 4.0742381595231755e-07, "loss": 0.84603083, "num_input_tokens_seen": 143350195, "step": 6648, "time_per_iteration": 2.4637646675109863 }, { "auxiliary_loss_clip": 0.01126819, "auxiliary_loss_mlp": 0.01027957, "balance_loss_clip": 1.04430795, "balance_loss_mlp": 1.02092361, "epoch": 0.7994949798593158, "flos": 20078772185760.0, "grad_norm": 1.5942413231844648, "language_loss": 0.78080171, "learning_rate": 4.06952723119359e-07, "loss": 0.80234951, "num_input_tokens_seen": 143370070, "step": 6649, "time_per_iteration": 2.5499112606048584 }, { "auxiliary_loss_clip": 0.01132662, "auxiliary_loss_mlp": 0.01026582, "balance_loss_clip": 1.0443536, "balance_loss_mlp": 1.01981401, "epoch": 0.7996152227499549, "flos": 38654214596160.0, "grad_norm": 1.8349849706825285, "language_loss": 0.67183203, "learning_rate": 4.0648187195040504e-07, "loss": 0.69342446, "num_input_tokens_seen": 143392275, "step": 6650, "time_per_iteration": 2.668245792388916 }, { "auxiliary_loss_clip": 0.0104656, "auxiliary_loss_mlp": 0.01001429, "balance_loss_clip": 1.00559092, "balance_loss_mlp": 1.00018966, "epoch": 0.799735465640594, "flos": 70243832736480.0, "grad_norm": 0.8106717747764739, "language_loss": 0.67587864, "learning_rate": 4.060112625168848e-07, "loss": 0.69635856, "num_input_tokens_seen": 143457385, "step": 6651, "time_per_iteration": 3.939751386642456 }, { "auxiliary_loss_clip": 0.01171135, "auxiliary_loss_mlp": 0.01027797, "balance_loss_clip": 1.05029082, "balance_loss_mlp": 1.02037954, "epoch": 0.7998557085312331, "flos": 24240999166080.0, "grad_norm": 1.8658843815949557, "language_loss": 0.73965561, "learning_rate": 4.055408948901886e-07, "loss": 0.76164496, "num_input_tokens_seen": 143478785, "step": 6652, "time_per_iteration": 2.4850478172302246 }, { "auxiliary_loss_clip": 0.01160282, "auxiliary_loss_mlp": 0.01027989, "balance_loss_clip": 1.04762459, "balance_loss_mlp": 1.02084804, "epoch": 0.7999759514218722, "flos": 27564029639520.0, "grad_norm": 1.740230837983742, "language_loss": 0.70985746, "learning_rate": 4.050707691416708e-07, "loss": 0.73174012, "num_input_tokens_seen": 143500095, "step": 6653, "time_per_iteration": 2.5216851234436035 }, { "auxiliary_loss_clip": 0.01046425, "auxiliary_loss_mlp": 0.01001097, "balance_loss_clip": 1.00561285, "balance_loss_mlp": 0.99981546, "epoch": 0.8000961943125112, "flos": 67337440561440.0, "grad_norm": 0.6728716976869306, "language_loss": 0.59768426, "learning_rate": 4.046008853426495e-07, "loss": 0.61815947, "num_input_tokens_seen": 143563410, "step": 6654, "time_per_iteration": 3.1641581058502197 }, { "auxiliary_loss_clip": 0.01123239, "auxiliary_loss_mlp": 0.01024712, "balance_loss_clip": 1.04282928, "balance_loss_mlp": 1.01648045, "epoch": 0.8002164372031504, "flos": 28733820185280.0, "grad_norm": 1.5116736434985416, "language_loss": 0.62707013, "learning_rate": 4.0413124356440464e-07, "loss": 0.64854968, "num_input_tokens_seen": 143587455, "step": 6655, "time_per_iteration": 2.639913558959961 }, { "auxiliary_loss_clip": 0.01115277, "auxiliary_loss_mlp": 0.01029151, "balance_loss_clip": 1.04122519, "balance_loss_mlp": 1.02161372, "epoch": 0.8003366800937894, "flos": 17639432944320.0, "grad_norm": 1.936362933699692, "language_loss": 0.82348078, "learning_rate": 4.0366184387818223e-07, "loss": 0.84492511, "num_input_tokens_seen": 143605915, "step": 6656, "time_per_iteration": 2.556685209274292 }, { "auxiliary_loss_clip": 0.01174643, "auxiliary_loss_mlp": 0.01026417, "balance_loss_clip": 1.04882073, "balance_loss_mlp": 1.01891899, "epoch": 0.8004569229844285, "flos": 25995307856640.0, "grad_norm": 1.7676700799779173, "language_loss": 0.84985876, "learning_rate": 4.0319268635518797e-07, "loss": 0.87186939, "num_input_tokens_seen": 143626490, "step": 6657, "time_per_iteration": 2.5412230491638184 }, { "auxiliary_loss_clip": 0.01155107, "auxiliary_loss_mlp": 0.01024954, "balance_loss_clip": 1.04554319, "balance_loss_mlp": 1.01826382, "epoch": 0.8005771658750677, "flos": 20812356078240.0, "grad_norm": 2.4585314597219106, "language_loss": 0.75229084, "learning_rate": 4.027237710665943e-07, "loss": 0.77409148, "num_input_tokens_seen": 143644955, "step": 6658, "time_per_iteration": 2.5059938430786133 }, { "auxiliary_loss_clip": 0.01130764, "auxiliary_loss_mlp": 0.01027078, "balance_loss_clip": 1.0424217, "balance_loss_mlp": 1.01967525, "epoch": 0.8006974087657067, "flos": 25812630222720.0, "grad_norm": 1.8454931217452755, "language_loss": 0.68924475, "learning_rate": 4.022550980835344e-07, "loss": 0.71082318, "num_input_tokens_seen": 143667200, "step": 6659, "time_per_iteration": 2.6242024898529053 }, { "auxiliary_loss_clip": 0.01127357, "auxiliary_loss_mlp": 0.01024923, "balance_loss_clip": 1.04187918, "balance_loss_mlp": 1.01779461, "epoch": 0.8008176516563458, "flos": 17164693781280.0, "grad_norm": 2.2877739270794017, "language_loss": 0.79204059, "learning_rate": 4.017866674771051e-07, "loss": 0.81356335, "num_input_tokens_seen": 143684685, "step": 6660, "time_per_iteration": 2.5367679595947266 }, { "auxiliary_loss_clip": 0.01105175, "auxiliary_loss_mlp": 0.01020469, "balance_loss_clip": 1.04023504, "balance_loss_mlp": 1.01330757, "epoch": 0.8009378945469849, "flos": 24207315445920.0, "grad_norm": 1.6813790309863368, "language_loss": 0.74528044, "learning_rate": 4.013184793183688e-07, "loss": 0.76653683, "num_input_tokens_seen": 143706780, "step": 6661, "time_per_iteration": 2.643355131149292 }, { "auxiliary_loss_clip": 0.01154816, "auxiliary_loss_mlp": 0.01025928, "balance_loss_clip": 1.04530406, "balance_loss_mlp": 1.01899302, "epoch": 0.801058137437624, "flos": 19787321244960.0, "grad_norm": 1.81215116501832, "language_loss": 0.72456133, "learning_rate": 4.008505336783472e-07, "loss": 0.74636877, "num_input_tokens_seen": 143724505, "step": 6662, "time_per_iteration": 3.2487266063690186 }, { "auxiliary_loss_clip": 0.0114432, "auxiliary_loss_mlp": 0.01029748, "balance_loss_clip": 1.04431677, "balance_loss_mlp": 1.02315903, "epoch": 0.801178380328263, "flos": 18659403486240.0, "grad_norm": 4.069437247576184, "language_loss": 0.806045, "learning_rate": 4.003828306280284e-07, "loss": 0.82778567, "num_input_tokens_seen": 143742180, "step": 6663, "time_per_iteration": 2.497455596923828 }, { "auxiliary_loss_clip": 0.0115467, "auxiliary_loss_mlp": 0.01019556, "balance_loss_clip": 1.04778969, "balance_loss_mlp": 1.013134, "epoch": 0.8012986232189022, "flos": 15706577070240.0, "grad_norm": 1.7724907203550258, "language_loss": 0.78049862, "learning_rate": 3.999153702383626e-07, "loss": 0.80224085, "num_input_tokens_seen": 143760070, "step": 6664, "time_per_iteration": 3.255389928817749 }, { "auxiliary_loss_clip": 0.01159147, "auxiliary_loss_mlp": 0.01026931, "balance_loss_clip": 1.04634571, "balance_loss_mlp": 1.01980805, "epoch": 0.8014188661095413, "flos": 28584143849280.0, "grad_norm": 1.8271735316531437, "language_loss": 0.73821706, "learning_rate": 3.9944815258026263e-07, "loss": 0.76007783, "num_input_tokens_seen": 143781890, "step": 6665, "time_per_iteration": 2.548827648162842 }, { "auxiliary_loss_clip": 0.01159208, "auxiliary_loss_mlp": 0.01025937, "balance_loss_clip": 1.04716778, "balance_loss_mlp": 1.01854575, "epoch": 0.8015391090001803, "flos": 29310364764960.0, "grad_norm": 1.635162456999472, "language_loss": 0.82840443, "learning_rate": 3.989811777246057e-07, "loss": 0.85025597, "num_input_tokens_seen": 143802060, "step": 6666, "time_per_iteration": 2.549189329147339 }, { "auxiliary_loss_clip": 0.01060958, "auxiliary_loss_mlp": 0.01001656, "balance_loss_clip": 1.00581694, "balance_loss_mlp": 1.0004344, "epoch": 0.8016593518908195, "flos": 70397352187200.0, "grad_norm": 0.8534042088866907, "language_loss": 0.66276258, "learning_rate": 3.985144457422305e-07, "loss": 0.68338871, "num_input_tokens_seen": 143856345, "step": 6667, "time_per_iteration": 3.710405111312866 }, { "auxiliary_loss_clip": 0.01169741, "auxiliary_loss_mlp": 0.0102492, "balance_loss_clip": 1.04824901, "balance_loss_mlp": 1.01805949, "epoch": 0.8017795947814585, "flos": 26026118220000.0, "grad_norm": 1.8374132213871273, "language_loss": 0.76620609, "learning_rate": 3.9804795670394096e-07, "loss": 0.78815269, "num_input_tokens_seen": 143876470, "step": 6668, "time_per_iteration": 2.5143327713012695 }, { "auxiliary_loss_clip": 0.01131933, "auxiliary_loss_mlp": 0.01023927, "balance_loss_clip": 1.04335356, "balance_loss_mlp": 1.01704884, "epoch": 0.8018998376720976, "flos": 22087184567040.0, "grad_norm": 1.513891900122831, "language_loss": 0.7040264, "learning_rate": 3.975817106805022e-07, "loss": 0.72558498, "num_input_tokens_seen": 143895170, "step": 6669, "time_per_iteration": 2.5600931644439697 }, { "auxiliary_loss_clip": 0.01128318, "auxiliary_loss_mlp": 0.01031496, "balance_loss_clip": 1.04451215, "balance_loss_mlp": 1.02438211, "epoch": 0.8020200805627368, "flos": 34568549797920.0, "grad_norm": 1.8080593575518589, "language_loss": 0.65340066, "learning_rate": 3.97115707742645e-07, "loss": 0.67499876, "num_input_tokens_seen": 143915845, "step": 6670, "time_per_iteration": 2.718494415283203 }, { "auxiliary_loss_clip": 0.01145284, "auxiliary_loss_mlp": 0.01025913, "balance_loss_clip": 1.04692388, "balance_loss_mlp": 1.01936555, "epoch": 0.8021403234533758, "flos": 20120357637120.0, "grad_norm": 1.9615543671849218, "language_loss": 0.65015608, "learning_rate": 3.966499479610599e-07, "loss": 0.67186809, "num_input_tokens_seen": 143933940, "step": 6671, "time_per_iteration": 2.5251080989837646 }, { "auxiliary_loss_clip": 0.01126004, "auxiliary_loss_mlp": 0.01026869, "balance_loss_clip": 1.04591131, "balance_loss_mlp": 1.01964211, "epoch": 0.8022605663440149, "flos": 27746204436000.0, "grad_norm": 1.7757498701821788, "language_loss": 0.64878178, "learning_rate": 3.9618443140640225e-07, "loss": 0.6703105, "num_input_tokens_seen": 143952850, "step": 6672, "time_per_iteration": 2.639082670211792 }, { "auxiliary_loss_clip": 0.01019892, "auxiliary_loss_mlp": 0.01001682, "balance_loss_clip": 1.00654984, "balance_loss_mlp": 1.00027561, "epoch": 0.802380809234654, "flos": 60245008461600.0, "grad_norm": 0.6885906175207537, "language_loss": 0.51363027, "learning_rate": 3.957191581492918e-07, "loss": 0.53384602, "num_input_tokens_seen": 144013610, "step": 6673, "time_per_iteration": 3.1786930561065674 }, { "auxiliary_loss_clip": 0.01135281, "auxiliary_loss_mlp": 0.01029129, "balance_loss_clip": 1.0443747, "balance_loss_mlp": 1.02108574, "epoch": 0.8025010521252931, "flos": 15080724892800.0, "grad_norm": 3.0382021588551353, "language_loss": 0.71146643, "learning_rate": 3.952541282603097e-07, "loss": 0.73311055, "num_input_tokens_seen": 144028715, "step": 6674, "time_per_iteration": 2.5419585704803467 }, { "auxiliary_loss_clip": 0.01152334, "auxiliary_loss_mlp": 0.01026565, "balance_loss_clip": 1.04619002, "balance_loss_mlp": 1.01965404, "epoch": 0.8026212950159322, "flos": 22163531247360.0, "grad_norm": 7.654523099396626, "language_loss": 0.83522326, "learning_rate": 3.9478934181000013e-07, "loss": 0.85701227, "num_input_tokens_seen": 144048740, "step": 6675, "time_per_iteration": 2.4889464378356934 }, { "auxiliary_loss_clip": 0.01174215, "auxiliary_loss_mlp": 0.01030799, "balance_loss_clip": 1.04887211, "balance_loss_mlp": 1.02343822, "epoch": 0.8027415379065713, "flos": 17675990021280.0, "grad_norm": 2.588460380281255, "language_loss": 0.84331566, "learning_rate": 3.943247988688714e-07, "loss": 0.86536586, "num_input_tokens_seen": 144067435, "step": 6676, "time_per_iteration": 2.4905996322631836 }, { "auxiliary_loss_clip": 0.01156575, "auxiliary_loss_mlp": 0.01025332, "balance_loss_clip": 1.04657543, "balance_loss_mlp": 1.01899362, "epoch": 0.8028617807972104, "flos": 21979596519840.0, "grad_norm": 1.8083075395651689, "language_loss": 0.72076982, "learning_rate": 3.938604995073933e-07, "loss": 0.74258888, "num_input_tokens_seen": 144085905, "step": 6677, "time_per_iteration": 3.2948508262634277 }, { "auxiliary_loss_clip": 0.01142175, "auxiliary_loss_mlp": 0.01024142, "balance_loss_clip": 1.04333544, "balance_loss_mlp": 1.01689959, "epoch": 0.8029820236878494, "flos": 26428461568320.0, "grad_norm": 1.8367228379754212, "language_loss": 0.65551543, "learning_rate": 3.9339644379600157e-07, "loss": 0.67717856, "num_input_tokens_seen": 144105735, "step": 6678, "time_per_iteration": 2.612734079360962 }, { "auxiliary_loss_clip": 0.01159276, "auxiliary_loss_mlp": 0.01023949, "balance_loss_clip": 1.05129886, "balance_loss_mlp": 1.01711512, "epoch": 0.8031022665784886, "flos": 17676492858720.0, "grad_norm": 1.9497336070451183, "language_loss": 0.71180093, "learning_rate": 3.929326318050907e-07, "loss": 0.73363316, "num_input_tokens_seen": 144123405, "step": 6679, "time_per_iteration": 2.4683446884155273 }, { "auxiliary_loss_clip": 0.01164571, "auxiliary_loss_mlp": 0.01024685, "balance_loss_clip": 1.04475713, "balance_loss_mlp": 1.01810765, "epoch": 0.8032225094691277, "flos": 15450282444960.0, "grad_norm": 1.848280044546713, "language_loss": 0.78799736, "learning_rate": 3.924690636050225e-07, "loss": 0.80988997, "num_input_tokens_seen": 144140815, "step": 6680, "time_per_iteration": 2.43861985206604 }, { "auxiliary_loss_clip": 0.01156035, "auxiliary_loss_mlp": 0.0102057, "balance_loss_clip": 1.04689479, "balance_loss_mlp": 1.01282382, "epoch": 0.8033427523597667, "flos": 26179206667200.0, "grad_norm": 2.161655914830005, "language_loss": 0.72997725, "learning_rate": 3.9200573926611915e-07, "loss": 0.75174326, "num_input_tokens_seen": 144162230, "step": 6681, "time_per_iteration": 2.5532655715942383 }, { "auxiliary_loss_clip": 0.01157607, "auxiliary_loss_mlp": 0.01020622, "balance_loss_clip": 1.05039239, "balance_loss_mlp": 1.01312411, "epoch": 0.8034629952504058, "flos": 21324909411840.0, "grad_norm": 1.9023785064638878, "language_loss": 0.72839034, "learning_rate": 3.9154265885866613e-07, "loss": 0.75017267, "num_input_tokens_seen": 144181540, "step": 6682, "time_per_iteration": 2.51423716545105 }, { "auxiliary_loss_clip": 0.01156269, "auxiliary_loss_mlp": 0.01028671, "balance_loss_clip": 1.04879248, "balance_loss_mlp": 1.02135479, "epoch": 0.8035832381410449, "flos": 21651588502080.0, "grad_norm": 2.4297253335946447, "language_loss": 0.74787617, "learning_rate": 3.9107982245291394e-07, "loss": 0.76972562, "num_input_tokens_seen": 144199665, "step": 6683, "time_per_iteration": 2.4794180393218994 }, { "auxiliary_loss_clip": 0.01130251, "auxiliary_loss_mlp": 0.01024445, "balance_loss_clip": 1.04815388, "balance_loss_mlp": 1.01701546, "epoch": 0.803703481031684, "flos": 20518821953760.0, "grad_norm": 2.0276078141442957, "language_loss": 0.77186567, "learning_rate": 3.9061723011907245e-07, "loss": 0.79341263, "num_input_tokens_seen": 144219020, "step": 6684, "time_per_iteration": 2.55173921585083 }, { "auxiliary_loss_clip": 0.01141588, "auxiliary_loss_mlp": 0.01026155, "balance_loss_clip": 1.04478669, "balance_loss_mlp": 1.01856697, "epoch": 0.803823723922323, "flos": 22854811349280.0, "grad_norm": 1.668423626598389, "language_loss": 0.79349416, "learning_rate": 3.901548819273179e-07, "loss": 0.8151716, "num_input_tokens_seen": 144239035, "step": 6685, "time_per_iteration": 2.526811122894287 }, { "auxiliary_loss_clip": 0.01158669, "auxiliary_loss_mlp": 0.01026154, "balance_loss_clip": 1.04948878, "balance_loss_mlp": 1.01896548, "epoch": 0.8039439668129622, "flos": 21362148911040.0, "grad_norm": 1.7250965395183633, "language_loss": 0.69388843, "learning_rate": 3.896927779477881e-07, "loss": 0.71573669, "num_input_tokens_seen": 144258295, "step": 6686, "time_per_iteration": 2.5046868324279785 }, { "auxiliary_loss_clip": 0.01127737, "auxiliary_loss_mlp": 0.01027578, "balance_loss_clip": 1.04389083, "balance_loss_mlp": 1.0206672, "epoch": 0.8040642097036013, "flos": 23802386076480.0, "grad_norm": 1.984933003959235, "language_loss": 0.67228985, "learning_rate": 3.892309182505833e-07, "loss": 0.69384301, "num_input_tokens_seen": 144276110, "step": 6687, "time_per_iteration": 2.590496063232422 }, { "auxiliary_loss_clip": 0.01169673, "auxiliary_loss_mlp": 0.01025201, "balance_loss_clip": 1.04776633, "balance_loss_mlp": 1.01844501, "epoch": 0.8041844525942403, "flos": 25922050034880.0, "grad_norm": 2.139368707078474, "language_loss": 0.85871828, "learning_rate": 3.887693029057675e-07, "loss": 0.88066709, "num_input_tokens_seen": 144295620, "step": 6688, "time_per_iteration": 2.482492208480835 }, { "auxiliary_loss_clip": 0.0114266, "auxiliary_loss_mlp": 0.01021631, "balance_loss_clip": 1.04590821, "balance_loss_mlp": 1.01520526, "epoch": 0.8043046954848795, "flos": 25191123997440.0, "grad_norm": 1.6953307954752368, "language_loss": 0.81201518, "learning_rate": 3.8830793198336684e-07, "loss": 0.83365798, "num_input_tokens_seen": 144315210, "step": 6689, "time_per_iteration": 3.3254053592681885 }, { "auxiliary_loss_clip": 0.01157781, "auxiliary_loss_mlp": 0.01029178, "balance_loss_clip": 1.04671931, "balance_loss_mlp": 1.02203107, "epoch": 0.8044249383755185, "flos": 41719190513280.0, "grad_norm": 1.6646296935201372, "language_loss": 0.70463198, "learning_rate": 3.878468055533721e-07, "loss": 0.72650152, "num_input_tokens_seen": 144337750, "step": 6690, "time_per_iteration": 3.4432296752929688 }, { "auxiliary_loss_clip": 0.0113431, "auxiliary_loss_mlp": 0.01024704, "balance_loss_clip": 1.04670334, "balance_loss_mlp": 1.01746833, "epoch": 0.8045451812661576, "flos": 20631438375360.0, "grad_norm": 2.5169865963880635, "language_loss": 0.84931481, "learning_rate": 3.8738592368573464e-07, "loss": 0.87090492, "num_input_tokens_seen": 144355305, "step": 6691, "time_per_iteration": 2.554055690765381 }, { "auxiliary_loss_clip": 0.01116389, "auxiliary_loss_mlp": 0.01024175, "balance_loss_clip": 1.04281974, "balance_loss_mlp": 1.01682329, "epoch": 0.8046654241567968, "flos": 29711810189280.0, "grad_norm": 1.8368117463519795, "language_loss": 0.87965983, "learning_rate": 3.8692528645037137e-07, "loss": 0.90106547, "num_input_tokens_seen": 144374485, "step": 6692, "time_per_iteration": 2.6021666526794434 }, { "auxiliary_loss_clip": 0.01169219, "auxiliary_loss_mlp": 0.0102205, "balance_loss_clip": 1.04826117, "balance_loss_mlp": 1.01532686, "epoch": 0.8047856670474358, "flos": 17671392650400.0, "grad_norm": 4.6314646945690825, "language_loss": 0.77492404, "learning_rate": 3.8646489391715907e-07, "loss": 0.79683673, "num_input_tokens_seen": 144388780, "step": 6693, "time_per_iteration": 3.135673761367798 }, { "auxiliary_loss_clip": 0.011435, "auxiliary_loss_mlp": 0.01025043, "balance_loss_clip": 1.0476141, "balance_loss_mlp": 1.0172348, "epoch": 0.8049059099380749, "flos": 17120701893600.0, "grad_norm": 2.490785036645909, "language_loss": 0.87874287, "learning_rate": 3.8600474615593903e-07, "loss": 0.9004283, "num_input_tokens_seen": 144403395, "step": 6694, "time_per_iteration": 2.496014356613159 }, { "auxiliary_loss_clip": 0.01032203, "auxiliary_loss_mlp": 0.01001971, "balance_loss_clip": 1.00633085, "balance_loss_mlp": 1.00073171, "epoch": 0.805026152828714, "flos": 62212912900320.0, "grad_norm": 0.7803998841671158, "language_loss": 0.59664178, "learning_rate": 3.8554484323651605e-07, "loss": 0.61698353, "num_input_tokens_seen": 144465265, "step": 6695, "time_per_iteration": 3.1942691802978516 }, { "auxiliary_loss_clip": 0.01155682, "auxiliary_loss_mlp": 0.00762362, "balance_loss_clip": 1.04912746, "balance_loss_mlp": 1.00056911, "epoch": 0.8051463957193531, "flos": 21688612499520.0, "grad_norm": 1.47542393146419, "language_loss": 0.79035509, "learning_rate": 3.85085185228657e-07, "loss": 0.80953556, "num_input_tokens_seen": 144484235, "step": 6696, "time_per_iteration": 2.5260491371154785 }, { "auxiliary_loss_clip": 0.01134824, "auxiliary_loss_mlp": 0.01021914, "balance_loss_clip": 1.04402494, "balance_loss_mlp": 1.01483297, "epoch": 0.8052666386099921, "flos": 32051463114720.0, "grad_norm": 2.1472833296493494, "language_loss": 0.73164117, "learning_rate": 3.8462577220209114e-07, "loss": 0.75320852, "num_input_tokens_seen": 144504610, "step": 6697, "time_per_iteration": 2.6021499633789062 }, { "auxiliary_loss_clip": 0.01061004, "auxiliary_loss_mlp": 0.01002387, "balance_loss_clip": 1.00573373, "balance_loss_mlp": 1.00111187, "epoch": 0.8053868815006313, "flos": 67157887703040.0, "grad_norm": 0.7073228912947155, "language_loss": 0.59009433, "learning_rate": 3.8416660422651127e-07, "loss": 0.61072826, "num_input_tokens_seen": 144574260, "step": 6698, "time_per_iteration": 3.1646265983581543 }, { "auxiliary_loss_clip": 0.01129646, "auxiliary_loss_mlp": 0.01029744, "balance_loss_clip": 1.04227543, "balance_loss_mlp": 1.02253461, "epoch": 0.8055071243912704, "flos": 23837003637600.0, "grad_norm": 1.800862578100664, "language_loss": 0.67941391, "learning_rate": 3.837076813715723e-07, "loss": 0.70100784, "num_input_tokens_seen": 144594145, "step": 6699, "time_per_iteration": 2.5638575553894043 }, { "auxiliary_loss_clip": 0.01125762, "auxiliary_loss_mlp": 0.01023672, "balance_loss_clip": 1.04310095, "balance_loss_mlp": 1.01592958, "epoch": 0.8056273672819094, "flos": 21324514325280.0, "grad_norm": 1.7292038978254805, "language_loss": 0.74792856, "learning_rate": 3.832490037068941e-07, "loss": 0.76942289, "num_input_tokens_seen": 144612935, "step": 6700, "time_per_iteration": 2.5652716159820557 }, { "auxiliary_loss_clip": 0.01094972, "auxiliary_loss_mlp": 0.01025982, "balance_loss_clip": 1.0399878, "balance_loss_mlp": 1.01856732, "epoch": 0.8057476101725486, "flos": 25768386916320.0, "grad_norm": 2.0389881933809018, "language_loss": 0.75947279, "learning_rate": 3.827905713020554e-07, "loss": 0.78068233, "num_input_tokens_seen": 144630580, "step": 6701, "time_per_iteration": 2.6312928199768066 }, { "auxiliary_loss_clip": 0.01129186, "auxiliary_loss_mlp": 0.01035126, "balance_loss_clip": 1.04083014, "balance_loss_mlp": 1.02721047, "epoch": 0.8058678530631876, "flos": 24535287546720.0, "grad_norm": 2.1149495572797927, "language_loss": 0.68373263, "learning_rate": 3.823323842266017e-07, "loss": 0.70537573, "num_input_tokens_seen": 144649975, "step": 6702, "time_per_iteration": 2.6036624908447266 }, { "auxiliary_loss_clip": 0.01156563, "auxiliary_loss_mlp": 0.01025925, "balance_loss_clip": 1.04472446, "balance_loss_mlp": 1.01833129, "epoch": 0.8059880959538267, "flos": 24753732084480.0, "grad_norm": 3.3048316848791393, "language_loss": 0.72673786, "learning_rate": 3.818744425500393e-07, "loss": 0.74856281, "num_input_tokens_seen": 144667990, "step": 6703, "time_per_iteration": 3.2921788692474365 }, { "auxiliary_loss_clip": 0.01120853, "auxiliary_loss_mlp": 0.01028248, "balance_loss_clip": 1.04118884, "balance_loss_mlp": 1.02129519, "epoch": 0.8061083388444659, "flos": 22196352960480.0, "grad_norm": 2.737046243570948, "language_loss": 0.80240619, "learning_rate": 3.8141674634183675e-07, "loss": 0.82389724, "num_input_tokens_seen": 144687020, "step": 6704, "time_per_iteration": 2.545689821243286 }, { "auxiliary_loss_clip": 0.01114019, "auxiliary_loss_mlp": 0.01026732, "balance_loss_clip": 1.04511821, "balance_loss_mlp": 1.02066433, "epoch": 0.8062285817351049, "flos": 30044200076160.0, "grad_norm": 1.9679088158715743, "language_loss": 0.66158569, "learning_rate": 3.809592956714278e-07, "loss": 0.68299317, "num_input_tokens_seen": 144710255, "step": 6705, "time_per_iteration": 2.6503419876098633 }, { "auxiliary_loss_clip": 0.01160642, "auxiliary_loss_mlp": 0.01022944, "balance_loss_clip": 1.04921615, "balance_loss_mlp": 1.01622331, "epoch": 0.806348824625744, "flos": 22782595119360.0, "grad_norm": 1.9667372883119139, "language_loss": 0.74551904, "learning_rate": 3.805020906082057e-07, "loss": 0.76735485, "num_input_tokens_seen": 144728830, "step": 6706, "time_per_iteration": 2.506087303161621 }, { "auxiliary_loss_clip": 0.01144816, "auxiliary_loss_mlp": 0.01028921, "balance_loss_clip": 1.0462327, "balance_loss_mlp": 1.0216701, "epoch": 0.8064690675163831, "flos": 23404604182080.0, "grad_norm": 2.1752607555097017, "language_loss": 0.81366003, "learning_rate": 3.8004513122152917e-07, "loss": 0.83539736, "num_input_tokens_seen": 144747140, "step": 6707, "time_per_iteration": 2.583200216293335 }, { "auxiliary_loss_clip": 0.01132386, "auxiliary_loss_mlp": 0.01030654, "balance_loss_clip": 1.0459075, "balance_loss_mlp": 1.02382684, "epoch": 0.8065893104070222, "flos": 24060907553280.0, "grad_norm": 7.269684839624203, "language_loss": 0.67521548, "learning_rate": 3.79588417580718e-07, "loss": 0.69684589, "num_input_tokens_seen": 144765250, "step": 6708, "time_per_iteration": 2.5535449981689453 }, { "auxiliary_loss_clip": 0.01155819, "auxiliary_loss_mlp": 0.01026681, "balance_loss_clip": 1.04674065, "balance_loss_mlp": 1.01984453, "epoch": 0.8067095532976613, "flos": 22305413603040.0, "grad_norm": 1.932528066459662, "language_loss": 0.76704127, "learning_rate": 3.791319497550558e-07, "loss": 0.78886628, "num_input_tokens_seen": 144783080, "step": 6709, "time_per_iteration": 2.5150530338287354 }, { "auxiliary_loss_clip": 0.01133274, "auxiliary_loss_mlp": 0.00761662, "balance_loss_clip": 1.04696608, "balance_loss_mlp": 1.0005703, "epoch": 0.8068297961883004, "flos": 17129501548800.0, "grad_norm": 2.674293336140767, "language_loss": 0.70444739, "learning_rate": 3.78675727813788e-07, "loss": 0.72339678, "num_input_tokens_seen": 144800645, "step": 6710, "time_per_iteration": 2.515026807785034 }, { "auxiliary_loss_clip": 0.01142198, "auxiliary_loss_mlp": 0.01025285, "balance_loss_clip": 1.04741359, "balance_loss_mlp": 1.01800752, "epoch": 0.8069500390789395, "flos": 22018847368800.0, "grad_norm": 1.7999066804238864, "language_loss": 0.73262322, "learning_rate": 3.782197518261225e-07, "loss": 0.75429809, "num_input_tokens_seen": 144820085, "step": 6711, "time_per_iteration": 2.5405845642089844 }, { "auxiliary_loss_clip": 0.01146749, "auxiliary_loss_mlp": 0.01026182, "balance_loss_clip": 1.0464834, "balance_loss_mlp": 1.018713, "epoch": 0.8070702819695785, "flos": 19244244883680.0, "grad_norm": 1.9145031972544289, "language_loss": 0.9536249, "learning_rate": 3.777640218612319e-07, "loss": 0.97535419, "num_input_tokens_seen": 144838070, "step": 6712, "time_per_iteration": 2.5055558681488037 }, { "auxiliary_loss_clip": 0.01148871, "auxiliary_loss_mlp": 0.01024997, "balance_loss_clip": 1.04536843, "balance_loss_mlp": 1.01834798, "epoch": 0.8071905248602176, "flos": 21544323707520.0, "grad_norm": 2.032509368718819, "language_loss": 0.71433043, "learning_rate": 3.773085379882488e-07, "loss": 0.73606908, "num_input_tokens_seen": 144857125, "step": 6713, "time_per_iteration": 2.502865791320801 }, { "auxiliary_loss_clip": 0.01154935, "auxiliary_loss_mlp": 0.00762871, "balance_loss_clip": 1.04469657, "balance_loss_mlp": 1.00054514, "epoch": 0.8073107677508568, "flos": 37268314115040.0, "grad_norm": 1.778088781530488, "language_loss": 0.76124346, "learning_rate": 3.768533002762715e-07, "loss": 0.78042156, "num_input_tokens_seen": 144880660, "step": 6714, "time_per_iteration": 3.3971633911132812 }, { "auxiliary_loss_clip": 0.01144668, "auxiliary_loss_mlp": 0.01024863, "balance_loss_clip": 1.04531431, "balance_loss_mlp": 1.01819634, "epoch": 0.8074310106414958, "flos": 28366273982880.0, "grad_norm": 1.9809164411052231, "language_loss": 0.77198952, "learning_rate": 3.763983087943572e-07, "loss": 0.79368484, "num_input_tokens_seen": 144900050, "step": 6715, "time_per_iteration": 2.596614122390747 }, { "auxiliary_loss_clip": 0.01143201, "auxiliary_loss_mlp": 0.00761583, "balance_loss_clip": 1.04217887, "balance_loss_mlp": 1.00051999, "epoch": 0.8075512535321349, "flos": 24281650776480.0, "grad_norm": 1.741797503461573, "language_loss": 0.80596006, "learning_rate": 3.759435636115282e-07, "loss": 0.82500786, "num_input_tokens_seen": 144920835, "step": 6716, "time_per_iteration": 2.5527970790863037 }, { "auxiliary_loss_clip": 0.01098487, "auxiliary_loss_mlp": 0.00762562, "balance_loss_clip": 1.04399598, "balance_loss_mlp": 1.00058007, "epoch": 0.807671496422774, "flos": 26030859258720.0, "grad_norm": 1.7620017338432017, "language_loss": 0.72842848, "learning_rate": 3.7548906479676967e-07, "loss": 0.74703896, "num_input_tokens_seen": 144940430, "step": 6717, "time_per_iteration": 3.4447383880615234 }, { "auxiliary_loss_clip": 0.01158871, "auxiliary_loss_mlp": 0.01021999, "balance_loss_clip": 1.04570937, "balance_loss_mlp": 1.01460838, "epoch": 0.8077917393134131, "flos": 23730744517920.0, "grad_norm": 1.6651866916129316, "language_loss": 0.71660334, "learning_rate": 3.7503481241902855e-07, "loss": 0.73841202, "num_input_tokens_seen": 144960405, "step": 6718, "time_per_iteration": 2.541361093521118 }, { "auxiliary_loss_clip": 0.01138441, "auxiliary_loss_mlp": 0.00761683, "balance_loss_clip": 1.04332805, "balance_loss_mlp": 1.00060201, "epoch": 0.8079119822040521, "flos": 18402031352160.0, "grad_norm": 1.6897104424937075, "language_loss": 0.8017872, "learning_rate": 3.745808065472145e-07, "loss": 0.82078844, "num_input_tokens_seen": 144977700, "step": 6719, "time_per_iteration": 3.2115235328674316 }, { "auxiliary_loss_clip": 0.01154851, "auxiliary_loss_mlp": 0.01029263, "balance_loss_clip": 1.05219293, "balance_loss_mlp": 1.02292442, "epoch": 0.8080322250946913, "flos": 23621791626240.0, "grad_norm": 1.6869628391035882, "language_loss": 0.76397634, "learning_rate": 3.741270472501994e-07, "loss": 0.78581738, "num_input_tokens_seen": 144998340, "step": 6720, "time_per_iteration": 2.5630624294281006 }, { "auxiliary_loss_clip": 0.0114084, "auxiliary_loss_mlp": 0.01026133, "balance_loss_clip": 1.04750645, "balance_loss_mlp": 1.01885509, "epoch": 0.8081524679853304, "flos": 22820696625600.0, "grad_norm": 1.6016246654671642, "language_loss": 0.72574639, "learning_rate": 3.736735345968183e-07, "loss": 0.74741614, "num_input_tokens_seen": 145017950, "step": 6721, "time_per_iteration": 2.5295515060424805 }, { "auxiliary_loss_clip": 0.01156591, "auxiliary_loss_mlp": 0.01029363, "balance_loss_clip": 1.04783392, "balance_loss_mlp": 1.0226779, "epoch": 0.8082727108759694, "flos": 17640007615680.0, "grad_norm": 1.6017005593674492, "language_loss": 0.78711116, "learning_rate": 3.7322026865586986e-07, "loss": 0.80897063, "num_input_tokens_seen": 145036985, "step": 6722, "time_per_iteration": 2.5203630924224854 }, { "auxiliary_loss_clip": 0.01163078, "auxiliary_loss_mlp": 0.0102587, "balance_loss_clip": 1.04927683, "balance_loss_mlp": 1.01869059, "epoch": 0.8083929537666086, "flos": 25958176108320.0, "grad_norm": 1.9478276956579155, "language_loss": 0.73282707, "learning_rate": 3.7276724949611206e-07, "loss": 0.75471652, "num_input_tokens_seen": 145057095, "step": 6723, "time_per_iteration": 2.532534599304199 }, { "auxiliary_loss_clip": 0.01145231, "auxiliary_loss_mlp": 0.01028122, "balance_loss_clip": 1.0467577, "balance_loss_mlp": 1.0203433, "epoch": 0.8085131966572476, "flos": 27089182725600.0, "grad_norm": 1.789950147176967, "language_loss": 0.7488488, "learning_rate": 3.723144771862694e-07, "loss": 0.77058232, "num_input_tokens_seen": 145077735, "step": 6724, "time_per_iteration": 2.5698745250701904 }, { "auxiliary_loss_clip": 0.01130629, "auxiliary_loss_mlp": 0.01024137, "balance_loss_clip": 1.04310536, "balance_loss_mlp": 1.01702332, "epoch": 0.8086334395478867, "flos": 23988547655520.0, "grad_norm": 1.5201780865169954, "language_loss": 0.77092201, "learning_rate": 3.718619517950263e-07, "loss": 0.79246968, "num_input_tokens_seen": 145098330, "step": 6725, "time_per_iteration": 2.5881073474884033 }, { "auxiliary_loss_clip": 0.01171643, "auxiliary_loss_mlp": 0.01026571, "balance_loss_clip": 1.05099201, "balance_loss_mlp": 1.01979363, "epoch": 0.8087536824385259, "flos": 20405882279520.0, "grad_norm": 1.9433448346245878, "language_loss": 0.76730347, "learning_rate": 3.714096733910301e-07, "loss": 0.78928554, "num_input_tokens_seen": 145115855, "step": 6726, "time_per_iteration": 2.4441277980804443 }, { "auxiliary_loss_clip": 0.01161064, "auxiliary_loss_mlp": 0.01029233, "balance_loss_clip": 1.04949951, "balance_loss_mlp": 1.02158308, "epoch": 0.8088739253291649, "flos": 25919643598560.0, "grad_norm": 2.496142634965004, "language_loss": 0.70078301, "learning_rate": 3.709576420428926e-07, "loss": 0.72268593, "num_input_tokens_seen": 145136655, "step": 6727, "time_per_iteration": 2.506868600845337 }, { "auxiliary_loss_clip": 0.01141732, "auxiliary_loss_mlp": 0.0102038, "balance_loss_clip": 1.04261947, "balance_loss_mlp": 1.01334631, "epoch": 0.808994168219804, "flos": 28402076803680.0, "grad_norm": 2.550332788884936, "language_loss": 0.73743856, "learning_rate": 3.7050585781918463e-07, "loss": 0.75905967, "num_input_tokens_seen": 145156955, "step": 6728, "time_per_iteration": 2.5585711002349854 }, { "auxiliary_loss_clip": 0.0116109, "auxiliary_loss_mlp": 0.01028593, "balance_loss_clip": 1.0486877, "balance_loss_mlp": 1.02099371, "epoch": 0.8091144111104431, "flos": 17421060240480.0, "grad_norm": 2.1387739235962044, "language_loss": 0.68986726, "learning_rate": 3.700543207884428e-07, "loss": 0.7117641, "num_input_tokens_seen": 145173865, "step": 6729, "time_per_iteration": 3.2555947303771973 }, { "auxiliary_loss_clip": 0.01154674, "auxiliary_loss_mlp": 0.01024067, "balance_loss_clip": 1.04765403, "balance_loss_mlp": 1.01743042, "epoch": 0.8092346540010822, "flos": 32153807285760.0, "grad_norm": 1.657873186833859, "language_loss": 0.71085155, "learning_rate": 3.6960303101916466e-07, "loss": 0.73263896, "num_input_tokens_seen": 145193780, "step": 6730, "time_per_iteration": 2.5411970615386963 }, { "auxiliary_loss_clip": 0.01060559, "auxiliary_loss_mlp": 0.00752695, "balance_loss_clip": 1.00553083, "balance_loss_mlp": 0.9999916, "epoch": 0.8093548968917212, "flos": 58035104347680.0, "grad_norm": 0.7398877858426172, "language_loss": 0.5564326, "learning_rate": 3.6915198857981047e-07, "loss": 0.57456511, "num_input_tokens_seen": 145258980, "step": 6731, "time_per_iteration": 3.1165616512298584 }, { "auxiliary_loss_clip": 0.01125313, "auxiliary_loss_mlp": 0.01028124, "balance_loss_clip": 1.044994, "balance_loss_mlp": 1.02090931, "epoch": 0.8094751397823604, "flos": 27381603424320.0, "grad_norm": 1.832300704709978, "language_loss": 0.68389261, "learning_rate": 3.687011935388027e-07, "loss": 0.70542693, "num_input_tokens_seen": 145281875, "step": 6732, "time_per_iteration": 2.607858180999756 }, { "auxiliary_loss_clip": 0.01154941, "auxiliary_loss_mlp": 0.0102302, "balance_loss_clip": 1.04711437, "balance_loss_mlp": 1.01600718, "epoch": 0.8095953826729995, "flos": 24061087138080.0, "grad_norm": 1.9255398105658004, "language_loss": 0.72745907, "learning_rate": 3.6825064596452646e-07, "loss": 0.74923873, "num_input_tokens_seen": 145302220, "step": 6733, "time_per_iteration": 2.5317509174346924 }, { "auxiliary_loss_clip": 0.01153409, "auxiliary_loss_mlp": 0.0102552, "balance_loss_clip": 1.04439139, "balance_loss_mlp": 1.01847517, "epoch": 0.8097156255636385, "flos": 23951415907200.0, "grad_norm": 1.583427640624157, "language_loss": 0.70358092, "learning_rate": 3.678003459253305e-07, "loss": 0.72537029, "num_input_tokens_seen": 145323070, "step": 6734, "time_per_iteration": 2.547762870788574 }, { "auxiliary_loss_clip": 0.01126408, "auxiliary_loss_mlp": 0.01026436, "balance_loss_clip": 1.04373145, "balance_loss_mlp": 1.01927137, "epoch": 0.8098358684542777, "flos": 21799145737440.0, "grad_norm": 3.4840341006332807, "language_loss": 0.73937619, "learning_rate": 3.673502934895236e-07, "loss": 0.76090467, "num_input_tokens_seen": 145342575, "step": 6735, "time_per_iteration": 2.5432536602020264 }, { "auxiliary_loss_clip": 0.01060766, "auxiliary_loss_mlp": 0.01000019, "balance_loss_clip": 1.00558424, "balance_loss_mlp": 0.99879736, "epoch": 0.8099561113449167, "flos": 68809528969920.0, "grad_norm": 0.6972440393524603, "language_loss": 0.57944632, "learning_rate": 3.669004887253802e-07, "loss": 0.60005414, "num_input_tokens_seen": 145408865, "step": 6736, "time_per_iteration": 3.190140724182129 }, { "auxiliary_loss_clip": 0.01147638, "auxiliary_loss_mlp": 0.01025798, "balance_loss_clip": 1.04942429, "balance_loss_mlp": 1.01959944, "epoch": 0.8100763542355558, "flos": 23586060639360.0, "grad_norm": 1.616514991181213, "language_loss": 0.78709602, "learning_rate": 3.664509317011335e-07, "loss": 0.80883044, "num_input_tokens_seen": 145429200, "step": 6737, "time_per_iteration": 2.585573673248291 }, { "auxiliary_loss_clip": 0.01156543, "auxiliary_loss_mlp": 0.0102833, "balance_loss_clip": 1.04858565, "balance_loss_mlp": 1.02112389, "epoch": 0.810196597126195, "flos": 31650412776960.0, "grad_norm": 1.9191004682012864, "language_loss": 0.73590982, "learning_rate": 3.6600162248498134e-07, "loss": 0.75775856, "num_input_tokens_seen": 145452830, "step": 6738, "time_per_iteration": 2.5706498622894287 }, { "auxiliary_loss_clip": 0.01080733, "auxiliary_loss_mlp": 0.01024384, "balance_loss_clip": 1.03661895, "balance_loss_mlp": 1.01776791, "epoch": 0.810316840016834, "flos": 24900463229760.0, "grad_norm": 1.7352348695123683, "language_loss": 0.76230896, "learning_rate": 3.6555256114508426e-07, "loss": 0.78336012, "num_input_tokens_seen": 145472625, "step": 6739, "time_per_iteration": 2.6338863372802734 }, { "auxiliary_loss_clip": 0.01141988, "auxiliary_loss_mlp": 0.01025321, "balance_loss_clip": 1.04272389, "balance_loss_mlp": 1.01783776, "epoch": 0.8104370829074731, "flos": 27965008143360.0, "grad_norm": 2.0070427240137976, "language_loss": 0.73079634, "learning_rate": 3.651037477495642e-07, "loss": 0.75246942, "num_input_tokens_seen": 145494075, "step": 6740, "time_per_iteration": 3.3621692657470703 }, { "auxiliary_loss_clip": 0.01166705, "auxiliary_loss_mlp": 0.01025071, "balance_loss_clip": 1.04539061, "balance_loss_mlp": 1.01783776, "epoch": 0.8105573257981122, "flos": 24640756493280.0, "grad_norm": 1.918032626573998, "language_loss": 0.68028295, "learning_rate": 3.6465518236650584e-07, "loss": 0.70220077, "num_input_tokens_seen": 145514220, "step": 6741, "time_per_iteration": 2.4727611541748047 }, { "auxiliary_loss_clip": 0.01124987, "auxiliary_loss_mlp": 0.01022063, "balance_loss_clip": 1.04198027, "balance_loss_mlp": 1.01524174, "epoch": 0.8106775686887513, "flos": 26358939110400.0, "grad_norm": 1.8732501565897772, "language_loss": 0.7818765, "learning_rate": 3.642068650639558e-07, "loss": 0.80334699, "num_input_tokens_seen": 145533965, "step": 6742, "time_per_iteration": 3.3919191360473633 }, { "auxiliary_loss_clip": 0.01132309, "auxiliary_loss_mlp": 0.01026769, "balance_loss_clip": 1.03934741, "balance_loss_mlp": 1.01992893, "epoch": 0.8107978115793903, "flos": 27271896276480.0, "grad_norm": 1.8593350892879101, "language_loss": 0.64701408, "learning_rate": 3.6375879590992334e-07, "loss": 0.66860485, "num_input_tokens_seen": 145554310, "step": 6743, "time_per_iteration": 2.563676595687866 }, { "auxiliary_loss_clip": 0.01136398, "auxiliary_loss_mlp": 0.01027002, "balance_loss_clip": 1.04363823, "balance_loss_mlp": 1.02018952, "epoch": 0.8109180544700295, "flos": 24934326534720.0, "grad_norm": 2.043025738543949, "language_loss": 0.81117946, "learning_rate": 3.6331097497238173e-07, "loss": 0.8328135, "num_input_tokens_seen": 145573755, "step": 6744, "time_per_iteration": 2.5688223838806152 }, { "auxiliary_loss_clip": 0.01123534, "auxiliary_loss_mlp": 0.01025524, "balance_loss_clip": 1.04355657, "balance_loss_mlp": 1.01921797, "epoch": 0.8110382973606686, "flos": 21105387365280.0, "grad_norm": 1.802107141060569, "language_loss": 0.7986058, "learning_rate": 3.628634023192627e-07, "loss": 0.82009631, "num_input_tokens_seen": 145594000, "step": 6745, "time_per_iteration": 3.3094489574432373 }, { "auxiliary_loss_clip": 0.01157506, "auxiliary_loss_mlp": 0.01021464, "balance_loss_clip": 1.04558122, "balance_loss_mlp": 1.01383471, "epoch": 0.8111585402513076, "flos": 15414084537600.0, "grad_norm": 2.1839413939628067, "language_loss": 0.75043643, "learning_rate": 3.624160780184644e-07, "loss": 0.7722261, "num_input_tokens_seen": 145611215, "step": 6746, "time_per_iteration": 2.4749577045440674 }, { "auxiliary_loss_clip": 0.01133081, "auxiliary_loss_mlp": 0.01024516, "balance_loss_clip": 1.04257655, "balance_loss_mlp": 1.01741099, "epoch": 0.8112787831419467, "flos": 24095740616160.0, "grad_norm": 1.760215139737111, "language_loss": 0.746714, "learning_rate": 3.6196900213784496e-07, "loss": 0.76828998, "num_input_tokens_seen": 145630530, "step": 6747, "time_per_iteration": 2.538444757461548 }, { "auxiliary_loss_clip": 0.01156465, "auxiliary_loss_mlp": 0.01030248, "balance_loss_clip": 1.04761541, "balance_loss_mlp": 1.02324438, "epoch": 0.8113990260325858, "flos": 20483378302560.0, "grad_norm": 2.132951711040773, "language_loss": 0.86483181, "learning_rate": 3.6152217474522527e-07, "loss": 0.88669896, "num_input_tokens_seen": 145647345, "step": 6748, "time_per_iteration": 2.5052618980407715 }, { "auxiliary_loss_clip": 0.01155696, "auxiliary_loss_mlp": 0.0102715, "balance_loss_clip": 1.04852784, "balance_loss_mlp": 1.02082014, "epoch": 0.8115192689232249, "flos": 24901145652000.0, "grad_norm": 1.6147146740839622, "language_loss": 0.72619724, "learning_rate": 3.6107559590838975e-07, "loss": 0.74802566, "num_input_tokens_seen": 145666330, "step": 6749, "time_per_iteration": 2.51669979095459 }, { "auxiliary_loss_clip": 0.01093386, "auxiliary_loss_mlp": 0.01028726, "balance_loss_clip": 1.0393126, "balance_loss_mlp": 1.02150476, "epoch": 0.811639511813864, "flos": 24057208106400.0, "grad_norm": 2.159111458131774, "language_loss": 0.66065407, "learning_rate": 3.606292656950822e-07, "loss": 0.68187523, "num_input_tokens_seen": 145684740, "step": 6750, "time_per_iteration": 2.6357221603393555 }, { "auxiliary_loss_clip": 0.01135692, "auxiliary_loss_mlp": 0.01021263, "balance_loss_clip": 1.04153275, "balance_loss_mlp": 1.01418161, "epoch": 0.8117597547045031, "flos": 23185153969440.0, "grad_norm": 1.900649833427253, "language_loss": 0.86622894, "learning_rate": 3.601831841730121e-07, "loss": 0.88779843, "num_input_tokens_seen": 145702660, "step": 6751, "time_per_iteration": 2.5319509506225586 }, { "auxiliary_loss_clip": 0.01156249, "auxiliary_loss_mlp": 0.01025593, "balance_loss_clip": 1.04877996, "balance_loss_mlp": 1.01823783, "epoch": 0.8118799975951422, "flos": 23040254589120.0, "grad_norm": 1.5609445313857448, "language_loss": 0.72329807, "learning_rate": 3.5973735140984916e-07, "loss": 0.74511647, "num_input_tokens_seen": 145722830, "step": 6752, "time_per_iteration": 2.5356736183166504 }, { "auxiliary_loss_clip": 0.01107036, "auxiliary_loss_mlp": 0.00761732, "balance_loss_clip": 1.0411942, "balance_loss_mlp": 1.00050187, "epoch": 0.8120002404857812, "flos": 24639966320160.0, "grad_norm": 3.0924034740570523, "language_loss": 0.79067028, "learning_rate": 3.5929176747322607e-07, "loss": 0.80935794, "num_input_tokens_seen": 145741935, "step": 6753, "time_per_iteration": 2.619138240814209 }, { "auxiliary_loss_clip": 0.01044456, "auxiliary_loss_mlp": 0.01001411, "balance_loss_clip": 1.00621557, "balance_loss_mlp": 1.00011146, "epoch": 0.8121204833764204, "flos": 57415753140000.0, "grad_norm": 0.8094468982681041, "language_loss": 0.56260824, "learning_rate": 3.588464324307372e-07, "loss": 0.58306694, "num_input_tokens_seen": 145805560, "step": 6754, "time_per_iteration": 3.1585421562194824 }, { "auxiliary_loss_clip": 0.01157274, "auxiliary_loss_mlp": 0.01024041, "balance_loss_clip": 1.04559612, "balance_loss_mlp": 1.01691842, "epoch": 0.8122407262670595, "flos": 19464593020320.0, "grad_norm": 2.748632413938099, "language_loss": 0.75266796, "learning_rate": 3.584013463499391e-07, "loss": 0.77448118, "num_input_tokens_seen": 145824180, "step": 6755, "time_per_iteration": 3.219677686691284 }, { "auxiliary_loss_clip": 0.01040548, "auxiliary_loss_mlp": 0.01001593, "balance_loss_clip": 1.00592387, "balance_loss_mlp": 1.0003413, "epoch": 0.8123609691576985, "flos": 56425335867840.0, "grad_norm": 0.7330281671633991, "language_loss": 0.64432752, "learning_rate": 3.579565092983521e-07, "loss": 0.66474891, "num_input_tokens_seen": 145885300, "step": 6756, "time_per_iteration": 2.982158899307251 }, { "auxiliary_loss_clip": 0.01170057, "auxiliary_loss_mlp": 0.01023601, "balance_loss_clip": 1.04856873, "balance_loss_mlp": 1.01639533, "epoch": 0.8124812120483377, "flos": 20631977129760.0, "grad_norm": 2.0429833055382316, "language_loss": 0.83891988, "learning_rate": 3.575119213434565e-07, "loss": 0.86085647, "num_input_tokens_seen": 145903815, "step": 6757, "time_per_iteration": 2.4592700004577637 }, { "auxiliary_loss_clip": 0.01151533, "auxiliary_loss_mlp": 0.01026213, "balance_loss_clip": 1.04641414, "balance_loss_mlp": 1.01869345, "epoch": 0.8126014549389767, "flos": 22492401272160.0, "grad_norm": 1.780869777849447, "language_loss": 0.81476784, "learning_rate": 3.5706758255269765e-07, "loss": 0.83654535, "num_input_tokens_seen": 145922270, "step": 6758, "time_per_iteration": 2.4881176948547363 }, { "auxiliary_loss_clip": 0.01144307, "auxiliary_loss_mlp": 0.010259, "balance_loss_clip": 1.04537678, "balance_loss_mlp": 1.01855087, "epoch": 0.8127216978296158, "flos": 23287965060960.0, "grad_norm": 1.8200582079454177, "language_loss": 0.69901502, "learning_rate": 3.566234929934795e-07, "loss": 0.72071707, "num_input_tokens_seen": 145941470, "step": 6759, "time_per_iteration": 2.5197877883911133 }, { "auxiliary_loss_clip": 0.01155028, "auxiliary_loss_mlp": 0.01025644, "balance_loss_clip": 1.04896188, "balance_loss_mlp": 1.01858377, "epoch": 0.812841940720255, "flos": 25154997924000.0, "grad_norm": 1.4489154846257553, "language_loss": 0.7165814, "learning_rate": 3.561796527331706e-07, "loss": 0.73838806, "num_input_tokens_seen": 145963145, "step": 6760, "time_per_iteration": 2.530008554458618 }, { "auxiliary_loss_clip": 0.01128234, "auxiliary_loss_mlp": 0.01027643, "balance_loss_clip": 1.04378569, "balance_loss_mlp": 1.02029347, "epoch": 0.812962183610894, "flos": 26648450535360.0, "grad_norm": 1.8190478716451055, "language_loss": 0.77597725, "learning_rate": 3.5573606183910163e-07, "loss": 0.79753602, "num_input_tokens_seen": 145983150, "step": 6761, "time_per_iteration": 2.5864951610565186 }, { "auxiliary_loss_clip": 0.01160815, "auxiliary_loss_mlp": 0.01025513, "balance_loss_clip": 1.0451808, "balance_loss_mlp": 1.01859033, "epoch": 0.8130824265015331, "flos": 24966968663040.0, "grad_norm": 1.7142853612042948, "language_loss": 0.78477907, "learning_rate": 3.5529272037856493e-07, "loss": 0.80664229, "num_input_tokens_seen": 146001365, "step": 6762, "time_per_iteration": 2.5051732063293457 }, { "auxiliary_loss_clip": 0.01014766, "auxiliary_loss_mlp": 0.0100231, "balance_loss_clip": 1.00579011, "balance_loss_mlp": 1.00096345, "epoch": 0.8132026693921722, "flos": 67622929286880.0, "grad_norm": 0.7082104852702995, "language_loss": 0.53941196, "learning_rate": 3.548496284188149e-07, "loss": 0.55958271, "num_input_tokens_seen": 146061570, "step": 6763, "time_per_iteration": 3.2356386184692383 }, { "auxiliary_loss_clip": 0.01106894, "auxiliary_loss_mlp": 0.01028195, "balance_loss_clip": 1.04407501, "balance_loss_mlp": 1.0214951, "epoch": 0.8133229122828113, "flos": 19495151964960.0, "grad_norm": 1.8318909792852636, "language_loss": 0.79124224, "learning_rate": 3.544067860270681e-07, "loss": 0.8125931, "num_input_tokens_seen": 146079145, "step": 6764, "time_per_iteration": 2.560074806213379 }, { "auxiliary_loss_clip": 0.01129362, "auxiliary_loss_mlp": 0.01025685, "balance_loss_clip": 1.04442751, "balance_loss_mlp": 1.01872075, "epoch": 0.8134431551734503, "flos": 20668139120160.0, "grad_norm": 1.9087524756405614, "language_loss": 0.70973426, "learning_rate": 3.539641932705029e-07, "loss": 0.73128474, "num_input_tokens_seen": 146097625, "step": 6765, "time_per_iteration": 2.5530991554260254 }, { "auxiliary_loss_clip": 0.01174171, "auxiliary_loss_mlp": 0.01026826, "balance_loss_clip": 1.04914522, "balance_loss_mlp": 1.01927972, "epoch": 0.8135633980640895, "flos": 21507335627040.0, "grad_norm": 3.2850198237631014, "language_loss": 0.77205443, "learning_rate": 3.53521850216262e-07, "loss": 0.7940644, "num_input_tokens_seen": 146117195, "step": 6766, "time_per_iteration": 3.2643659114837646 }, { "auxiliary_loss_clip": 0.01169886, "auxiliary_loss_mlp": 0.01027263, "balance_loss_clip": 1.04926836, "balance_loss_mlp": 1.02004457, "epoch": 0.8136836409547286, "flos": 20554445189760.0, "grad_norm": 1.8821881139436334, "language_loss": 0.76294553, "learning_rate": 3.530797569314461e-07, "loss": 0.78491706, "num_input_tokens_seen": 146136220, "step": 6767, "time_per_iteration": 2.454932451248169 }, { "auxiliary_loss_clip": 0.01169454, "auxiliary_loss_mlp": 0.01025815, "balance_loss_clip": 1.04848528, "balance_loss_mlp": 1.01850724, "epoch": 0.8138038838453676, "flos": 20299048488480.0, "grad_norm": 2.4225601800601453, "language_loss": 0.77790105, "learning_rate": 3.5263791348312235e-07, "loss": 0.79985374, "num_input_tokens_seen": 146155415, "step": 6768, "time_per_iteration": 3.246840238571167 }, { "auxiliary_loss_clip": 0.01138509, "auxiliary_loss_mlp": 0.01021856, "balance_loss_clip": 1.04409003, "balance_loss_mlp": 1.01470935, "epoch": 0.8139241267360068, "flos": 29789844966720.0, "grad_norm": 2.0396524213721534, "language_loss": 0.70577836, "learning_rate": 3.521963199383171e-07, "loss": 0.727382, "num_input_tokens_seen": 146178370, "step": 6769, "time_per_iteration": 2.5742225646972656 }, { "auxiliary_loss_clip": 0.01114628, "auxiliary_loss_mlp": 0.01024644, "balance_loss_clip": 1.04302144, "balance_loss_mlp": 1.01733041, "epoch": 0.8140443696266458, "flos": 19713273250080.0, "grad_norm": 2.053870711855918, "language_loss": 0.76988053, "learning_rate": 3.517549763640197e-07, "loss": 0.79127324, "num_input_tokens_seen": 146196010, "step": 6770, "time_per_iteration": 2.5817041397094727 }, { "auxiliary_loss_clip": 0.01152343, "auxiliary_loss_mlp": 0.00761676, "balance_loss_clip": 1.04896116, "balance_loss_mlp": 1.00052953, "epoch": 0.8141646125172849, "flos": 27160572865440.0, "grad_norm": 1.8668248927579918, "language_loss": 0.71160442, "learning_rate": 3.513138828271829e-07, "loss": 0.73074466, "num_input_tokens_seen": 146215880, "step": 6771, "time_per_iteration": 3.3035244941711426 }, { "auxiliary_loss_clip": 0.01124819, "auxiliary_loss_mlp": 0.01028965, "balance_loss_clip": 1.04475045, "balance_loss_mlp": 1.02235174, "epoch": 0.8142848554079241, "flos": 39673107629280.0, "grad_norm": 1.9313610842721558, "language_loss": 0.69853938, "learning_rate": 3.508730393947179e-07, "loss": 0.72007722, "num_input_tokens_seen": 146239135, "step": 6772, "time_per_iteration": 2.6860013008117676 }, { "auxiliary_loss_clip": 0.01126974, "auxiliary_loss_mlp": 0.01024957, "balance_loss_clip": 1.04403007, "balance_loss_mlp": 1.01804876, "epoch": 0.8144050982985631, "flos": 22237291906560.0, "grad_norm": 1.7555554743855137, "language_loss": 0.72061431, "learning_rate": 3.504324461335024e-07, "loss": 0.74213362, "num_input_tokens_seen": 146259245, "step": 6773, "time_per_iteration": 2.5612192153930664 }, { "auxiliary_loss_clip": 0.01106583, "auxiliary_loss_mlp": 0.0103162, "balance_loss_clip": 1.04057443, "balance_loss_mlp": 1.02393699, "epoch": 0.8145253411892022, "flos": 23038243239360.0, "grad_norm": 2.175931156827821, "language_loss": 0.88039964, "learning_rate": 3.499921031103732e-07, "loss": 0.90178168, "num_input_tokens_seen": 146280015, "step": 6774, "time_per_iteration": 2.6177802085876465 }, { "auxiliary_loss_clip": 0.01130505, "auxiliary_loss_mlp": 0.01026172, "balance_loss_clip": 1.04082954, "balance_loss_mlp": 1.0189774, "epoch": 0.8146455840798413, "flos": 24827672328480.0, "grad_norm": 1.6483187216276487, "language_loss": 0.78363097, "learning_rate": 3.4955201039212987e-07, "loss": 0.80519772, "num_input_tokens_seen": 146300935, "step": 6775, "time_per_iteration": 2.578719139099121 }, { "auxiliary_loss_clip": 0.011575, "auxiliary_loss_mlp": 0.01027525, "balance_loss_clip": 1.04886723, "balance_loss_mlp": 1.02034247, "epoch": 0.8147658269704804, "flos": 19974524415840.0, "grad_norm": 2.051280164700113, "language_loss": 0.65510768, "learning_rate": 3.4911216804553465e-07, "loss": 0.67695791, "num_input_tokens_seen": 146319835, "step": 6776, "time_per_iteration": 2.4791202545166016 }, { "auxiliary_loss_clip": 0.01141629, "auxiliary_loss_mlp": 0.0102934, "balance_loss_clip": 1.0451405, "balance_loss_mlp": 1.02159429, "epoch": 0.8148860698611194, "flos": 21178034598720.0, "grad_norm": 2.0158344204982863, "language_loss": 0.69893205, "learning_rate": 3.4867257613731017e-07, "loss": 0.72064167, "num_input_tokens_seen": 146339030, "step": 6777, "time_per_iteration": 2.5274264812469482 }, { "auxiliary_loss_clip": 0.01143025, "auxiliary_loss_mlp": 0.01025992, "balance_loss_clip": 1.0456996, "balance_loss_mlp": 1.01933384, "epoch": 0.8150063127517585, "flos": 19606906379520.0, "grad_norm": 2.071266593116309, "language_loss": 0.85659683, "learning_rate": 3.4823323473414343e-07, "loss": 0.87828702, "num_input_tokens_seen": 146358550, "step": 6778, "time_per_iteration": 2.5153019428253174 }, { "auxiliary_loss_clip": 0.01131151, "auxiliary_loss_mlp": 0.0102771, "balance_loss_clip": 1.04484332, "balance_loss_mlp": 1.01958001, "epoch": 0.8151265556423977, "flos": 22638378161280.0, "grad_norm": 1.766469301290576, "language_loss": 0.76048738, "learning_rate": 3.477941439026812e-07, "loss": 0.78207594, "num_input_tokens_seen": 146376770, "step": 6779, "time_per_iteration": 2.5578603744506836 }, { "auxiliary_loss_clip": 0.01137475, "auxiliary_loss_mlp": 0.01023055, "balance_loss_clip": 1.04561019, "balance_loss_mlp": 1.01634049, "epoch": 0.8152467985330367, "flos": 17968051550400.0, "grad_norm": 3.415129074318693, "language_loss": 0.73007286, "learning_rate": 3.473553037095349e-07, "loss": 0.75167817, "num_input_tokens_seen": 146395795, "step": 6780, "time_per_iteration": 2.496469736099243 }, { "auxiliary_loss_clip": 0.01134257, "auxiliary_loss_mlp": 0.0102369, "balance_loss_clip": 1.04230452, "balance_loss_mlp": 1.01772976, "epoch": 0.8153670414236758, "flos": 24969015929760.0, "grad_norm": 1.8868536460873762, "language_loss": 0.83363575, "learning_rate": 3.469167142212743e-07, "loss": 0.85521525, "num_input_tokens_seen": 146417640, "step": 6781, "time_per_iteration": 3.2791285514831543 }, { "auxiliary_loss_clip": 0.01156469, "auxiliary_loss_mlp": 0.01029122, "balance_loss_clip": 1.04711461, "balance_loss_mlp": 1.0217427, "epoch": 0.8154872843143149, "flos": 31066074216960.0, "grad_norm": 2.6971326950153376, "language_loss": 0.62946117, "learning_rate": 3.4647837550443337e-07, "loss": 0.651317, "num_input_tokens_seen": 146436205, "step": 6782, "time_per_iteration": 2.5589375495910645 }, { "auxiliary_loss_clip": 0.01128343, "auxiliary_loss_mlp": 0.01027044, "balance_loss_clip": 1.04438317, "balance_loss_mlp": 1.02026057, "epoch": 0.815607527204954, "flos": 19391658451200.0, "grad_norm": 2.0421063390843486, "language_loss": 0.74870563, "learning_rate": 3.460402876255086e-07, "loss": 0.7702595, "num_input_tokens_seen": 146453595, "step": 6783, "time_per_iteration": 2.5600531101226807 }, { "auxiliary_loss_clip": 0.01158233, "auxiliary_loss_mlp": 0.01027504, "balance_loss_clip": 1.04694295, "balance_loss_mlp": 1.01979065, "epoch": 0.815727770095593, "flos": 26140422738720.0, "grad_norm": 2.339392648335478, "language_loss": 0.71999037, "learning_rate": 3.456024506509574e-07, "loss": 0.74184775, "num_input_tokens_seen": 146474515, "step": 6784, "time_per_iteration": 2.561579465866089 }, { "auxiliary_loss_clip": 0.01159168, "auxiliary_loss_mlp": 0.00761967, "balance_loss_clip": 1.05096698, "balance_loss_mlp": 1.00052905, "epoch": 0.8158480129862322, "flos": 25337531890080.0, "grad_norm": 1.5241741731209464, "language_loss": 0.7387259, "learning_rate": 3.4516486464719873e-07, "loss": 0.75793731, "num_input_tokens_seen": 146493905, "step": 6785, "time_per_iteration": 2.5486209392547607 }, { "auxiliary_loss_clip": 0.01107255, "auxiliary_loss_mlp": 0.01024794, "balance_loss_clip": 1.04039824, "balance_loss_mlp": 1.01734376, "epoch": 0.8159682558768713, "flos": 34423650417600.0, "grad_norm": 1.88488428410064, "language_loss": 0.62192798, "learning_rate": 3.4472752968061445e-07, "loss": 0.64324844, "num_input_tokens_seen": 146518335, "step": 6786, "time_per_iteration": 2.6880874633789062 }, { "auxiliary_loss_clip": 0.01155173, "auxiliary_loss_mlp": 0.01027353, "balance_loss_clip": 1.04551327, "balance_loss_mlp": 1.02068043, "epoch": 0.8160884987675103, "flos": 18653225769120.0, "grad_norm": 1.918149432404924, "language_loss": 0.73607767, "learning_rate": 3.442904458175475e-07, "loss": 0.75790298, "num_input_tokens_seen": 146535655, "step": 6787, "time_per_iteration": 2.476292848587036 }, { "auxiliary_loss_clip": 0.01151465, "auxiliary_loss_mlp": 0.01025353, "balance_loss_clip": 1.04454041, "balance_loss_mlp": 1.01802814, "epoch": 0.8162087416581495, "flos": 31430531560800.0, "grad_norm": 1.5377381866976836, "language_loss": 0.75948012, "learning_rate": 3.438536131243044e-07, "loss": 0.78124833, "num_input_tokens_seen": 146556815, "step": 6788, "time_per_iteration": 2.5788843631744385 }, { "auxiliary_loss_clip": 0.01145462, "auxiliary_loss_mlp": 0.0102406, "balance_loss_clip": 1.04570901, "balance_loss_mlp": 1.0167315, "epoch": 0.8163289845487885, "flos": 37593915696480.0, "grad_norm": 5.847407924638171, "language_loss": 0.62013102, "learning_rate": 3.434170316671503e-07, "loss": 0.64182621, "num_input_tokens_seen": 146581845, "step": 6789, "time_per_iteration": 2.655043363571167 }, { "auxiliary_loss_clip": 0.01123228, "auxiliary_loss_mlp": 0.01023149, "balance_loss_clip": 1.04680729, "balance_loss_mlp": 1.0165298, "epoch": 0.8164492274394276, "flos": 13953992393760.0, "grad_norm": 2.574611816697075, "language_loss": 0.90118724, "learning_rate": 3.4298070151231583e-07, "loss": 0.92265105, "num_input_tokens_seen": 146597245, "step": 6790, "time_per_iteration": 2.5366852283477783 }, { "auxiliary_loss_clip": 0.01145293, "auxiliary_loss_mlp": 0.0102424, "balance_loss_clip": 1.04547262, "balance_loss_mlp": 1.01709986, "epoch": 0.8165694703300668, "flos": 28986559031520.0, "grad_norm": 1.8293384486137156, "language_loss": 0.59584248, "learning_rate": 3.425446227259916e-07, "loss": 0.61753786, "num_input_tokens_seen": 146618210, "step": 6791, "time_per_iteration": 3.3536031246185303 }, { "auxiliary_loss_clip": 0.0114017, "auxiliary_loss_mlp": 0.01022847, "balance_loss_clip": 1.04348302, "balance_loss_mlp": 1.0161829, "epoch": 0.8166897132207058, "flos": 25118369013120.0, "grad_norm": 2.2591125489613213, "language_loss": 0.82139903, "learning_rate": 3.421087953743296e-07, "loss": 0.8430292, "num_input_tokens_seen": 146637975, "step": 6792, "time_per_iteration": 2.548323392868042 }, { "auxiliary_loss_clip": 0.01154357, "auxiliary_loss_mlp": 0.01026336, "balance_loss_clip": 1.04382205, "balance_loss_mlp": 1.0192436, "epoch": 0.8168099561113449, "flos": 23148596892480.0, "grad_norm": 2.0347035687735455, "language_loss": 0.80204666, "learning_rate": 3.416732195234464e-07, "loss": 0.82385361, "num_input_tokens_seen": 146658030, "step": 6793, "time_per_iteration": 2.48720121383667 }, { "auxiliary_loss_clip": 0.01157762, "auxiliary_loss_mlp": 0.01022523, "balance_loss_clip": 1.04621065, "balance_loss_mlp": 1.01568079, "epoch": 0.816930199001984, "flos": 18407670314880.0, "grad_norm": 1.50479304032651, "language_loss": 0.79298592, "learning_rate": 3.4123789523941613e-07, "loss": 0.81478876, "num_input_tokens_seen": 146677855, "step": 6794, "time_per_iteration": 3.2886016368865967 }, { "auxiliary_loss_clip": 0.01147941, "auxiliary_loss_mlp": 0.0101892, "balance_loss_clip": 1.04321456, "balance_loss_mlp": 1.01265216, "epoch": 0.8170504418926231, "flos": 21251328337440.0, "grad_norm": 1.5423416911582666, "language_loss": 0.63480616, "learning_rate": 3.4080282258827884e-07, "loss": 0.65647471, "num_input_tokens_seen": 146696230, "step": 6795, "time_per_iteration": 2.482235908508301 }, { "auxiliary_loss_clip": 0.01159309, "auxiliary_loss_mlp": 0.01029066, "balance_loss_clip": 1.04807711, "balance_loss_mlp": 1.02236938, "epoch": 0.8171706847832622, "flos": 19099237752480.0, "grad_norm": 1.961571429961644, "language_loss": 0.72538054, "learning_rate": 3.403680016360342e-07, "loss": 0.74726427, "num_input_tokens_seen": 146714835, "step": 6796, "time_per_iteration": 2.4849960803985596 }, { "auxiliary_loss_clip": 0.01149473, "auxiliary_loss_mlp": 0.01027227, "balance_loss_clip": 1.0469749, "balance_loss_mlp": 1.02012539, "epoch": 0.8172909276739013, "flos": 21470132044800.0, "grad_norm": 2.121013130898158, "language_loss": 0.67731178, "learning_rate": 3.3993343244864403e-07, "loss": 0.69907874, "num_input_tokens_seen": 146734425, "step": 6797, "time_per_iteration": 3.22629976272583 }, { "auxiliary_loss_clip": 0.01153599, "auxiliary_loss_mlp": 0.01024538, "balance_loss_clip": 1.04746652, "balance_loss_mlp": 1.01804662, "epoch": 0.8174111705645404, "flos": 27599796543360.0, "grad_norm": 2.680722241235299, "language_loss": 0.72768366, "learning_rate": 3.394991150920323e-07, "loss": 0.74946511, "num_input_tokens_seen": 146757545, "step": 6798, "time_per_iteration": 2.541679859161377 }, { "auxiliary_loss_clip": 0.01115622, "auxiliary_loss_mlp": 0.00762497, "balance_loss_clip": 1.04358459, "balance_loss_mlp": 1.00053525, "epoch": 0.8175314134551794, "flos": 14064597465600.0, "grad_norm": 1.8493882962571424, "language_loss": 0.74173522, "learning_rate": 3.3906504963208396e-07, "loss": 0.76051641, "num_input_tokens_seen": 146774240, "step": 6799, "time_per_iteration": 2.538292646408081 }, { "auxiliary_loss_clip": 0.011084, "auxiliary_loss_mlp": 0.01027472, "balance_loss_clip": 1.04365206, "balance_loss_mlp": 1.02009249, "epoch": 0.8176516563458186, "flos": 22708080204000.0, "grad_norm": 1.9608114576545845, "language_loss": 0.66507643, "learning_rate": 3.3863123613464774e-07, "loss": 0.68643516, "num_input_tokens_seen": 146793140, "step": 6800, "time_per_iteration": 2.5769925117492676 }, { "auxiliary_loss_clip": 0.01140713, "auxiliary_loss_mlp": 0.01024885, "balance_loss_clip": 1.04079318, "balance_loss_mlp": 1.01819706, "epoch": 0.8177718992364577, "flos": 21945409962240.0, "grad_norm": 1.6211983272642092, "language_loss": 0.75069141, "learning_rate": 3.381976746655317e-07, "loss": 0.77234745, "num_input_tokens_seen": 146812895, "step": 6801, "time_per_iteration": 2.518495559692383 }, { "auxiliary_loss_clip": 0.0110638, "auxiliary_loss_mlp": 0.01026402, "balance_loss_clip": 1.04504812, "balance_loss_mlp": 1.01958597, "epoch": 0.8178921421270967, "flos": 22017446607360.0, "grad_norm": 1.936859749154302, "language_loss": 0.67124361, "learning_rate": 3.3776436529050756e-07, "loss": 0.6925714, "num_input_tokens_seen": 146832445, "step": 6802, "time_per_iteration": 2.5711967945098877 }, { "auxiliary_loss_clip": 0.01165699, "auxiliary_loss_mlp": 0.01026613, "balance_loss_clip": 1.04630578, "balance_loss_mlp": 1.01987767, "epoch": 0.8180123850177359, "flos": 33183116237280.0, "grad_norm": 1.7681803598331234, "language_loss": 0.72468019, "learning_rate": 3.373313080753073e-07, "loss": 0.74660331, "num_input_tokens_seen": 146856505, "step": 6803, "time_per_iteration": 2.5734736919403076 }, { "auxiliary_loss_clip": 0.01147244, "auxiliary_loss_mlp": 0.01024465, "balance_loss_clip": 1.04274583, "balance_loss_mlp": 1.01790857, "epoch": 0.8181326279083749, "flos": 22091171349600.0, "grad_norm": 3.2377552252115396, "language_loss": 0.77345914, "learning_rate": 3.3689850308562527e-07, "loss": 0.79517621, "num_input_tokens_seen": 146876950, "step": 6804, "time_per_iteration": 2.494135856628418 }, { "auxiliary_loss_clip": 0.0110657, "auxiliary_loss_mlp": 0.01025342, "balance_loss_clip": 1.04581761, "balance_loss_mlp": 1.01848793, "epoch": 0.818252870799014, "flos": 15705750980160.0, "grad_norm": 2.9889425127614437, "language_loss": 0.77416599, "learning_rate": 3.364659503871183e-07, "loss": 0.79548514, "num_input_tokens_seen": 146894885, "step": 6805, "time_per_iteration": 2.5666370391845703 }, { "auxiliary_loss_clip": 0.01120779, "auxiliary_loss_mlp": 0.01022884, "balance_loss_clip": 1.04069185, "balance_loss_mlp": 1.01665258, "epoch": 0.8183731136896532, "flos": 18770690980320.0, "grad_norm": 1.8755781559636917, "language_loss": 0.83524418, "learning_rate": 3.3603365004540417e-07, "loss": 0.85668081, "num_input_tokens_seen": 146913180, "step": 6806, "time_per_iteration": 2.548175573348999 }, { "auxiliary_loss_clip": 0.01168839, "auxiliary_loss_mlp": 0.01035474, "balance_loss_clip": 1.04823089, "balance_loss_mlp": 1.02790713, "epoch": 0.8184933565802922, "flos": 26541796329120.0, "grad_norm": 2.516791475744295, "language_loss": 0.76937842, "learning_rate": 3.356016021260624e-07, "loss": 0.79142153, "num_input_tokens_seen": 146933510, "step": 6807, "time_per_iteration": 3.3102474212646484 }, { "auxiliary_loss_clip": 0.01157612, "auxiliary_loss_mlp": 0.01023141, "balance_loss_clip": 1.04808736, "balance_loss_mlp": 1.0161463, "epoch": 0.8186135994709313, "flos": 17530120883040.0, "grad_norm": 4.460254379707718, "language_loss": 0.655595, "learning_rate": 3.35169806694634e-07, "loss": 0.67740256, "num_input_tokens_seen": 146951760, "step": 6808, "time_per_iteration": 2.493436813354492 }, { "auxiliary_loss_clip": 0.01033125, "auxiliary_loss_mlp": 0.01000641, "balance_loss_clip": 1.01041567, "balance_loss_mlp": 0.99955004, "epoch": 0.8187338423615703, "flos": 63480312966720.0, "grad_norm": 0.7212229978020668, "language_loss": 0.60653549, "learning_rate": 3.3473826381662186e-07, "loss": 0.6268732, "num_input_tokens_seen": 147022900, "step": 6809, "time_per_iteration": 3.3051674365997314 }, { "auxiliary_loss_clip": 0.0114991, "auxiliary_loss_mlp": 0.01023678, "balance_loss_clip": 1.04758763, "balance_loss_mlp": 1.01736844, "epoch": 0.8188540852522095, "flos": 17529977215200.0, "grad_norm": 2.005916697883979, "language_loss": 0.81664294, "learning_rate": 3.3430697355749216e-07, "loss": 0.83837879, "num_input_tokens_seen": 147040590, "step": 6810, "time_per_iteration": 2.4462687969207764 }, { "auxiliary_loss_clip": 0.01107912, "auxiliary_loss_mlp": 0.01025691, "balance_loss_clip": 1.04110336, "balance_loss_mlp": 1.01866007, "epoch": 0.8189743281428485, "flos": 14392533649440.0, "grad_norm": 2.1955477966239187, "language_loss": 0.7553525, "learning_rate": 3.3387593598266907e-07, "loss": 0.77668858, "num_input_tokens_seen": 147057200, "step": 6811, "time_per_iteration": 2.54500150680542 }, { "auxiliary_loss_clip": 0.01118021, "auxiliary_loss_mlp": 0.01026427, "balance_loss_clip": 1.04047465, "balance_loss_mlp": 1.0190537, "epoch": 0.8190945710334876, "flos": 25080483008640.0, "grad_norm": 1.8080010716429211, "language_loss": 0.78190213, "learning_rate": 3.3344515115754225e-07, "loss": 0.80334663, "num_input_tokens_seen": 147076180, "step": 6812, "time_per_iteration": 2.578929901123047 }, { "auxiliary_loss_clip": 0.01128572, "auxiliary_loss_mlp": 0.01025289, "balance_loss_clip": 1.04313409, "balance_loss_mlp": 1.01856875, "epoch": 0.8192148139241268, "flos": 21507156042240.0, "grad_norm": 2.606324616411954, "language_loss": 0.80068916, "learning_rate": 3.33014619147461e-07, "loss": 0.82222778, "num_input_tokens_seen": 147094205, "step": 6813, "time_per_iteration": 2.5533065795898438 }, { "auxiliary_loss_clip": 0.01141663, "auxiliary_loss_mlp": 0.01029195, "balance_loss_clip": 1.04691255, "balance_loss_mlp": 1.02171445, "epoch": 0.8193350568147658, "flos": 23952170163360.0, "grad_norm": 1.7750943382172297, "language_loss": 0.7153874, "learning_rate": 3.325843400177362e-07, "loss": 0.73709595, "num_input_tokens_seen": 147115545, "step": 6814, "time_per_iteration": 2.540536403656006 }, { "auxiliary_loss_clip": 0.01157548, "auxiliary_loss_mlp": 0.00762198, "balance_loss_clip": 1.04600775, "balance_loss_mlp": 1.00063646, "epoch": 0.8194552997054049, "flos": 20559473564160.0, "grad_norm": 2.5315007347272216, "language_loss": 0.73676491, "learning_rate": 3.32154313833642e-07, "loss": 0.75596237, "num_input_tokens_seen": 147135700, "step": 6815, "time_per_iteration": 2.5016348361968994 }, { "auxiliary_loss_clip": 0.01169292, "auxiliary_loss_mlp": 0.01027658, "balance_loss_clip": 1.04721141, "balance_loss_mlp": 1.02027869, "epoch": 0.819575542596044, "flos": 26031757182720.0, "grad_norm": 2.3714264360647666, "language_loss": 0.5934757, "learning_rate": 3.3172454066041164e-07, "loss": 0.61544526, "num_input_tokens_seen": 147155205, "step": 6816, "time_per_iteration": 2.4800665378570557 }, { "auxiliary_loss_clip": 0.01098745, "auxiliary_loss_mlp": 0.00761037, "balance_loss_clip": 1.04412961, "balance_loss_mlp": 1.00053966, "epoch": 0.8196957854866831, "flos": 29096948601600.0, "grad_norm": 1.8514552660844832, "language_loss": 0.76139855, "learning_rate": 3.3129502056324234e-07, "loss": 0.7799964, "num_input_tokens_seen": 147176570, "step": 6817, "time_per_iteration": 2.674217462539673 }, { "auxiliary_loss_clip": 0.0100306, "auxiliary_loss_mlp": 0.0100731, "balance_loss_clip": 1.00865054, "balance_loss_mlp": 1.00602853, "epoch": 0.8198160283773221, "flos": 69033641999040.0, "grad_norm": 0.7931442237089144, "language_loss": 0.59760392, "learning_rate": 3.3086575360729165e-07, "loss": 0.61770767, "num_input_tokens_seen": 147234105, "step": 6818, "time_per_iteration": 3.949476718902588 }, { "auxiliary_loss_clip": 0.01140277, "auxiliary_loss_mlp": 0.01024989, "balance_loss_clip": 1.04493785, "balance_loss_mlp": 1.01787806, "epoch": 0.8199362712679613, "flos": 16618061640960.0, "grad_norm": 1.9237768129084265, "language_loss": 0.71408218, "learning_rate": 3.3043673985767906e-07, "loss": 0.73573482, "num_input_tokens_seen": 147253170, "step": 6819, "time_per_iteration": 2.670532703399658 }, { "auxiliary_loss_clip": 0.01113665, "auxiliary_loss_mlp": 0.01027724, "balance_loss_clip": 1.03897488, "balance_loss_mlp": 1.02055311, "epoch": 0.8200565141586004, "flos": 21757668036960.0, "grad_norm": 2.7343257252273103, "language_loss": 0.77788514, "learning_rate": 3.3000797937948564e-07, "loss": 0.799299, "num_input_tokens_seen": 147271465, "step": 6820, "time_per_iteration": 3.63244891166687 }, { "auxiliary_loss_clip": 0.01026568, "auxiliary_loss_mlp": 0.01000945, "balance_loss_clip": 1.00626683, "balance_loss_mlp": 0.99968702, "epoch": 0.8201767570492394, "flos": 69807123245760.0, "grad_norm": 0.9326889854303613, "language_loss": 0.65012652, "learning_rate": 3.295794722377534e-07, "loss": 0.67040169, "num_input_tokens_seen": 147335070, "step": 6821, "time_per_iteration": 3.2350106239318848 }, { "auxiliary_loss_clip": 0.01165073, "auxiliary_loss_mlp": 0.01023234, "balance_loss_clip": 1.04592693, "balance_loss_mlp": 1.01657033, "epoch": 0.8202969999398786, "flos": 23111896147680.0, "grad_norm": 8.579405893907522, "language_loss": 0.79834998, "learning_rate": 3.291512184974876e-07, "loss": 0.82023305, "num_input_tokens_seen": 147355460, "step": 6822, "time_per_iteration": 2.4770689010620117 }, { "auxiliary_loss_clip": 0.01134976, "auxiliary_loss_mlp": 0.01029746, "balance_loss_clip": 1.03999007, "balance_loss_mlp": 1.02194405, "epoch": 0.8204172428305176, "flos": 28220620346400.0, "grad_norm": 1.9047605725270458, "language_loss": 0.66686916, "learning_rate": 3.2872321822365346e-07, "loss": 0.68851638, "num_input_tokens_seen": 147375675, "step": 6823, "time_per_iteration": 3.3467695713043213 }, { "auxiliary_loss_clip": 0.01153092, "auxiliary_loss_mlp": 0.01023225, "balance_loss_clip": 1.04720879, "balance_loss_mlp": 1.01620638, "epoch": 0.8205374857211567, "flos": 20887014661440.0, "grad_norm": 1.9981658063131893, "language_loss": 0.73320502, "learning_rate": 3.282954714811783e-07, "loss": 0.75496823, "num_input_tokens_seen": 147394580, "step": 6824, "time_per_iteration": 2.479489803314209 }, { "auxiliary_loss_clip": 0.01124906, "auxiliary_loss_mlp": 0.01027987, "balance_loss_clip": 1.03880703, "balance_loss_mlp": 1.02075124, "epoch": 0.8206577286117959, "flos": 13152143136960.0, "grad_norm": 2.302752628301598, "language_loss": 0.70717871, "learning_rate": 3.2786797833495093e-07, "loss": 0.72870767, "num_input_tokens_seen": 147409935, "step": 6825, "time_per_iteration": 2.5152816772460938 }, { "auxiliary_loss_clip": 0.01165971, "auxiliary_loss_mlp": 0.01024019, "balance_loss_clip": 1.04620957, "balance_loss_mlp": 1.01760888, "epoch": 0.8207779715024349, "flos": 25265638912800.0, "grad_norm": 1.845099700051377, "language_loss": 0.7282936, "learning_rate": 3.274407388498213e-07, "loss": 0.75019348, "num_input_tokens_seen": 147428065, "step": 6826, "time_per_iteration": 2.505995988845825 }, { "auxiliary_loss_clip": 0.01120364, "auxiliary_loss_mlp": 0.01024775, "balance_loss_clip": 1.0411787, "balance_loss_mlp": 1.01817715, "epoch": 0.820898214393074, "flos": 19610246656800.0, "grad_norm": 3.155384510661952, "language_loss": 0.74634862, "learning_rate": 3.270137530906021e-07, "loss": 0.76780003, "num_input_tokens_seen": 147447300, "step": 6827, "time_per_iteration": 2.5881717205047607 }, { "auxiliary_loss_clip": 0.01105346, "auxiliary_loss_mlp": 0.01021786, "balance_loss_clip": 1.04441333, "balance_loss_mlp": 1.01479459, "epoch": 0.8210184572837131, "flos": 15596618503680.0, "grad_norm": 1.751826777854533, "language_loss": 0.83324438, "learning_rate": 3.265870211220665e-07, "loss": 0.85451567, "num_input_tokens_seen": 147465135, "step": 6828, "time_per_iteration": 2.543896198272705 }, { "auxiliary_loss_clip": 0.01123191, "auxiliary_loss_mlp": 0.01024697, "balance_loss_clip": 1.04386663, "balance_loss_mlp": 1.01710367, "epoch": 0.8211387001743522, "flos": 20813936424480.0, "grad_norm": 1.8997070906258087, "language_loss": 0.81881863, "learning_rate": 3.2616054300894934e-07, "loss": 0.84029752, "num_input_tokens_seen": 147484585, "step": 6829, "time_per_iteration": 2.583735704421997 }, { "auxiliary_loss_clip": 0.01131415, "auxiliary_loss_mlp": 0.01027336, "balance_loss_clip": 1.04462874, "balance_loss_mlp": 1.02006984, "epoch": 0.8212589430649913, "flos": 27704583067680.0, "grad_norm": 2.0679822159032217, "language_loss": 0.84519351, "learning_rate": 3.2573431881594693e-07, "loss": 0.866781, "num_input_tokens_seen": 147504130, "step": 6830, "time_per_iteration": 2.5609548091888428 }, { "auxiliary_loss_clip": 0.01095359, "auxiliary_loss_mlp": 0.01027061, "balance_loss_clip": 1.03798199, "balance_loss_mlp": 1.02001011, "epoch": 0.8213791859556304, "flos": 22455628693440.0, "grad_norm": 1.9084050843854643, "language_loss": 0.6624999, "learning_rate": 3.2530834860771663e-07, "loss": 0.68372416, "num_input_tokens_seen": 147523510, "step": 6831, "time_per_iteration": 2.6016111373901367 }, { "auxiliary_loss_clip": 0.01154487, "auxiliary_loss_mlp": 0.01029847, "balance_loss_clip": 1.04596233, "balance_loss_mlp": 1.0228492, "epoch": 0.8214994288462695, "flos": 16654475050080.0, "grad_norm": 2.1519280795379574, "language_loss": 0.74364805, "learning_rate": 3.248826324488794e-07, "loss": 0.76549143, "num_input_tokens_seen": 147540805, "step": 6832, "time_per_iteration": 2.467503547668457 }, { "auxiliary_loss_clip": 0.01169067, "auxiliary_loss_mlp": 0.01023273, "balance_loss_clip": 1.05034924, "balance_loss_mlp": 1.01661849, "epoch": 0.8216196717369085, "flos": 25221790692960.0, "grad_norm": 1.797472045835808, "language_loss": 0.87758756, "learning_rate": 3.244571704040138e-07, "loss": 0.89951098, "num_input_tokens_seen": 147560965, "step": 6833, "time_per_iteration": 3.249724864959717 }, { "auxiliary_loss_clip": 0.01149207, "auxiliary_loss_mlp": 0.01027477, "balance_loss_clip": 1.04399109, "balance_loss_mlp": 1.02043438, "epoch": 0.8217399146275477, "flos": 25371933949440.0, "grad_norm": 2.1129049516965495, "language_loss": 0.73479605, "learning_rate": 3.2403196253766374e-07, "loss": 0.75656289, "num_input_tokens_seen": 147580045, "step": 6834, "time_per_iteration": 2.522212266921997 }, { "auxiliary_loss_clip": 0.01153811, "auxiliary_loss_mlp": 0.01027212, "balance_loss_clip": 1.04707575, "balance_loss_mlp": 1.01950502, "epoch": 0.8218601575181868, "flos": 25629629336160.0, "grad_norm": 2.527985956137898, "language_loss": 0.79099929, "learning_rate": 3.2360700891433254e-07, "loss": 0.81280947, "num_input_tokens_seen": 147599070, "step": 6835, "time_per_iteration": 2.5077695846557617 }, { "auxiliary_loss_clip": 0.0101923, "auxiliary_loss_mlp": 0.01002677, "balance_loss_clip": 1.00640464, "balance_loss_mlp": 1.00130641, "epoch": 0.8219804004088258, "flos": 67660240620000.0, "grad_norm": 0.7929644319272571, "language_loss": 0.57319427, "learning_rate": 3.231823095984847e-07, "loss": 0.59341335, "num_input_tokens_seen": 147653710, "step": 6836, "time_per_iteration": 3.0720415115356445 }, { "auxiliary_loss_clip": 0.01140997, "auxiliary_loss_mlp": 0.01023567, "balance_loss_clip": 1.04706311, "balance_loss_mlp": 1.016361, "epoch": 0.822100643299465, "flos": 19464269767680.0, "grad_norm": 1.9276759155541334, "language_loss": 0.7596004, "learning_rate": 3.2275786465454814e-07, "loss": 0.78124607, "num_input_tokens_seen": 147670360, "step": 6837, "time_per_iteration": 2.500192880630493 }, { "auxiliary_loss_clip": 0.01123762, "auxiliary_loss_mlp": 0.01025537, "balance_loss_clip": 1.04238784, "balance_loss_mlp": 1.01852453, "epoch": 0.822220886190104, "flos": 24681372186720.0, "grad_norm": 2.915026642360982, "language_loss": 0.75484276, "learning_rate": 3.2233367414690917e-07, "loss": 0.77633572, "num_input_tokens_seen": 147692550, "step": 6838, "time_per_iteration": 2.5786030292510986 }, { "auxiliary_loss_clip": 0.011227, "auxiliary_loss_mlp": 0.01031745, "balance_loss_clip": 1.0418067, "balance_loss_mlp": 1.02435446, "epoch": 0.8223411290807431, "flos": 27819067171200.0, "grad_norm": 2.6631566509410116, "language_loss": 0.84740239, "learning_rate": 3.219097381399183e-07, "loss": 0.86894685, "num_input_tokens_seen": 147709725, "step": 6839, "time_per_iteration": 2.594273328781128 }, { "auxiliary_loss_clip": 0.01148386, "auxiliary_loss_mlp": 0.0102763, "balance_loss_clip": 1.04591882, "balance_loss_mlp": 1.02017093, "epoch": 0.8224613719713821, "flos": 23218550353920.0, "grad_norm": 1.6843160726308053, "language_loss": 0.8093003, "learning_rate": 3.2148605669788584e-07, "loss": 0.83106041, "num_input_tokens_seen": 147729615, "step": 6840, "time_per_iteration": 2.538177490234375 }, { "auxiliary_loss_clip": 0.01145154, "auxiliary_loss_mlp": 0.01022929, "balance_loss_clip": 1.04954326, "balance_loss_mlp": 1.01649785, "epoch": 0.8225816148620213, "flos": 15706253817600.0, "grad_norm": 2.7393184382397533, "language_loss": 0.77313042, "learning_rate": 3.2106262988508405e-07, "loss": 0.79481125, "num_input_tokens_seen": 147747665, "step": 6841, "time_per_iteration": 2.491722345352173 }, { "auxiliary_loss_clip": 0.01142286, "auxiliary_loss_mlp": 0.01024938, "balance_loss_clip": 1.04604125, "balance_loss_mlp": 1.01746392, "epoch": 0.8227018577526604, "flos": 18515114694240.0, "grad_norm": 2.0402783378757507, "language_loss": 0.74138021, "learning_rate": 3.206394577657465e-07, "loss": 0.7630524, "num_input_tokens_seen": 147765445, "step": 6842, "time_per_iteration": 2.5008842945098877 }, { "auxiliary_loss_clip": 0.01159668, "auxiliary_loss_mlp": 0.01029255, "balance_loss_clip": 1.04696059, "balance_loss_mlp": 1.02124977, "epoch": 0.8228221006432994, "flos": 22236789069120.0, "grad_norm": 2.3313783573374987, "language_loss": 0.72481918, "learning_rate": 3.202165404040675e-07, "loss": 0.74670845, "num_input_tokens_seen": 147783365, "step": 6843, "time_per_iteration": 2.487236499786377 }, { "auxiliary_loss_clip": 0.01099882, "auxiliary_loss_mlp": 0.01025488, "balance_loss_clip": 1.04290438, "balance_loss_mlp": 1.01840687, "epoch": 0.8229423435339386, "flos": 24097536464160.0, "grad_norm": 2.0133993099237237, "language_loss": 0.74905717, "learning_rate": 3.1979387786420396e-07, "loss": 0.77031082, "num_input_tokens_seen": 147803605, "step": 6844, "time_per_iteration": 3.6284875869750977 }, { "auxiliary_loss_clip": 0.01141191, "auxiliary_loss_mlp": 0.01020202, "balance_loss_clip": 1.04266787, "balance_loss_mlp": 1.01327038, "epoch": 0.8230625864245776, "flos": 23878553172000.0, "grad_norm": 2.526982134167111, "language_loss": 0.82175618, "learning_rate": 3.1937147021027346e-07, "loss": 0.84337008, "num_input_tokens_seen": 147822060, "step": 6845, "time_per_iteration": 2.541440486907959 }, { "auxiliary_loss_clip": 0.0115436, "auxiliary_loss_mlp": 0.01021825, "balance_loss_clip": 1.04838693, "balance_loss_mlp": 1.01505136, "epoch": 0.8231828293152167, "flos": 16581109477440.0, "grad_norm": 2.47069921880133, "language_loss": 0.76450038, "learning_rate": 3.189493175063547e-07, "loss": 0.78626221, "num_input_tokens_seen": 147839295, "step": 6846, "time_per_iteration": 3.2438442707061768 }, { "auxiliary_loss_clip": 0.01141622, "auxiliary_loss_mlp": 0.0102496, "balance_loss_clip": 1.04593158, "balance_loss_mlp": 1.01798916, "epoch": 0.8233030722058559, "flos": 18880074875520.0, "grad_norm": 1.7267929170981973, "language_loss": 0.6732502, "learning_rate": 3.1852741981648776e-07, "loss": 0.69491607, "num_input_tokens_seen": 147857945, "step": 6847, "time_per_iteration": 2.500100612640381 }, { "auxiliary_loss_clip": 0.01117193, "auxiliary_loss_mlp": 0.01031147, "balance_loss_clip": 1.04351497, "balance_loss_mlp": 1.02402997, "epoch": 0.8234233150964949, "flos": 28439028967200.0, "grad_norm": 5.76926832585386, "language_loss": 0.69912642, "learning_rate": 3.1810577720467404e-07, "loss": 0.72060984, "num_input_tokens_seen": 147879675, "step": 6848, "time_per_iteration": 3.328939199447632 }, { "auxiliary_loss_clip": 0.01144795, "auxiliary_loss_mlp": 0.01033965, "balance_loss_clip": 1.04555917, "balance_loss_mlp": 1.0264672, "epoch": 0.823543557987134, "flos": 33765946284960.0, "grad_norm": 1.5740710548017167, "language_loss": 0.56568247, "learning_rate": 3.176843897348769e-07, "loss": 0.58747011, "num_input_tokens_seen": 147902870, "step": 6849, "time_per_iteration": 2.6383917331695557 }, { "auxiliary_loss_clip": 0.01136589, "auxiliary_loss_mlp": 0.01025711, "balance_loss_clip": 1.04511881, "balance_loss_mlp": 1.01850748, "epoch": 0.8236638008777731, "flos": 17092369800480.0, "grad_norm": 2.8076253162064853, "language_loss": 0.75722659, "learning_rate": 3.1726325747102034e-07, "loss": 0.7788496, "num_input_tokens_seen": 147921245, "step": 6850, "time_per_iteration": 2.480943202972412 }, { "auxiliary_loss_clip": 0.01105272, "auxiliary_loss_mlp": 0.01027292, "balance_loss_clip": 1.03814423, "balance_loss_mlp": 1.02074456, "epoch": 0.8237840437684122, "flos": 61639989545280.0, "grad_norm": 1.5336092887585686, "language_loss": 0.64309353, "learning_rate": 3.1684238047698974e-07, "loss": 0.66441917, "num_input_tokens_seen": 147949515, "step": 6851, "time_per_iteration": 2.9708166122436523 }, { "auxiliary_loss_clip": 0.01142935, "auxiliary_loss_mlp": 0.01029481, "balance_loss_clip": 1.04473722, "balance_loss_mlp": 1.02217913, "epoch": 0.8239042866590512, "flos": 27309028024800.0, "grad_norm": 2.099881748433975, "language_loss": 0.53093529, "learning_rate": 3.1642175881663155e-07, "loss": 0.55265945, "num_input_tokens_seen": 147969245, "step": 6852, "time_per_iteration": 2.5494678020477295 }, { "auxiliary_loss_clip": 0.01166929, "auxiliary_loss_mlp": 0.01026183, "balance_loss_clip": 1.04604304, "balance_loss_mlp": 1.01942647, "epoch": 0.8240245295496904, "flos": 21726354836160.0, "grad_norm": 2.2460185979810356, "language_loss": 0.83917749, "learning_rate": 3.160013925537537e-07, "loss": 0.86110866, "num_input_tokens_seen": 147990080, "step": 6853, "time_per_iteration": 2.459442377090454 }, { "auxiliary_loss_clip": 0.01125654, "auxiliary_loss_mlp": 0.01024136, "balance_loss_clip": 1.04421508, "balance_loss_mlp": 1.01684916, "epoch": 0.8241447724403295, "flos": 20009321561760.0, "grad_norm": 1.7956181108843592, "language_loss": 0.75267744, "learning_rate": 3.155812817521266e-07, "loss": 0.77417529, "num_input_tokens_seen": 148010455, "step": 6854, "time_per_iteration": 2.549028158187866 }, { "auxiliary_loss_clip": 0.01146795, "auxiliary_loss_mlp": 0.01030342, "balance_loss_clip": 1.04962897, "balance_loss_mlp": 1.02301073, "epoch": 0.8242650153309685, "flos": 22272986976480.0, "grad_norm": 2.512719223940755, "language_loss": 0.77738053, "learning_rate": 3.151614264754787e-07, "loss": 0.7991519, "num_input_tokens_seen": 148028400, "step": 6855, "time_per_iteration": 2.532679319381714 }, { "auxiliary_loss_clip": 0.01166762, "auxiliary_loss_mlp": 0.01028604, "balance_loss_clip": 1.04437673, "balance_loss_mlp": 1.02124918, "epoch": 0.8243852582216077, "flos": 22309975056960.0, "grad_norm": 2.2655277294312723, "language_loss": 0.79173595, "learning_rate": 3.147418267875035e-07, "loss": 0.81368959, "num_input_tokens_seen": 148046530, "step": 6856, "time_per_iteration": 2.437441825866699 }, { "auxiliary_loss_clip": 0.01093721, "auxiliary_loss_mlp": 0.00762453, "balance_loss_clip": 1.03770328, "balance_loss_mlp": 1.00054383, "epoch": 0.8245055011122467, "flos": 24645425698080.0, "grad_norm": 2.041541335089569, "language_loss": 0.65229172, "learning_rate": 3.1432248275185315e-07, "loss": 0.6708535, "num_input_tokens_seen": 148067040, "step": 6857, "time_per_iteration": 2.6518216133117676 }, { "auxiliary_loss_clip": 0.0115032, "auxiliary_loss_mlp": 0.01027575, "balance_loss_clip": 1.04451752, "balance_loss_mlp": 1.02030349, "epoch": 0.8246257440028858, "flos": 17487278338080.0, "grad_norm": 2.049672461404937, "language_loss": 0.76711202, "learning_rate": 3.139033944321412e-07, "loss": 0.78889096, "num_input_tokens_seen": 148084400, "step": 6858, "time_per_iteration": 2.456519842147827 }, { "auxiliary_loss_clip": 0.01158071, "auxiliary_loss_mlp": 0.01026001, "balance_loss_clip": 1.04646754, "balance_loss_mlp": 1.01838899, "epoch": 0.824745986893525, "flos": 25010134460640.0, "grad_norm": 2.040261182569947, "language_loss": 0.79092443, "learning_rate": 3.1348456189194507e-07, "loss": 0.81276512, "num_input_tokens_seen": 148104860, "step": 6859, "time_per_iteration": 3.27828049659729 }, { "auxiliary_loss_clip": 0.01116749, "auxiliary_loss_mlp": 0.01023833, "balance_loss_clip": 1.04100871, "balance_loss_mlp": 1.01645398, "epoch": 0.824866229784164, "flos": 18772702330080.0, "grad_norm": 1.5831321164269916, "language_loss": 0.82899654, "learning_rate": 3.1306598519479876e-07, "loss": 0.85040236, "num_input_tokens_seen": 148124680, "step": 6860, "time_per_iteration": 2.5636606216430664 }, { "auxiliary_loss_clip": 0.01138284, "auxiliary_loss_mlp": 0.01022448, "balance_loss_clip": 1.0466423, "balance_loss_mlp": 1.01551592, "epoch": 0.8249864726748031, "flos": 23842175679840.0, "grad_norm": 1.6475328058157024, "language_loss": 0.78264105, "learning_rate": 3.1264766440420177e-07, "loss": 0.80424833, "num_input_tokens_seen": 148147150, "step": 6861, "time_per_iteration": 2.5647552013397217 }, { "auxiliary_loss_clip": 0.01152407, "auxiliary_loss_mlp": 0.01025288, "balance_loss_clip": 1.04766679, "balance_loss_mlp": 1.01853526, "epoch": 0.8251067155654422, "flos": 20303107104960.0, "grad_norm": 2.102518838809454, "language_loss": 0.69227064, "learning_rate": 3.122295995836124e-07, "loss": 0.71404761, "num_input_tokens_seen": 148167020, "step": 6862, "time_per_iteration": 2.478348970413208 }, { "auxiliary_loss_clip": 0.01156995, "auxiliary_loss_mlp": 0.01026011, "balance_loss_clip": 1.04413998, "balance_loss_mlp": 1.01868892, "epoch": 0.8252269584560813, "flos": 25009703457120.0, "grad_norm": 1.7779074255426384, "language_loss": 0.77352411, "learning_rate": 3.118117907964508e-07, "loss": 0.79535419, "num_input_tokens_seen": 148188965, "step": 6863, "time_per_iteration": 2.5430917739868164 }, { "auxiliary_loss_clip": 0.01131759, "auxiliary_loss_mlp": 0.01025001, "balance_loss_clip": 1.04267108, "balance_loss_mlp": 1.01830137, "epoch": 0.8253472013467203, "flos": 17128567707840.0, "grad_norm": 2.036203654098269, "language_loss": 0.80248165, "learning_rate": 3.1139423810609856e-07, "loss": 0.82404923, "num_input_tokens_seen": 148205660, "step": 6864, "time_per_iteration": 2.553483724594116 }, { "auxiliary_loss_clip": 0.01164761, "auxiliary_loss_mlp": 0.01026922, "balance_loss_clip": 1.04416382, "balance_loss_mlp": 1.01980853, "epoch": 0.8254674442373595, "flos": 22414797498240.0, "grad_norm": 2.043870837156182, "language_loss": 0.75453877, "learning_rate": 3.1097694157589714e-07, "loss": 0.77645564, "num_input_tokens_seen": 148225545, "step": 6865, "time_per_iteration": 2.4805188179016113 }, { "auxiliary_loss_clip": 0.01154408, "auxiliary_loss_mlp": 0.01028089, "balance_loss_clip": 1.04904258, "balance_loss_mlp": 1.02032876, "epoch": 0.8255876871279986, "flos": 24786769299360.0, "grad_norm": 7.374972667490136, "language_loss": 0.7621007, "learning_rate": 3.105599012691511e-07, "loss": 0.78392565, "num_input_tokens_seen": 148243975, "step": 6866, "time_per_iteration": 2.5108442306518555 }, { "auxiliary_loss_clip": 0.01152564, "auxiliary_loss_mlp": 0.01021677, "balance_loss_clip": 1.04713225, "balance_loss_mlp": 1.01452994, "epoch": 0.8257079300186376, "flos": 27455435917440.0, "grad_norm": 1.6552922581834848, "language_loss": 0.82324553, "learning_rate": 3.101431172491249e-07, "loss": 0.84498793, "num_input_tokens_seen": 148265520, "step": 6867, "time_per_iteration": 2.5388402938842773 }, { "auxiliary_loss_clip": 0.0112756, "auxiliary_loss_mlp": 0.00762398, "balance_loss_clip": 1.04051793, "balance_loss_mlp": 1.00053811, "epoch": 0.8258281729092768, "flos": 16471869250080.0, "grad_norm": 3.8709164670144602, "language_loss": 0.71964872, "learning_rate": 3.097265895790444e-07, "loss": 0.7385484, "num_input_tokens_seen": 148283730, "step": 6868, "time_per_iteration": 2.5157949924468994 }, { "auxiliary_loss_clip": 0.01128315, "auxiliary_loss_mlp": 0.01029852, "balance_loss_clip": 1.04357862, "balance_loss_mlp": 1.02241921, "epoch": 0.8259484157999158, "flos": 21433826386560.0, "grad_norm": 2.0252771026376277, "language_loss": 0.83025527, "learning_rate": 3.093103183220962e-07, "loss": 0.85183692, "num_input_tokens_seen": 148303775, "step": 6869, "time_per_iteration": 2.5444908142089844 }, { "auxiliary_loss_clip": 0.01050523, "auxiliary_loss_mlp": 0.01001837, "balance_loss_clip": 1.00585914, "balance_loss_mlp": 1.00053203, "epoch": 0.8260686586905549, "flos": 58322353004160.0, "grad_norm": 0.8220561110955935, "language_loss": 0.59412956, "learning_rate": 3.0889430354142796e-07, "loss": 0.61465311, "num_input_tokens_seen": 148365285, "step": 6870, "time_per_iteration": 3.8198695182800293 }, { "auxiliary_loss_clip": 0.01125824, "auxiliary_loss_mlp": 0.01025465, "balance_loss_clip": 1.0396533, "balance_loss_mlp": 1.01781225, "epoch": 0.826188901581194, "flos": 27527292977760.0, "grad_norm": 1.9069264097922591, "language_loss": 0.69815934, "learning_rate": 3.084785453001497e-07, "loss": 0.7196722, "num_input_tokens_seen": 148386200, "step": 6871, "time_per_iteration": 2.595294237136841 }, { "auxiliary_loss_clip": 0.01140144, "auxiliary_loss_mlp": 0.00762038, "balance_loss_clip": 1.04630232, "balance_loss_mlp": 1.00047696, "epoch": 0.8263091444718331, "flos": 23696055122880.0, "grad_norm": 2.1553804455275305, "language_loss": 0.82130671, "learning_rate": 3.080630436613314e-07, "loss": 0.84032845, "num_input_tokens_seen": 148403970, "step": 6872, "time_per_iteration": 3.314009666442871 }, { "auxiliary_loss_clip": 0.0114471, "auxiliary_loss_mlp": 0.01023232, "balance_loss_clip": 1.04332113, "balance_loss_mlp": 1.01631832, "epoch": 0.8264293873624722, "flos": 17165160701760.0, "grad_norm": 2.2996515701747664, "language_loss": 0.86112869, "learning_rate": 3.076477986880039e-07, "loss": 0.88280809, "num_input_tokens_seen": 148421765, "step": 6873, "time_per_iteration": 2.468606948852539 }, { "auxiliary_loss_clip": 0.01141682, "auxiliary_loss_mlp": 0.01023226, "balance_loss_clip": 1.04658556, "balance_loss_mlp": 1.01647592, "epoch": 0.8265496302531112, "flos": 24098649889920.0, "grad_norm": 2.929815672743224, "language_loss": 0.69425964, "learning_rate": 3.0723281044315986e-07, "loss": 0.71590871, "num_input_tokens_seen": 148443720, "step": 6874, "time_per_iteration": 3.232511043548584 }, { "auxiliary_loss_clip": 0.01163018, "auxiliary_loss_mlp": 0.01021334, "balance_loss_clip": 1.0447247, "balance_loss_mlp": 1.01459312, "epoch": 0.8266698731437504, "flos": 14099897448960.0, "grad_norm": 4.488711708575554, "language_loss": 0.76206577, "learning_rate": 3.068180789897521e-07, "loss": 0.78390932, "num_input_tokens_seen": 148462130, "step": 6875, "time_per_iteration": 2.4415061473846436 }, { "auxiliary_loss_clip": 0.01159323, "auxiliary_loss_mlp": 0.01025601, "balance_loss_clip": 1.04664683, "balance_loss_mlp": 1.01883626, "epoch": 0.8267901160343895, "flos": 30777568466880.0, "grad_norm": 3.8366907339599523, "language_loss": 0.8148396, "learning_rate": 3.064036043906966e-07, "loss": 0.83668882, "num_input_tokens_seen": 148485570, "step": 6876, "time_per_iteration": 2.57330584526062 }, { "auxiliary_loss_clip": 0.01133787, "auxiliary_loss_mlp": 0.0102721, "balance_loss_clip": 1.04368019, "balance_loss_mlp": 1.02000344, "epoch": 0.8269103589250285, "flos": 40624920557760.0, "grad_norm": 1.9518884690181428, "language_loss": 0.678253, "learning_rate": 3.059893867088668e-07, "loss": 0.69986296, "num_input_tokens_seen": 148509715, "step": 6877, "time_per_iteration": 2.735619306564331 }, { "auxiliary_loss_clip": 0.01151789, "auxiliary_loss_mlp": 0.01020533, "balance_loss_clip": 1.04530048, "balance_loss_mlp": 1.01406574, "epoch": 0.8270306018156677, "flos": 30263650288800.0, "grad_norm": 2.9519162331514885, "language_loss": 0.66823316, "learning_rate": 3.055754260071004e-07, "loss": 0.68995643, "num_input_tokens_seen": 148532010, "step": 6878, "time_per_iteration": 2.5443990230560303 }, { "auxiliary_loss_clip": 0.01156361, "auxiliary_loss_mlp": 0.01025987, "balance_loss_clip": 1.04721272, "balance_loss_mlp": 1.01926684, "epoch": 0.8271508447063067, "flos": 25226603565600.0, "grad_norm": 1.9584737551052696, "language_loss": 0.73449856, "learning_rate": 3.051617223481948e-07, "loss": 0.75632203, "num_input_tokens_seen": 148553330, "step": 6879, "time_per_iteration": 2.5345911979675293 }, { "auxiliary_loss_clip": 0.01132966, "auxiliary_loss_mlp": 0.01026114, "balance_loss_clip": 1.04362798, "balance_loss_mlp": 1.01867199, "epoch": 0.8272710875969458, "flos": 17566606126080.0, "grad_norm": 1.8442316889007018, "language_loss": 0.75250083, "learning_rate": 3.047482757949078e-07, "loss": 0.7740916, "num_input_tokens_seen": 148570960, "step": 6880, "time_per_iteration": 2.5137674808502197 }, { "auxiliary_loss_clip": 0.01122089, "auxiliary_loss_mlp": 0.00761557, "balance_loss_clip": 1.04071617, "balance_loss_mlp": 1.0005635, "epoch": 0.827391330487585, "flos": 19755469289760.0, "grad_norm": 1.9134243770784658, "language_loss": 0.85780478, "learning_rate": 3.043350864099605e-07, "loss": 0.87664127, "num_input_tokens_seen": 148589520, "step": 6881, "time_per_iteration": 2.564901351928711 }, { "auxiliary_loss_clip": 0.01156305, "auxiliary_loss_mlp": 0.01024758, "balance_loss_clip": 1.04559398, "balance_loss_mlp": 1.0181179, "epoch": 0.827511573378224, "flos": 16835177251200.0, "grad_norm": 2.092426477727026, "language_loss": 0.80569559, "learning_rate": 3.039221542560315e-07, "loss": 0.8275063, "num_input_tokens_seen": 148606085, "step": 6882, "time_per_iteration": 2.4451074600219727 }, { "auxiliary_loss_clip": 0.0115369, "auxiliary_loss_mlp": 0.01026861, "balance_loss_clip": 1.04593861, "balance_loss_mlp": 1.02030778, "epoch": 0.8276318162688631, "flos": 18369245556000.0, "grad_norm": 1.8419875507959642, "language_loss": 0.73367524, "learning_rate": 3.0350947939576356e-07, "loss": 0.75548077, "num_input_tokens_seen": 148625240, "step": 6883, "time_per_iteration": 2.4718382358551025 }, { "auxiliary_loss_clip": 0.01160071, "auxiliary_loss_mlp": 0.01023075, "balance_loss_clip": 1.0470984, "balance_loss_mlp": 1.01599669, "epoch": 0.8277520591595022, "flos": 19352694937920.0, "grad_norm": 1.6155713008174595, "language_loss": 0.72048748, "learning_rate": 3.0309706189175876e-07, "loss": 0.74231899, "num_input_tokens_seen": 148645075, "step": 6884, "time_per_iteration": 2.489914655685425 }, { "auxiliary_loss_clip": 0.01041356, "auxiliary_loss_mlp": 0.01000473, "balance_loss_clip": 1.00579834, "balance_loss_mlp": 0.9990961, "epoch": 0.8278723020501413, "flos": 67918869847680.0, "grad_norm": 0.7647187888091778, "language_loss": 0.57344496, "learning_rate": 3.0268490180658045e-07, "loss": 0.59386331, "num_input_tokens_seen": 148707855, "step": 6885, "time_per_iteration": 3.79870867729187 }, { "auxiliary_loss_clip": 0.01172407, "auxiliary_loss_mlp": 0.01022042, "balance_loss_clip": 1.05027449, "balance_loss_mlp": 1.01528585, "epoch": 0.8279925449407803, "flos": 18185741832000.0, "grad_norm": 3.932250122478064, "language_loss": 0.79055107, "learning_rate": 3.0227299920275305e-07, "loss": 0.81249559, "num_input_tokens_seen": 148724170, "step": 6886, "time_per_iteration": 2.426084518432617 }, { "auxiliary_loss_clip": 0.01132478, "auxiliary_loss_mlp": 0.01024998, "balance_loss_clip": 1.04759455, "balance_loss_mlp": 1.01738679, "epoch": 0.8281127878314195, "flos": 20631438375360.0, "grad_norm": 2.1146764359665147, "language_loss": 0.8565051, "learning_rate": 3.018613541427613e-07, "loss": 0.87807977, "num_input_tokens_seen": 148743690, "step": 6887, "time_per_iteration": 2.550806760787964 }, { "auxiliary_loss_clip": 0.01167261, "auxiliary_loss_mlp": 0.01027621, "balance_loss_clip": 1.04555392, "balance_loss_mlp": 1.020486, "epoch": 0.8282330307220586, "flos": 18004285374720.0, "grad_norm": 1.617525947474326, "language_loss": 0.73235852, "learning_rate": 3.0144996668905243e-07, "loss": 0.75430727, "num_input_tokens_seen": 148761070, "step": 6888, "time_per_iteration": 2.411306619644165 }, { "auxiliary_loss_clip": 0.01099791, "auxiliary_loss_mlp": 0.00761826, "balance_loss_clip": 1.03783131, "balance_loss_mlp": 1.00059009, "epoch": 0.8283532736126976, "flos": 20084123812800.0, "grad_norm": 2.366278595651219, "language_loss": 0.81960976, "learning_rate": 3.010388369040331e-07, "loss": 0.83822596, "num_input_tokens_seen": 148779730, "step": 6889, "time_per_iteration": 2.7323644161224365 }, { "auxiliary_loss_clip": 0.01155221, "auxiliary_loss_mlp": 0.01025999, "balance_loss_clip": 1.04711986, "balance_loss_mlp": 1.01842356, "epoch": 0.8284735165033368, "flos": 31868426311200.0, "grad_norm": 2.0059121113310607, "language_loss": 0.82922971, "learning_rate": 3.0062796485007156e-07, "loss": 0.85104191, "num_input_tokens_seen": 148800670, "step": 6890, "time_per_iteration": 2.779634475708008 }, { "auxiliary_loss_clip": 0.01168055, "auxiliary_loss_mlp": 0.00762419, "balance_loss_clip": 1.04690051, "balance_loss_mlp": 1.00060081, "epoch": 0.8285937593939758, "flos": 26651323892160.0, "grad_norm": 2.5784167957974597, "language_loss": 0.65622747, "learning_rate": 3.002173505894965e-07, "loss": 0.67553222, "num_input_tokens_seen": 148819820, "step": 6891, "time_per_iteration": 2.494631767272949 }, { "auxiliary_loss_clip": 0.01159966, "auxiliary_loss_mlp": 0.01025225, "balance_loss_clip": 1.04702878, "balance_loss_mlp": 1.01725292, "epoch": 0.8287140022846149, "flos": 20193687292800.0, "grad_norm": 2.8305106360842553, "language_loss": 0.62283677, "learning_rate": 2.998069941845973e-07, "loss": 0.64468873, "num_input_tokens_seen": 148838890, "step": 6892, "time_per_iteration": 2.489881992340088 }, { "auxiliary_loss_clip": 0.01060467, "auxiliary_loss_mlp": 0.00999616, "balance_loss_clip": 1.0055933, "balance_loss_mlp": 0.99841231, "epoch": 0.8288342451752541, "flos": 70755990983520.0, "grad_norm": 0.7139726682738038, "language_loss": 0.57518244, "learning_rate": 2.993968956976258e-07, "loss": 0.59578323, "num_input_tokens_seen": 148906635, "step": 6893, "time_per_iteration": 3.1511220932006836 }, { "auxiliary_loss_clip": 0.0117432, "auxiliary_loss_mlp": 0.0102642, "balance_loss_clip": 1.04852843, "balance_loss_mlp": 1.01895785, "epoch": 0.8289544880658931, "flos": 24572239710240.0, "grad_norm": 1.9251831362636556, "language_loss": 0.69940233, "learning_rate": 2.9898705519079313e-07, "loss": 0.72140974, "num_input_tokens_seen": 148925740, "step": 6894, "time_per_iteration": 2.4789419174194336 }, { "auxiliary_loss_clip": 0.01132709, "auxiliary_loss_mlp": 0.0102391, "balance_loss_clip": 1.04271376, "balance_loss_mlp": 1.01691258, "epoch": 0.8290747309565322, "flos": 22273382063040.0, "grad_norm": 1.6880202029381548, "language_loss": 0.75044501, "learning_rate": 2.985774727262715e-07, "loss": 0.77201116, "num_input_tokens_seen": 148944585, "step": 6895, "time_per_iteration": 3.262873649597168 }, { "auxiliary_loss_clip": 0.01168877, "auxiliary_loss_mlp": 0.0102429, "balance_loss_clip": 1.04776609, "balance_loss_mlp": 1.01796901, "epoch": 0.8291949738471713, "flos": 23255574351360.0, "grad_norm": 1.8264609280521409, "language_loss": 0.81264144, "learning_rate": 2.981681483661949e-07, "loss": 0.83457303, "num_input_tokens_seen": 148964170, "step": 6896, "time_per_iteration": 2.4569475650787354 }, { "auxiliary_loss_clip": 0.01154424, "auxiliary_loss_mlp": 0.01026761, "balance_loss_clip": 1.04750061, "balance_loss_mlp": 1.01971245, "epoch": 0.8293152167378104, "flos": 52555774616640.0, "grad_norm": 1.9136083452411747, "language_loss": 0.71308434, "learning_rate": 2.9775908217265633e-07, "loss": 0.73489612, "num_input_tokens_seen": 148989405, "step": 6897, "time_per_iteration": 2.7668659687042236 }, { "auxiliary_loss_clip": 0.01008527, "auxiliary_loss_mlp": 0.01004733, "balance_loss_clip": 1.00557017, "balance_loss_mlp": 1.00345123, "epoch": 0.8294354596284494, "flos": 63356167200960.0, "grad_norm": 0.8252833019453656, "language_loss": 0.5036723, "learning_rate": 2.9735027420771253e-07, "loss": 0.5238049, "num_input_tokens_seen": 149049740, "step": 6898, "time_per_iteration": 4.00630521774292 }, { "auxiliary_loss_clip": 0.01134588, "auxiliary_loss_mlp": 0.01022076, "balance_loss_clip": 1.04656911, "balance_loss_mlp": 1.01611233, "epoch": 0.8295557025190886, "flos": 24827025823200.0, "grad_norm": 1.7249758691361172, "language_loss": 0.71482998, "learning_rate": 2.969417245333774e-07, "loss": 0.73639661, "num_input_tokens_seen": 149069120, "step": 6899, "time_per_iteration": 2.7287371158599854 }, { "auxiliary_loss_clip": 0.01122058, "auxiliary_loss_mlp": 0.01025238, "balance_loss_clip": 1.04451275, "balance_loss_mlp": 1.01870525, "epoch": 0.8296759454097277, "flos": 25118584514880.0, "grad_norm": 2.5904924839328265, "language_loss": 0.77953666, "learning_rate": 2.9653343321162915e-07, "loss": 0.80100965, "num_input_tokens_seen": 149088630, "step": 6900, "time_per_iteration": 3.3042988777160645 }, { "auxiliary_loss_clip": 0.01127703, "auxiliary_loss_mlp": 0.01028298, "balance_loss_clip": 1.04534745, "balance_loss_mlp": 1.02085388, "epoch": 0.8297961883003667, "flos": 24132585028800.0, "grad_norm": 2.066530234193199, "language_loss": 0.6470893, "learning_rate": 2.9612540030440446e-07, "loss": 0.66864932, "num_input_tokens_seen": 149109175, "step": 6901, "time_per_iteration": 2.5634655952453613 }, { "auxiliary_loss_clip": 0.01040776, "auxiliary_loss_mlp": 0.01001651, "balance_loss_clip": 1.00502801, "balance_loss_mlp": 1.00035751, "epoch": 0.8299164311910058, "flos": 67446573037920.0, "grad_norm": 0.8737274610417293, "language_loss": 0.64104503, "learning_rate": 2.9571762587360206e-07, "loss": 0.66146928, "num_input_tokens_seen": 149165560, "step": 6902, "time_per_iteration": 3.072280168533325 }, { "auxiliary_loss_clip": 0.01107754, "auxiliary_loss_mlp": 0.01021601, "balance_loss_clip": 1.0359993, "balance_loss_mlp": 1.01478565, "epoch": 0.8300366740816449, "flos": 25228686749280.0, "grad_norm": 1.6017049417375775, "language_loss": 0.74109763, "learning_rate": 2.953101099810806e-07, "loss": 0.76239121, "num_input_tokens_seen": 149185165, "step": 6903, "time_per_iteration": 2.6294186115264893 }, { "auxiliary_loss_clip": 0.01150975, "auxiliary_loss_mlp": 0.01026744, "balance_loss_clip": 1.04840684, "balance_loss_mlp": 1.0199132, "epoch": 0.830156916972284, "flos": 18041022036480.0, "grad_norm": 2.1002792678417697, "language_loss": 0.82483387, "learning_rate": 2.9490285268865965e-07, "loss": 0.84661102, "num_input_tokens_seen": 149202655, "step": 6904, "time_per_iteration": 2.4576597213745117 }, { "auxiliary_loss_clip": 0.01161375, "auxiliary_loss_mlp": 0.01031868, "balance_loss_clip": 1.04976332, "balance_loss_mlp": 1.0238421, "epoch": 0.830277159862923, "flos": 26322489784320.0, "grad_norm": 2.2903369687776594, "language_loss": 0.7969287, "learning_rate": 2.9449585405812085e-07, "loss": 0.81886119, "num_input_tokens_seen": 149220035, "step": 6905, "time_per_iteration": 2.517280340194702 }, { "auxiliary_loss_clip": 0.01126899, "auxiliary_loss_mlp": 0.01023838, "balance_loss_clip": 1.04531693, "balance_loss_mlp": 1.01684976, "epoch": 0.8303974027535622, "flos": 19938865262880.0, "grad_norm": 1.8200946830700513, "language_loss": 0.73815405, "learning_rate": 2.940891141512043e-07, "loss": 0.75966144, "num_input_tokens_seen": 149238055, "step": 6906, "time_per_iteration": 2.5345711708068848 }, { "auxiliary_loss_clip": 0.01139639, "auxiliary_loss_mlp": 0.01031575, "balance_loss_clip": 1.04414737, "balance_loss_mlp": 1.02467561, "epoch": 0.8305176456442013, "flos": 17165555788320.0, "grad_norm": 2.496488980497204, "language_loss": 0.72068954, "learning_rate": 2.9368263302961385e-07, "loss": 0.74240166, "num_input_tokens_seen": 149256755, "step": 6907, "time_per_iteration": 2.4864211082458496 }, { "auxiliary_loss_clip": 0.01097202, "auxiliary_loss_mlp": 0.01024353, "balance_loss_clip": 1.03847361, "balance_loss_mlp": 1.01728964, "epoch": 0.8306378885348403, "flos": 25627617986400.0, "grad_norm": 1.7561942547883584, "language_loss": 0.79814941, "learning_rate": 2.9327641075501075e-07, "loss": 0.81936491, "num_input_tokens_seen": 149275745, "step": 6908, "time_per_iteration": 2.6637260913848877 }, { "auxiliary_loss_clip": 0.01132458, "auxiliary_loss_mlp": 0.01026839, "balance_loss_clip": 1.04144359, "balance_loss_mlp": 1.01980019, "epoch": 0.8307581314254795, "flos": 33947869662720.0, "grad_norm": 3.8720718364024114, "language_loss": 0.66352051, "learning_rate": 2.9287044738901866e-07, "loss": 0.68511349, "num_input_tokens_seen": 149293730, "step": 6909, "time_per_iteration": 2.609489679336548 }, { "auxiliary_loss_clip": 0.01155687, "auxiliary_loss_mlp": 0.00761766, "balance_loss_clip": 1.04584098, "balance_loss_mlp": 1.00057304, "epoch": 0.8308783743161186, "flos": 17562727094400.0, "grad_norm": 3.747398577823654, "language_loss": 0.90826619, "learning_rate": 2.9246474299322274e-07, "loss": 0.92744064, "num_input_tokens_seen": 149309290, "step": 6910, "time_per_iteration": 2.4741580486297607 }, { "auxiliary_loss_clip": 0.01022072, "auxiliary_loss_mlp": 0.01000611, "balance_loss_clip": 1.00501776, "balance_loss_mlp": 0.99934149, "epoch": 0.8309986172067576, "flos": 69412897130400.0, "grad_norm": 0.8911240890592962, "language_loss": 0.63168585, "learning_rate": 2.920592976291678e-07, "loss": 0.65191269, "num_input_tokens_seen": 149366620, "step": 6911, "time_per_iteration": 3.7742691040039062 }, { "auxiliary_loss_clip": 0.01152787, "auxiliary_loss_mlp": 0.01034078, "balance_loss_clip": 1.04515243, "balance_loss_mlp": 1.02673507, "epoch": 0.8311188600973968, "flos": 22309759555200.0, "grad_norm": 2.346676998824798, "language_loss": 0.80880558, "learning_rate": 2.916541113583595e-07, "loss": 0.83067423, "num_input_tokens_seen": 149385120, "step": 6912, "time_per_iteration": 2.488903760910034 }, { "auxiliary_loss_clip": 0.01129634, "auxiliary_loss_mlp": 0.01023046, "balance_loss_clip": 1.04742491, "balance_loss_mlp": 1.01614368, "epoch": 0.8312391029880358, "flos": 18770080392000.0, "grad_norm": 2.3555474782398376, "language_loss": 0.66421056, "learning_rate": 2.912491842422642e-07, "loss": 0.68573737, "num_input_tokens_seen": 149402825, "step": 6913, "time_per_iteration": 2.5229272842407227 }, { "auxiliary_loss_clip": 0.01156379, "auxiliary_loss_mlp": 0.01025578, "balance_loss_clip": 1.04726458, "balance_loss_mlp": 1.01907229, "epoch": 0.8313593458786749, "flos": 20376652262400.0, "grad_norm": 1.7518520489042286, "language_loss": 0.70935225, "learning_rate": 2.9084451634230857e-07, "loss": 0.73117185, "num_input_tokens_seen": 149422125, "step": 6914, "time_per_iteration": 2.4934585094451904 }, { "auxiliary_loss_clip": 0.01124676, "auxiliary_loss_mlp": 0.01024155, "balance_loss_clip": 1.04156983, "balance_loss_mlp": 1.0169251, "epoch": 0.831479588769314, "flos": 32124074431200.0, "grad_norm": 2.1787310796318327, "language_loss": 0.71185637, "learning_rate": 2.9044010771988125e-07, "loss": 0.73334461, "num_input_tokens_seen": 149441940, "step": 6915, "time_per_iteration": 2.616304397583008 }, { "auxiliary_loss_clip": 0.01133337, "auxiliary_loss_mlp": 0.01024073, "balance_loss_clip": 1.04338932, "balance_loss_mlp": 1.01770401, "epoch": 0.8315998316599531, "flos": 45185935107360.0, "grad_norm": 1.9593048931079364, "language_loss": 0.71833956, "learning_rate": 2.900359584363303e-07, "loss": 0.7399137, "num_input_tokens_seen": 149465045, "step": 6916, "time_per_iteration": 2.7073140144348145 }, { "auxiliary_loss_clip": 0.01109261, "auxiliary_loss_mlp": 0.01028251, "balance_loss_clip": 1.04352534, "balance_loss_mlp": 1.02108908, "epoch": 0.8317200745505922, "flos": 18363750261120.0, "grad_norm": 2.161305158825501, "language_loss": 0.84476465, "learning_rate": 2.8963206855296494e-07, "loss": 0.86613977, "num_input_tokens_seen": 149481285, "step": 6917, "time_per_iteration": 2.5288937091827393 }, { "auxiliary_loss_clip": 0.011559, "auxiliary_loss_mlp": 0.0102512, "balance_loss_clip": 1.04528689, "balance_loss_mlp": 1.01848912, "epoch": 0.8318403174412313, "flos": 24206561189760.0, "grad_norm": 1.6923489096942095, "language_loss": 0.77075064, "learning_rate": 2.892284381310548e-07, "loss": 0.79256082, "num_input_tokens_seen": 149502700, "step": 6918, "time_per_iteration": 2.521066904067993 }, { "auxiliary_loss_clip": 0.01138748, "auxiliary_loss_mlp": 0.01024086, "balance_loss_clip": 1.04583526, "balance_loss_mlp": 1.01710629, "epoch": 0.8319605603318704, "flos": 22418784280800.0, "grad_norm": 2.428685252707848, "language_loss": 0.71887159, "learning_rate": 2.888250672318302e-07, "loss": 0.74049991, "num_input_tokens_seen": 149520100, "step": 6919, "time_per_iteration": 2.5051558017730713 }, { "auxiliary_loss_clip": 0.01172878, "auxiliary_loss_mlp": 0.01028526, "balance_loss_clip": 1.0512116, "balance_loss_mlp": 1.02173996, "epoch": 0.8320808032225094, "flos": 37414506505920.0, "grad_norm": 1.6462230074315547, "language_loss": 0.68599451, "learning_rate": 2.884219559164831e-07, "loss": 0.70800853, "num_input_tokens_seen": 149543245, "step": 6920, "time_per_iteration": 2.601572275161743 }, { "auxiliary_loss_clip": 0.01153656, "auxiliary_loss_mlp": 0.01024653, "balance_loss_clip": 1.0472008, "balance_loss_mlp": 1.01724446, "epoch": 0.8322010461131486, "flos": 12787398457440.0, "grad_norm": 3.333754077783793, "language_loss": 0.81374353, "learning_rate": 2.880191042461635e-07, "loss": 0.83552665, "num_input_tokens_seen": 149559185, "step": 6921, "time_per_iteration": 3.1868557929992676 }, { "auxiliary_loss_clip": 0.01118985, "auxiliary_loss_mlp": 0.01025224, "balance_loss_clip": 1.04200888, "balance_loss_mlp": 1.01853895, "epoch": 0.8323212890037877, "flos": 15815458128000.0, "grad_norm": 1.7199692424585662, "language_loss": 0.80213773, "learning_rate": 2.876165122819849e-07, "loss": 0.82357979, "num_input_tokens_seen": 149577165, "step": 6922, "time_per_iteration": 2.56501841545105 }, { "auxiliary_loss_clip": 0.0116683, "auxiliary_loss_mlp": 0.01022334, "balance_loss_clip": 1.04755902, "balance_loss_mlp": 1.01530981, "epoch": 0.8324415318944267, "flos": 21719279195040.0, "grad_norm": 1.6087927951157535, "language_loss": 0.79420894, "learning_rate": 2.872141800850201e-07, "loss": 0.8161006, "num_input_tokens_seen": 149594340, "step": 6923, "time_per_iteration": 3.2050986289978027 }, { "auxiliary_loss_clip": 0.01167445, "auxiliary_loss_mlp": 0.01027504, "balance_loss_clip": 1.04717422, "balance_loss_mlp": 1.02050352, "epoch": 0.8325617747850659, "flos": 34198704910080.0, "grad_norm": 1.6001857996456594, "language_loss": 0.73032993, "learning_rate": 2.868121077163024e-07, "loss": 0.7522794, "num_input_tokens_seen": 149613895, "step": 6924, "time_per_iteration": 2.554245948791504 }, { "auxiliary_loss_clip": 0.0115485, "auxiliary_loss_mlp": 0.01024475, "balance_loss_clip": 1.0441066, "balance_loss_mlp": 1.01713777, "epoch": 0.8326820176757049, "flos": 18369461057760.0, "grad_norm": 1.778280750390388, "language_loss": 0.72174513, "learning_rate": 2.864102952368257e-07, "loss": 0.74353838, "num_input_tokens_seen": 149631820, "step": 6925, "time_per_iteration": 2.4746973514556885 }, { "auxiliary_loss_clip": 0.01098796, "auxiliary_loss_mlp": 0.01024488, "balance_loss_clip": 1.03634882, "balance_loss_mlp": 1.01760995, "epoch": 0.832802260566344, "flos": 35991330608640.0, "grad_norm": 1.2373682964085762, "language_loss": 0.59222937, "learning_rate": 2.860087427075444e-07, "loss": 0.61346221, "num_input_tokens_seen": 149656070, "step": 6926, "time_per_iteration": 3.4455268383026123 }, { "auxiliary_loss_clip": 0.01135746, "auxiliary_loss_mlp": 0.01030322, "balance_loss_clip": 1.04401851, "balance_loss_mlp": 1.02347088, "epoch": 0.8329225034569832, "flos": 14244437659680.0, "grad_norm": 2.4127617439936113, "language_loss": 0.85990536, "learning_rate": 2.856074501893744e-07, "loss": 0.88156605, "num_input_tokens_seen": 149671270, "step": 6927, "time_per_iteration": 2.4905600547790527 }, { "auxiliary_loss_clip": 0.0115962, "auxiliary_loss_mlp": 0.01025122, "balance_loss_clip": 1.05042028, "balance_loss_mlp": 1.01816368, "epoch": 0.8330427463476222, "flos": 18077471362560.0, "grad_norm": 1.7146491836584707, "language_loss": 0.81458563, "learning_rate": 2.8520641774319054e-07, "loss": 0.83643305, "num_input_tokens_seen": 149689360, "step": 6928, "time_per_iteration": 2.453645706176758 }, { "auxiliary_loss_clip": 0.01141935, "auxiliary_loss_mlp": 0.01030583, "balance_loss_clip": 1.04171658, "balance_loss_mlp": 1.02328706, "epoch": 0.8331629892382613, "flos": 18040842451680.0, "grad_norm": 2.196799495868883, "language_loss": 0.75572693, "learning_rate": 2.848056454298309e-07, "loss": 0.77745211, "num_input_tokens_seen": 149706685, "step": 6929, "time_per_iteration": 2.500077962875366 }, { "auxiliary_loss_clip": 0.01139737, "auxiliary_loss_mlp": 0.01028648, "balance_loss_clip": 1.04605579, "balance_loss_mlp": 1.02168357, "epoch": 0.8332832321289004, "flos": 17457401815680.0, "grad_norm": 2.1939113557567818, "language_loss": 0.65302444, "learning_rate": 2.844051333100905e-07, "loss": 0.67470831, "num_input_tokens_seen": 149724230, "step": 6930, "time_per_iteration": 2.4651148319244385 }, { "auxiliary_loss_clip": 0.01141022, "auxiliary_loss_mlp": 0.01024211, "balance_loss_clip": 1.04779005, "balance_loss_mlp": 1.01801765, "epoch": 0.8334034750195395, "flos": 15084855343200.0, "grad_norm": 2.9487731997181665, "language_loss": 0.83660662, "learning_rate": 2.840048814447269e-07, "loss": 0.85825896, "num_input_tokens_seen": 149742395, "step": 6931, "time_per_iteration": 2.4895706176757812 }, { "auxiliary_loss_clip": 0.01132067, "auxiliary_loss_mlp": 0.01023975, "balance_loss_clip": 1.04324853, "balance_loss_mlp": 1.01729369, "epoch": 0.8335237179101785, "flos": 19427173936320.0, "grad_norm": 2.3317985100450023, "language_loss": 0.74125421, "learning_rate": 2.836048898944587e-07, "loss": 0.76281458, "num_input_tokens_seen": 149760820, "step": 6932, "time_per_iteration": 2.4908902645111084 }, { "auxiliary_loss_clip": 0.01138367, "auxiliary_loss_mlp": 0.01026071, "balance_loss_clip": 1.04408383, "balance_loss_mlp": 1.01913595, "epoch": 0.8336439608008177, "flos": 21762049906080.0, "grad_norm": 2.179390938198981, "language_loss": 0.72637159, "learning_rate": 2.832051587199642e-07, "loss": 0.748016, "num_input_tokens_seen": 149778075, "step": 6933, "time_per_iteration": 2.5285215377807617 }, { "auxiliary_loss_clip": 0.0105288, "auxiliary_loss_mlp": 0.0100187, "balance_loss_clip": 1.00695992, "balance_loss_mlp": 1.00056422, "epoch": 0.8337642036914568, "flos": 59702794107360.0, "grad_norm": 0.8087054716166233, "language_loss": 0.57796335, "learning_rate": 2.828056879818821e-07, "loss": 0.59851086, "num_input_tokens_seen": 149837150, "step": 6934, "time_per_iteration": 3.0373599529266357 }, { "auxiliary_loss_clip": 0.01124573, "auxiliary_loss_mlp": 0.01023818, "balance_loss_clip": 1.03879881, "balance_loss_mlp": 1.01747918, "epoch": 0.8338844465820958, "flos": 27162189128640.0, "grad_norm": 2.9396247547132743, "language_loss": 0.83272207, "learning_rate": 2.824064777408117e-07, "loss": 0.85420597, "num_input_tokens_seen": 149856940, "step": 6935, "time_per_iteration": 2.5983407497406006 }, { "auxiliary_loss_clip": 0.01154455, "auxiliary_loss_mlp": 0.01026091, "balance_loss_clip": 1.04681134, "balance_loss_mlp": 1.01924801, "epoch": 0.8340046894727349, "flos": 30481268736480.0, "grad_norm": 1.9322899873007309, "language_loss": 0.75823903, "learning_rate": 2.8200752805731263e-07, "loss": 0.7800445, "num_input_tokens_seen": 149879930, "step": 6936, "time_per_iteration": 2.5654783248901367 }, { "auxiliary_loss_clip": 0.01154027, "auxiliary_loss_mlp": 0.01026651, "balance_loss_clip": 1.04817724, "balance_loss_mlp": 1.02007616, "epoch": 0.834124932363374, "flos": 27126170806080.0, "grad_norm": 1.4934770143574472, "language_loss": 0.80950499, "learning_rate": 2.8160883899190625e-07, "loss": 0.83131176, "num_input_tokens_seen": 149903200, "step": 6937, "time_per_iteration": 3.293727397918701 }, { "auxiliary_loss_clip": 0.01117477, "auxiliary_loss_mlp": 0.01023898, "balance_loss_clip": 1.04366994, "balance_loss_mlp": 1.01710296, "epoch": 0.8342451752540131, "flos": 24569869190880.0, "grad_norm": 4.748169163038076, "language_loss": 0.73298383, "learning_rate": 2.8121041060507234e-07, "loss": 0.75439751, "num_input_tokens_seen": 149922230, "step": 6938, "time_per_iteration": 2.5739452838897705 }, { "auxiliary_loss_clip": 0.01158928, "auxiliary_loss_mlp": 0.01025059, "balance_loss_clip": 1.04650688, "balance_loss_mlp": 1.0181303, "epoch": 0.8343654181446521, "flos": 26615090067840.0, "grad_norm": 1.7013093330732754, "language_loss": 0.71573877, "learning_rate": 2.808122429572528e-07, "loss": 0.73757863, "num_input_tokens_seen": 149942435, "step": 6939, "time_per_iteration": 2.5336825847625732 }, { "auxiliary_loss_clip": 0.0113235, "auxiliary_loss_mlp": 0.01026986, "balance_loss_clip": 1.04405701, "balance_loss_mlp": 1.02032197, "epoch": 0.8344856610352913, "flos": 20777271596640.0, "grad_norm": 2.7900815246093393, "language_loss": 0.76227212, "learning_rate": 2.804143361088489e-07, "loss": 0.78386545, "num_input_tokens_seen": 149961615, "step": 6940, "time_per_iteration": 2.5473520755767822 }, { "auxiliary_loss_clip": 0.01134076, "auxiliary_loss_mlp": 0.01032148, "balance_loss_clip": 1.04314899, "balance_loss_mlp": 1.02556205, "epoch": 0.8346059039259304, "flos": 26095963930560.0, "grad_norm": 2.491949428255499, "language_loss": 0.77935797, "learning_rate": 2.8001669012022277e-07, "loss": 0.80102021, "num_input_tokens_seen": 149979585, "step": 6941, "time_per_iteration": 2.5410187244415283 }, { "auxiliary_loss_clip": 0.01153972, "auxiliary_loss_mlp": 0.01026617, "balance_loss_clip": 1.04864407, "balance_loss_mlp": 1.01940501, "epoch": 0.8347261468165694, "flos": 29027713479360.0, "grad_norm": 2.017191957169963, "language_loss": 0.69484711, "learning_rate": 2.7961930505169795e-07, "loss": 0.71665299, "num_input_tokens_seen": 150003830, "step": 6942, "time_per_iteration": 2.597820997238159 }, { "auxiliary_loss_clip": 0.01158145, "auxiliary_loss_mlp": 0.00762075, "balance_loss_clip": 1.04962754, "balance_loss_mlp": 1.00061727, "epoch": 0.8348463897072086, "flos": 26396465945280.0, "grad_norm": 1.9768464035711812, "language_loss": 0.76427722, "learning_rate": 2.792221809635558e-07, "loss": 0.78347945, "num_input_tokens_seen": 150024460, "step": 6943, "time_per_iteration": 2.519117832183838 }, { "auxiliary_loss_clip": 0.01087638, "auxiliary_loss_mlp": 0.0102661, "balance_loss_clip": 1.04078758, "balance_loss_mlp": 1.01951456, "epoch": 0.8349666325978476, "flos": 23367723852480.0, "grad_norm": 3.8114780440233944, "language_loss": 0.74918157, "learning_rate": 2.788253179160411e-07, "loss": 0.77032405, "num_input_tokens_seen": 150045620, "step": 6944, "time_per_iteration": 2.6384761333465576 }, { "auxiliary_loss_clip": 0.01140828, "auxiliary_loss_mlp": 0.01021961, "balance_loss_clip": 1.0459702, "balance_loss_mlp": 1.01552975, "epoch": 0.8350868754884867, "flos": 12896530933920.0, "grad_norm": 2.1825580062610412, "language_loss": 0.64948267, "learning_rate": 2.7842871596935725e-07, "loss": 0.67111051, "num_input_tokens_seen": 150064135, "step": 6945, "time_per_iteration": 2.4849581718444824 }, { "auxiliary_loss_clip": 0.01155296, "auxiliary_loss_mlp": 0.01023561, "balance_loss_clip": 1.04640865, "balance_loss_mlp": 1.01626801, "epoch": 0.8352071183791259, "flos": 26505526587840.0, "grad_norm": 1.5392796633060828, "language_loss": 0.69149888, "learning_rate": 2.780323751836682e-07, "loss": 0.71328741, "num_input_tokens_seen": 150085350, "step": 6946, "time_per_iteration": 2.558701753616333 }, { "auxiliary_loss_clip": 0.01141148, "auxiliary_loss_mlp": 0.00761475, "balance_loss_clip": 1.04476976, "balance_loss_mlp": 1.00058115, "epoch": 0.8353273612697649, "flos": 20668067286240.0, "grad_norm": 1.6630886295999496, "language_loss": 0.78649253, "learning_rate": 2.7763629561909876e-07, "loss": 0.80551875, "num_input_tokens_seen": 150106180, "step": 6947, "time_per_iteration": 3.2699694633483887 }, { "auxiliary_loss_clip": 0.01163611, "auxiliary_loss_mlp": 0.01024252, "balance_loss_clip": 1.04394412, "balance_loss_mlp": 1.01755857, "epoch": 0.835447604160404, "flos": 19754140362240.0, "grad_norm": 2.681383287557435, "language_loss": 0.76735771, "learning_rate": 2.772404773357335e-07, "loss": 0.78923643, "num_input_tokens_seen": 150125585, "step": 6948, "time_per_iteration": 2.439502477645874 }, { "auxiliary_loss_clip": 0.01118173, "auxiliary_loss_mlp": 0.01027471, "balance_loss_clip": 1.04222012, "balance_loss_mlp": 1.02063179, "epoch": 0.8355678470510431, "flos": 23435845548960.0, "grad_norm": 1.9326496183143926, "language_loss": 0.78568959, "learning_rate": 2.7684492039361853e-07, "loss": 0.80714607, "num_input_tokens_seen": 150144810, "step": 6949, "time_per_iteration": 2.555885076522827 }, { "auxiliary_loss_clip": 0.01170485, "auxiliary_loss_mlp": 0.01027347, "balance_loss_clip": 1.0491426, "balance_loss_mlp": 1.02012897, "epoch": 0.8356880899416822, "flos": 21214591675680.0, "grad_norm": 1.8130323198202405, "language_loss": 0.83628905, "learning_rate": 2.764496248527586e-07, "loss": 0.85826743, "num_input_tokens_seen": 150163785, "step": 6950, "time_per_iteration": 3.2205300331115723 }, { "auxiliary_loss_clip": 0.01130913, "auxiliary_loss_mlp": 0.01026079, "balance_loss_clip": 1.04300904, "balance_loss_mlp": 1.01924515, "epoch": 0.8358083328323213, "flos": 28037547625920.0, "grad_norm": 2.0072835788421766, "language_loss": 0.78252852, "learning_rate": 2.760545907731211e-07, "loss": 0.80409849, "num_input_tokens_seen": 150184360, "step": 6951, "time_per_iteration": 2.5956809520721436 }, { "auxiliary_loss_clip": 0.01154311, "auxiliary_loss_mlp": 0.01024736, "balance_loss_clip": 1.04394507, "balance_loss_mlp": 1.0177356, "epoch": 0.8359285757229604, "flos": 27783659436960.0, "grad_norm": 2.8656243376517567, "language_loss": 0.68029594, "learning_rate": 2.75659818214631e-07, "loss": 0.70208645, "num_input_tokens_seen": 150205465, "step": 6952, "time_per_iteration": 3.29482364654541 }, { "auxiliary_loss_clip": 0.01142432, "auxiliary_loss_mlp": 0.01025925, "balance_loss_clip": 1.04505658, "balance_loss_mlp": 1.01912975, "epoch": 0.8360488186135995, "flos": 21435119397120.0, "grad_norm": 1.7852871622649207, "language_loss": 0.77935946, "learning_rate": 2.752653072371749e-07, "loss": 0.80104303, "num_input_tokens_seen": 150224900, "step": 6953, "time_per_iteration": 2.535031318664551 }, { "auxiliary_loss_clip": 0.01121444, "auxiliary_loss_mlp": 0.01022988, "balance_loss_clip": 1.04440165, "balance_loss_mlp": 1.01614833, "epoch": 0.8361690615042385, "flos": 27632330920800.0, "grad_norm": 1.6579510048822519, "language_loss": 0.7473774, "learning_rate": 2.7487105790060105e-07, "loss": 0.76882166, "num_input_tokens_seen": 150244310, "step": 6954, "time_per_iteration": 2.5812580585479736 }, { "auxiliary_loss_clip": 0.01155443, "auxiliary_loss_mlp": 0.01023882, "balance_loss_clip": 1.0450778, "balance_loss_mlp": 1.01730204, "epoch": 0.8362893043948777, "flos": 39202534833600.0, "grad_norm": 1.7871926959387119, "language_loss": 0.69415396, "learning_rate": 2.7447707026471587e-07, "loss": 0.71594721, "num_input_tokens_seen": 150267285, "step": 6955, "time_per_iteration": 2.6356921195983887 }, { "auxiliary_loss_clip": 0.01126783, "auxiliary_loss_mlp": 0.01026616, "balance_loss_clip": 1.04225349, "balance_loss_mlp": 1.0204705, "epoch": 0.8364095472855168, "flos": 24785332620960.0, "grad_norm": 2.215135732135125, "language_loss": 0.80065733, "learning_rate": 2.740833443892874e-07, "loss": 0.82219136, "num_input_tokens_seen": 150285455, "step": 6956, "time_per_iteration": 2.5666685104370117 }, { "auxiliary_loss_clip": 0.01139273, "auxiliary_loss_mlp": 0.01020965, "balance_loss_clip": 1.04357982, "balance_loss_mlp": 1.01420259, "epoch": 0.8365297901761558, "flos": 22743416104320.0, "grad_norm": 1.7000174274473476, "language_loss": 0.79548359, "learning_rate": 2.7368988033404327e-07, "loss": 0.81708592, "num_input_tokens_seen": 150302970, "step": 6957, "time_per_iteration": 2.5329930782318115 }, { "auxiliary_loss_clip": 0.01128082, "auxiliary_loss_mlp": 0.01022154, "balance_loss_clip": 1.04384875, "balance_loss_mlp": 1.01559401, "epoch": 0.836650033066795, "flos": 28396006837440.0, "grad_norm": 1.5287716103582347, "language_loss": 0.84448469, "learning_rate": 2.732966781586712e-07, "loss": 0.86598706, "num_input_tokens_seen": 150322715, "step": 6958, "time_per_iteration": 2.5938045978546143 }, { "auxiliary_loss_clip": 0.01146787, "auxiliary_loss_mlp": 0.01020008, "balance_loss_clip": 1.0417608, "balance_loss_mlp": 1.01347542, "epoch": 0.836770275957434, "flos": 22236860903040.0, "grad_norm": 1.5693983947727068, "language_loss": 0.66656041, "learning_rate": 2.729037379228205e-07, "loss": 0.68822837, "num_input_tokens_seen": 150342900, "step": 6959, "time_per_iteration": 2.5025408267974854 }, { "auxiliary_loss_clip": 0.01140138, "auxiliary_loss_mlp": 0.01028644, "balance_loss_clip": 1.04789674, "balance_loss_mlp": 1.02163458, "epoch": 0.8368905188480731, "flos": 22491934351680.0, "grad_norm": 1.4932014442931714, "language_loss": 0.8060571, "learning_rate": 2.725110596860998e-07, "loss": 0.82774484, "num_input_tokens_seen": 150363580, "step": 6960, "time_per_iteration": 2.5233211517333984 }, { "auxiliary_loss_clip": 0.01108517, "auxiliary_loss_mlp": 0.0102426, "balance_loss_clip": 1.04346776, "balance_loss_mlp": 1.01811767, "epoch": 0.8370107617387123, "flos": 13370408089920.0, "grad_norm": 1.8762149070824337, "language_loss": 0.70165825, "learning_rate": 2.7211864350807776e-07, "loss": 0.72298604, "num_input_tokens_seen": 150381780, "step": 6961, "time_per_iteration": 2.5559399127960205 }, { "auxiliary_loss_clip": 0.01167886, "auxiliary_loss_mlp": 0.01027887, "balance_loss_clip": 1.04689813, "balance_loss_mlp": 1.01994717, "epoch": 0.8371310046293513, "flos": 25261292960640.0, "grad_norm": 1.66079857586492, "language_loss": 0.73442322, "learning_rate": 2.717264894482836e-07, "loss": 0.75638098, "num_input_tokens_seen": 150402120, "step": 6962, "time_per_iteration": 2.483895778656006 }, { "auxiliary_loss_clip": 0.01158343, "auxiliary_loss_mlp": 0.01023985, "balance_loss_clip": 1.047894, "balance_loss_mlp": 1.01681209, "epoch": 0.8372512475199904, "flos": 19792708788960.0, "grad_norm": 1.8336297220756452, "language_loss": 0.80555785, "learning_rate": 2.7133459756620646e-07, "loss": 0.82738107, "num_input_tokens_seen": 150419315, "step": 6963, "time_per_iteration": 3.2132692337036133 }, { "auxiliary_loss_clip": 0.01149276, "auxiliary_loss_mlp": 0.01036189, "balance_loss_clip": 1.04581237, "balance_loss_mlp": 1.02887535, "epoch": 0.8373714904106295, "flos": 19391227447680.0, "grad_norm": 1.7182385035976422, "language_loss": 0.73670518, "learning_rate": 2.7094296792129733e-07, "loss": 0.75855982, "num_input_tokens_seen": 150438915, "step": 6964, "time_per_iteration": 2.4799294471740723 }, { "auxiliary_loss_clip": 0.01153445, "auxiliary_loss_mlp": 0.01023004, "balance_loss_clip": 1.04557538, "balance_loss_mlp": 1.01660228, "epoch": 0.8374917333012686, "flos": 14975938368480.0, "grad_norm": 1.9196929473112088, "language_loss": 0.7539016, "learning_rate": 2.7055160057296424e-07, "loss": 0.77566612, "num_input_tokens_seen": 150456155, "step": 6965, "time_per_iteration": 2.4287357330322266 }, { "auxiliary_loss_clip": 0.01126448, "auxiliary_loss_mlp": 0.01025344, "balance_loss_clip": 1.04357219, "balance_loss_mlp": 1.0177772, "epoch": 0.8376119761919076, "flos": 30331843819200.0, "grad_norm": 1.6688668054405331, "language_loss": 0.72086412, "learning_rate": 2.7016049558057896e-07, "loss": 0.74238205, "num_input_tokens_seen": 150478115, "step": 6966, "time_per_iteration": 2.601121187210083 }, { "auxiliary_loss_clip": 0.01155207, "auxiliary_loss_mlp": 0.01023185, "balance_loss_clip": 1.04917264, "balance_loss_mlp": 1.01594853, "epoch": 0.8377322190825467, "flos": 29423340356160.0, "grad_norm": 1.7768144938771722, "language_loss": 0.70894527, "learning_rate": 2.6976965300347074e-07, "loss": 0.73072922, "num_input_tokens_seen": 150500725, "step": 6967, "time_per_iteration": 2.523071050643921 }, { "auxiliary_loss_clip": 0.01134865, "auxiliary_loss_mlp": 0.01026225, "balance_loss_clip": 1.04255533, "balance_loss_mlp": 1.0194422, "epoch": 0.8378524619731859, "flos": 26687090796000.0, "grad_norm": 2.3128474938977583, "language_loss": 0.69342846, "learning_rate": 2.693790729009309e-07, "loss": 0.71503937, "num_input_tokens_seen": 150522335, "step": 6968, "time_per_iteration": 2.54447078704834 }, { "auxiliary_loss_clip": 0.01141101, "auxiliary_loss_mlp": 0.01025903, "balance_loss_clip": 1.04535222, "balance_loss_mlp": 1.01894128, "epoch": 0.8379727048638249, "flos": 20703870107040.0, "grad_norm": 2.044666230088868, "language_loss": 0.88073921, "learning_rate": 2.6898875533220946e-07, "loss": 0.9024092, "num_input_tokens_seen": 150541640, "step": 6969, "time_per_iteration": 2.4949989318847656 }, { "auxiliary_loss_clip": 0.01162344, "auxiliary_loss_mlp": 0.01023222, "balance_loss_clip": 1.04686952, "balance_loss_mlp": 1.01697516, "epoch": 0.838092947754464, "flos": 20084087895840.0, "grad_norm": 1.7546980613193495, "language_loss": 0.81711864, "learning_rate": 2.685987003565171e-07, "loss": 0.83897424, "num_input_tokens_seen": 150559680, "step": 6970, "time_per_iteration": 2.4532790184020996 }, { "auxiliary_loss_clip": 0.01123934, "auxiliary_loss_mlp": 0.01027683, "balance_loss_clip": 1.04791665, "balance_loss_mlp": 1.02076244, "epoch": 0.8382131906451031, "flos": 18113274183360.0, "grad_norm": 2.598615639229556, "language_loss": 0.75174105, "learning_rate": 2.6820890803302566e-07, "loss": 0.7732572, "num_input_tokens_seen": 150575205, "step": 6971, "time_per_iteration": 2.49640154838562 }, { "auxiliary_loss_clip": 0.01137659, "auxiliary_loss_mlp": 0.01021788, "balance_loss_clip": 1.04802632, "balance_loss_mlp": 1.01419163, "epoch": 0.8383334335357422, "flos": 17092657136160.0, "grad_norm": 2.384082024398171, "language_loss": 0.81668621, "learning_rate": 2.6781937842086557e-07, "loss": 0.83828062, "num_input_tokens_seen": 150593995, "step": 6972, "time_per_iteration": 3.2547686100006104 }, { "auxiliary_loss_clip": 0.01155855, "auxiliary_loss_mlp": 0.01027186, "balance_loss_clip": 1.04540086, "balance_loss_mlp": 1.02045703, "epoch": 0.8384536764263812, "flos": 20704732114080.0, "grad_norm": 1.8418623245036703, "language_loss": 0.67426854, "learning_rate": 2.6743011157912933e-07, "loss": 0.69609898, "num_input_tokens_seen": 150613715, "step": 6973, "time_per_iteration": 2.4806535243988037 }, { "auxiliary_loss_clip": 0.01109692, "auxiliary_loss_mlp": 0.01025257, "balance_loss_clip": 1.0389936, "balance_loss_mlp": 1.0179826, "epoch": 0.8385739193170204, "flos": 28986846367200.0, "grad_norm": 2.336985445658409, "language_loss": 0.64963055, "learning_rate": 2.6704110756686725e-07, "loss": 0.67098004, "num_input_tokens_seen": 150634540, "step": 6974, "time_per_iteration": 2.700810670852661 }, { "auxiliary_loss_clip": 0.01134625, "auxiliary_loss_mlp": 0.00761852, "balance_loss_clip": 1.04191852, "balance_loss_mlp": 1.00053239, "epoch": 0.8386941622076595, "flos": 23438072400480.0, "grad_norm": 1.6002931604977133, "language_loss": 0.83799791, "learning_rate": 2.6665236644309085e-07, "loss": 0.85696268, "num_input_tokens_seen": 150654850, "step": 6975, "time_per_iteration": 3.3822884559631348 }, { "auxiliary_loss_clip": 0.01153269, "auxiliary_loss_mlp": 0.0102177, "balance_loss_clip": 1.04506195, "balance_loss_mlp": 1.01513064, "epoch": 0.8388144050982985, "flos": 23002727754240.0, "grad_norm": 2.0274319051069587, "language_loss": 0.79989159, "learning_rate": 2.662638882667727e-07, "loss": 0.82164198, "num_input_tokens_seen": 150673790, "step": 6976, "time_per_iteration": 2.5504190921783447 }, { "auxiliary_loss_clip": 0.01171199, "auxiliary_loss_mlp": 0.01026927, "balance_loss_clip": 1.04775286, "balance_loss_mlp": 1.01969993, "epoch": 0.8389346479889377, "flos": 24280357765920.0, "grad_norm": 1.8098158370636166, "language_loss": 0.72844422, "learning_rate": 2.658756730968443e-07, "loss": 0.75042546, "num_input_tokens_seen": 150692255, "step": 6977, "time_per_iteration": 2.5105953216552734 }, { "auxiliary_loss_clip": 0.01145488, "auxiliary_loss_mlp": 0.01028659, "balance_loss_clip": 1.04761171, "balance_loss_mlp": 1.02202559, "epoch": 0.8390548908795767, "flos": 21215022679200.0, "grad_norm": 1.9724066338295658, "language_loss": 0.88276285, "learning_rate": 2.654877209921975e-07, "loss": 0.9045043, "num_input_tokens_seen": 150709790, "step": 6978, "time_per_iteration": 2.5274274349212646 }, { "auxiliary_loss_clip": 0.01116627, "auxiliary_loss_mlp": 0.0103211, "balance_loss_clip": 1.03914452, "balance_loss_mlp": 1.02401614, "epoch": 0.8391751337702158, "flos": 35627304268320.0, "grad_norm": 2.534791850545048, "language_loss": 0.6279372, "learning_rate": 2.651000320116843e-07, "loss": 0.64942455, "num_input_tokens_seen": 150730675, "step": 6979, "time_per_iteration": 3.493276834487915 }, { "auxiliary_loss_clip": 0.01124335, "auxiliary_loss_mlp": 0.00762348, "balance_loss_clip": 1.04362977, "balance_loss_mlp": 1.00059819, "epoch": 0.839295376660855, "flos": 21325232664480.0, "grad_norm": 1.8221881274860348, "language_loss": 0.75846726, "learning_rate": 2.647126062141163e-07, "loss": 0.77733409, "num_input_tokens_seen": 150749750, "step": 6980, "time_per_iteration": 2.586946725845337 }, { "auxiliary_loss_clip": 0.01142359, "auxiliary_loss_mlp": 0.01021864, "balance_loss_clip": 1.04253316, "balance_loss_mlp": 1.01541734, "epoch": 0.839415619551494, "flos": 18442539294720.0, "grad_norm": 1.777164263919217, "language_loss": 0.83953404, "learning_rate": 2.643254436582669e-07, "loss": 0.86117625, "num_input_tokens_seen": 150769240, "step": 6981, "time_per_iteration": 2.543046236038208 }, { "auxiliary_loss_clip": 0.01115881, "auxiliary_loss_mlp": 0.01020919, "balance_loss_clip": 1.0449419, "balance_loss_mlp": 1.01352847, "epoch": 0.8395358624421331, "flos": 23221962465120.0, "grad_norm": 2.613894537195128, "language_loss": 0.82297105, "learning_rate": 2.6393854440286743e-07, "loss": 0.84433907, "num_input_tokens_seen": 150788410, "step": 6982, "time_per_iteration": 2.5851826667785645 }, { "auxiliary_loss_clip": 0.01168017, "auxiliary_loss_mlp": 0.01020974, "balance_loss_clip": 1.04860556, "balance_loss_mlp": 1.01409888, "epoch": 0.8396561053327722, "flos": 24381660345120.0, "grad_norm": 1.877746116153726, "language_loss": 0.70401704, "learning_rate": 2.6355190850661045e-07, "loss": 0.72590691, "num_input_tokens_seen": 150805245, "step": 6983, "time_per_iteration": 2.4750893115997314 }, { "auxiliary_loss_clip": 0.01139964, "auxiliary_loss_mlp": 0.0102426, "balance_loss_clip": 1.04634309, "balance_loss_mlp": 1.01750994, "epoch": 0.8397763482234113, "flos": 22237758827040.0, "grad_norm": 2.5404674940250636, "language_loss": 0.86289674, "learning_rate": 2.631655360281486e-07, "loss": 0.88453895, "num_input_tokens_seen": 150824920, "step": 6984, "time_per_iteration": 2.526937961578369 }, { "auxiliary_loss_clip": 0.01156745, "auxiliary_loss_mlp": 0.00762401, "balance_loss_clip": 1.04798174, "balance_loss_mlp": 1.00057054, "epoch": 0.8398965911140504, "flos": 22163746749120.0, "grad_norm": 2.1107717072481127, "language_loss": 0.65860856, "learning_rate": 2.6277942702609323e-07, "loss": 0.67780006, "num_input_tokens_seen": 150844400, "step": 6985, "time_per_iteration": 2.5468502044677734 }, { "auxiliary_loss_clip": 0.01129736, "auxiliary_loss_mlp": 0.01025877, "balance_loss_clip": 1.0451473, "balance_loss_mlp": 1.01888525, "epoch": 0.8400168340046895, "flos": 21542779278240.0, "grad_norm": 1.967845661233219, "language_loss": 0.87476355, "learning_rate": 2.623935815590186e-07, "loss": 0.89631963, "num_input_tokens_seen": 150862780, "step": 6986, "time_per_iteration": 2.545387029647827 }, { "auxiliary_loss_clip": 0.01141188, "auxiliary_loss_mlp": 0.01023645, "balance_loss_clip": 1.0457015, "balance_loss_mlp": 1.01700544, "epoch": 0.8401370768953286, "flos": 22491970268640.0, "grad_norm": 2.122376349563058, "language_loss": 0.80844676, "learning_rate": 2.6200799968545516e-07, "loss": 0.83009505, "num_input_tokens_seen": 150883075, "step": 6987, "time_per_iteration": 2.58320951461792 }, { "auxiliary_loss_clip": 0.01041014, "auxiliary_loss_mlp": 0.01000987, "balance_loss_clip": 1.01040578, "balance_loss_mlp": 0.99986631, "epoch": 0.8402573197859676, "flos": 59238901866240.0, "grad_norm": 0.7874134674629126, "language_loss": 0.56378031, "learning_rate": 2.616226814638969e-07, "loss": 0.58420026, "num_input_tokens_seen": 150948180, "step": 6988, "time_per_iteration": 3.1621196269989014 }, { "auxiliary_loss_clip": 0.01138766, "auxiliary_loss_mlp": 0.01032072, "balance_loss_clip": 1.04625559, "balance_loss_mlp": 1.0252142, "epoch": 0.8403775626766068, "flos": 22674611985600.0, "grad_norm": 2.0323692106881137, "language_loss": 0.77173007, "learning_rate": 2.612376269527954e-07, "loss": 0.79343843, "num_input_tokens_seen": 150967885, "step": 6989, "time_per_iteration": 2.582479238510132 }, { "auxiliary_loss_clip": 0.01137944, "auxiliary_loss_mlp": 0.01029451, "balance_loss_clip": 1.04727483, "balance_loss_mlp": 1.02277207, "epoch": 0.8404978055672458, "flos": 19609707902400.0, "grad_norm": 2.5843481200037277, "language_loss": 0.6757586, "learning_rate": 2.608528362105635e-07, "loss": 0.69743252, "num_input_tokens_seen": 150987255, "step": 6990, "time_per_iteration": 3.238218307495117 }, { "auxiliary_loss_clip": 0.01126485, "auxiliary_loss_mlp": 0.01023473, "balance_loss_clip": 1.04096317, "balance_loss_mlp": 1.01721764, "epoch": 0.8406180484578849, "flos": 27526933808160.0, "grad_norm": 1.9615895195463855, "language_loss": 0.73094654, "learning_rate": 2.6046830929557374e-07, "loss": 0.75244612, "num_input_tokens_seen": 151006905, "step": 6991, "time_per_iteration": 2.658236265182495 }, { "auxiliary_loss_clip": 0.01123052, "auxiliary_loss_mlp": 0.01023875, "balance_loss_clip": 1.04455721, "balance_loss_mlp": 1.01631427, "epoch": 0.8407382913485241, "flos": 22127477007840.0, "grad_norm": 3.9064956954240437, "language_loss": 0.84990174, "learning_rate": 2.6008404626615776e-07, "loss": 0.87137103, "num_input_tokens_seen": 151025405, "step": 6992, "time_per_iteration": 2.5996949672698975 }, { "auxiliary_loss_clip": 0.0115776, "auxiliary_loss_mlp": 0.0103049, "balance_loss_clip": 1.04828584, "balance_loss_mlp": 1.02372479, "epoch": 0.8408585342391631, "flos": 13918476908640.0, "grad_norm": 2.3685153554379816, "language_loss": 0.73650503, "learning_rate": 2.597000471806092e-07, "loss": 0.75838751, "num_input_tokens_seen": 151041970, "step": 6993, "time_per_iteration": 2.511948585510254 }, { "auxiliary_loss_clip": 0.0113659, "auxiliary_loss_mlp": 0.01026567, "balance_loss_clip": 1.0470897, "balance_loss_mlp": 1.01930106, "epoch": 0.8409787771298022, "flos": 20187868745280.0, "grad_norm": 2.222137047464089, "language_loss": 0.73232734, "learning_rate": 2.593163120971793e-07, "loss": 0.75395888, "num_input_tokens_seen": 151060835, "step": 6994, "time_per_iteration": 2.5118870735168457 }, { "auxiliary_loss_clip": 0.01100743, "auxiliary_loss_mlp": 0.01022603, "balance_loss_clip": 1.03633356, "balance_loss_mlp": 1.01546872, "epoch": 0.8410990200204413, "flos": 23142526926240.0, "grad_norm": 2.158249008013952, "language_loss": 0.68768001, "learning_rate": 2.5893284107408165e-07, "loss": 0.70891345, "num_input_tokens_seen": 151078205, "step": 6995, "time_per_iteration": 2.640712022781372 }, { "auxiliary_loss_clip": 0.0111628, "auxiliary_loss_mlp": 0.01023863, "balance_loss_clip": 1.04514301, "balance_loss_mlp": 1.01682043, "epoch": 0.8412192629110804, "flos": 24027223833120.0, "grad_norm": 1.7799602969099562, "language_loss": 0.77695113, "learning_rate": 2.5854963416948726e-07, "loss": 0.7983526, "num_input_tokens_seen": 151100470, "step": 6996, "time_per_iteration": 2.675088882446289 }, { "auxiliary_loss_clip": 0.01105872, "auxiliary_loss_mlp": 0.01028665, "balance_loss_clip": 1.037009, "balance_loss_mlp": 1.0214622, "epoch": 0.8413395058017195, "flos": 25591707414720.0, "grad_norm": 1.9675202314413707, "language_loss": 0.69349205, "learning_rate": 2.5816669144152816e-07, "loss": 0.71483743, "num_input_tokens_seen": 151121650, "step": 6997, "time_per_iteration": 2.6788721084594727 }, { "auxiliary_loss_clip": 0.01060356, "auxiliary_loss_mlp": 0.01000255, "balance_loss_clip": 1.00538456, "balance_loss_mlp": 0.99898511, "epoch": 0.8414597486923585, "flos": 63635406375360.0, "grad_norm": 0.8523147919767519, "language_loss": 0.66397595, "learning_rate": 2.5778401294829777e-07, "loss": 0.684582, "num_input_tokens_seen": 151180390, "step": 6998, "time_per_iteration": 3.8614869117736816 }, { "auxiliary_loss_clip": 0.01150719, "auxiliary_loss_mlp": 0.0076175, "balance_loss_clip": 1.04551816, "balance_loss_mlp": 1.00054431, "epoch": 0.8415799915829977, "flos": 19098734915040.0, "grad_norm": 3.7657732465742817, "language_loss": 0.64817715, "learning_rate": 2.574015987478473e-07, "loss": 0.66730183, "num_input_tokens_seen": 151198520, "step": 6999, "time_per_iteration": 2.490067481994629 }, { "auxiliary_loss_clip": 0.01148268, "auxiliary_loss_mlp": 0.01030189, "balance_loss_clip": 1.04704571, "balance_loss_mlp": 1.02282178, "epoch": 0.8417002344736367, "flos": 19821615553440.0, "grad_norm": 1.9390562542304088, "language_loss": 0.86866695, "learning_rate": 2.570194488981887e-07, "loss": 0.89045149, "num_input_tokens_seen": 151215065, "step": 7000, "time_per_iteration": 2.5122361183166504 }, { "auxiliary_loss_clip": 0.01060364, "auxiliary_loss_mlp": 0.01000879, "balance_loss_clip": 1.00550389, "balance_loss_mlp": 0.99962181, "epoch": 0.8418204773642758, "flos": 62161528507200.0, "grad_norm": 0.841013446187068, "language_loss": 0.60353458, "learning_rate": 2.566375634572939e-07, "loss": 0.62414706, "num_input_tokens_seen": 151275705, "step": 7001, "time_per_iteration": 3.7803537845611572 }, { "auxiliary_loss_clip": 0.01130352, "auxiliary_loss_mlp": 0.01020557, "balance_loss_clip": 1.0423373, "balance_loss_mlp": 1.01344013, "epoch": 0.841940720254915, "flos": 17092908554880.0, "grad_norm": 1.862803483845238, "language_loss": 0.76385438, "learning_rate": 2.562559424830943e-07, "loss": 0.78536344, "num_input_tokens_seen": 151293665, "step": 7002, "time_per_iteration": 2.527277946472168 }, { "auxiliary_loss_clip": 0.01135183, "auxiliary_loss_mlp": 0.01023932, "balance_loss_clip": 1.04315948, "balance_loss_mlp": 1.01633215, "epoch": 0.842060963145554, "flos": 16283588570400.0, "grad_norm": 2.9564055984150315, "language_loss": 0.70528537, "learning_rate": 2.5587458603348256e-07, "loss": 0.7268765, "num_input_tokens_seen": 151310955, "step": 7003, "time_per_iteration": 2.513589859008789 }, { "auxiliary_loss_clip": 0.0111978, "auxiliary_loss_mlp": 0.010222, "balance_loss_clip": 1.04285371, "balance_loss_mlp": 1.01536047, "epoch": 0.8421812060361931, "flos": 21908242296960.0, "grad_norm": 1.7831599818354915, "language_loss": 0.84153557, "learning_rate": 2.554934941663085e-07, "loss": 0.86295533, "num_input_tokens_seen": 151328490, "step": 7004, "time_per_iteration": 3.2437305450439453 }, { "auxiliary_loss_clip": 0.01127138, "auxiliary_loss_mlp": 0.0102542, "balance_loss_clip": 1.04459047, "balance_loss_mlp": 1.01816607, "epoch": 0.8423014489268322, "flos": 27777697221600.0, "grad_norm": 7.4755422487238965, "language_loss": 0.73224568, "learning_rate": 2.5511266693938484e-07, "loss": 0.75377131, "num_input_tokens_seen": 151346950, "step": 7005, "time_per_iteration": 2.5815865993499756 }, { "auxiliary_loss_clip": 0.01139026, "auxiliary_loss_mlp": 0.01026091, "balance_loss_clip": 1.04678583, "balance_loss_mlp": 1.018471, "epoch": 0.8424216918174713, "flos": 25117614756960.0, "grad_norm": 1.4463926813686587, "language_loss": 0.7759347, "learning_rate": 2.547321044104822e-07, "loss": 0.79758584, "num_input_tokens_seen": 151368445, "step": 7006, "time_per_iteration": 2.5388495922088623 }, { "auxiliary_loss_clip": 0.01171869, "auxiliary_loss_mlp": 0.01025117, "balance_loss_clip": 1.04931355, "balance_loss_mlp": 1.01756823, "epoch": 0.8425419347081103, "flos": 24748452291360.0, "grad_norm": 1.6522775630440172, "language_loss": 0.76774335, "learning_rate": 2.5435180663733113e-07, "loss": 0.7897132, "num_input_tokens_seen": 151388745, "step": 7007, "time_per_iteration": 2.496305227279663 }, { "auxiliary_loss_clip": 0.01115393, "auxiliary_loss_mlp": 0.01028065, "balance_loss_clip": 1.04058814, "balance_loss_mlp": 1.02115655, "epoch": 0.8426621775987495, "flos": 24820919940000.0, "grad_norm": 2.34905099235097, "language_loss": 0.71989632, "learning_rate": 2.539717736776241e-07, "loss": 0.74133092, "num_input_tokens_seen": 151404970, "step": 7008, "time_per_iteration": 2.585958957672119 }, { "auxiliary_loss_clip": 0.01152577, "auxiliary_loss_mlp": 0.01029634, "balance_loss_clip": 1.04799366, "balance_loss_mlp": 1.02254093, "epoch": 0.8427824204893886, "flos": 23550078233760.0, "grad_norm": 1.6773090882666235, "language_loss": 0.76298428, "learning_rate": 2.535920055890097e-07, "loss": 0.78480637, "num_input_tokens_seen": 151426265, "step": 7009, "time_per_iteration": 2.4943273067474365 }, { "auxiliary_loss_clip": 0.01106472, "auxiliary_loss_mlp": 0.01028106, "balance_loss_clip": 1.03961432, "balance_loss_mlp": 1.02054203, "epoch": 0.8429026633800276, "flos": 16143861232320.0, "grad_norm": 4.722384828431493, "language_loss": 0.64596874, "learning_rate": 2.5321250242910006e-07, "loss": 0.66731453, "num_input_tokens_seen": 151444180, "step": 7010, "time_per_iteration": 2.546658754348755 }, { "auxiliary_loss_clip": 0.01167784, "auxiliary_loss_mlp": 0.01027106, "balance_loss_clip": 1.04901969, "balance_loss_mlp": 1.0201596, "epoch": 0.8430229062706668, "flos": 22198543895040.0, "grad_norm": 1.6382938237461329, "language_loss": 0.86505675, "learning_rate": 2.5283326425546493e-07, "loss": 0.88700575, "num_input_tokens_seen": 151463290, "step": 7011, "time_per_iteration": 2.464017152786255 }, { "auxiliary_loss_clip": 0.01120836, "auxiliary_loss_mlp": 0.01023515, "balance_loss_clip": 1.04572117, "balance_loss_mlp": 1.01663327, "epoch": 0.8431431491613058, "flos": 35330322115680.0, "grad_norm": 1.959808790342644, "language_loss": 0.69453633, "learning_rate": 2.5245429112563443e-07, "loss": 0.71597981, "num_input_tokens_seen": 151483965, "step": 7012, "time_per_iteration": 2.6580584049224854 }, { "auxiliary_loss_clip": 0.01153123, "auxiliary_loss_mlp": 0.01028135, "balance_loss_clip": 1.0470469, "balance_loss_mlp": 1.02100039, "epoch": 0.8432633920519449, "flos": 25812378804000.0, "grad_norm": 1.8553876720476663, "language_loss": 0.82184106, "learning_rate": 2.5207558309709865e-07, "loss": 0.84365368, "num_input_tokens_seen": 151503700, "step": 7013, "time_per_iteration": 2.530855894088745 }, { "auxiliary_loss_clip": 0.01034898, "auxiliary_loss_mlp": 0.00752949, "balance_loss_clip": 1.0056746, "balance_loss_mlp": 1.00021923, "epoch": 0.8433836349425841, "flos": 64959542461920.0, "grad_norm": 0.6539909856829615, "language_loss": 0.56341505, "learning_rate": 2.516971402273065e-07, "loss": 0.58129346, "num_input_tokens_seen": 151569765, "step": 7014, "time_per_iteration": 3.1620428562164307 }, { "auxiliary_loss_clip": 0.01139291, "auxiliary_loss_mlp": 0.01027739, "balance_loss_clip": 1.04329729, "balance_loss_mlp": 1.02121508, "epoch": 0.8435038778332231, "flos": 20229992951040.0, "grad_norm": 2.03040605131105, "language_loss": 0.68320853, "learning_rate": 2.513189625736687e-07, "loss": 0.70487881, "num_input_tokens_seen": 151586660, "step": 7015, "time_per_iteration": 3.217637300491333 }, { "auxiliary_loss_clip": 0.01131765, "auxiliary_loss_mlp": 0.01025894, "balance_loss_clip": 1.04412234, "balance_loss_mlp": 1.01869106, "epoch": 0.8436241207238622, "flos": 20992232189280.0, "grad_norm": 2.168340888653465, "language_loss": 0.7156564, "learning_rate": 2.509410501935534e-07, "loss": 0.73723292, "num_input_tokens_seen": 151602295, "step": 7016, "time_per_iteration": 2.5659327507019043 }, { "auxiliary_loss_clip": 0.01144519, "auxiliary_loss_mlp": 0.01031369, "balance_loss_clip": 1.04609787, "balance_loss_mlp": 1.02418327, "epoch": 0.8437443636145013, "flos": 14682260576160.0, "grad_norm": 2.1682910868580314, "language_loss": 0.75113666, "learning_rate": 2.5056340314429116e-07, "loss": 0.77289551, "num_input_tokens_seen": 151619760, "step": 7017, "time_per_iteration": 2.493039846420288 }, { "auxiliary_loss_clip": 0.01114179, "auxiliary_loss_mlp": 0.01019384, "balance_loss_clip": 1.04099035, "balance_loss_mlp": 1.01222301, "epoch": 0.8438646065051404, "flos": 21608817791040.0, "grad_norm": 2.349888136628024, "language_loss": 0.80353469, "learning_rate": 2.5018602148316904e-07, "loss": 0.82487035, "num_input_tokens_seen": 151635795, "step": 7018, "time_per_iteration": 2.5944833755493164 }, { "auxiliary_loss_clip": 0.01167116, "auxiliary_loss_mlp": 0.01025938, "balance_loss_clip": 1.04960251, "balance_loss_mlp": 1.01947725, "epoch": 0.8439848493957794, "flos": 23289940493760.0, "grad_norm": 2.112828492226817, "language_loss": 0.80220991, "learning_rate": 2.498089052674359e-07, "loss": 0.82414043, "num_input_tokens_seen": 151653770, "step": 7019, "time_per_iteration": 2.4628329277038574 }, { "auxiliary_loss_clip": 0.01155606, "auxiliary_loss_mlp": 0.01029381, "balance_loss_clip": 1.04792881, "balance_loss_mlp": 1.0217396, "epoch": 0.8441050922864186, "flos": 19719343216320.0, "grad_norm": 1.7789224217067643, "language_loss": 0.75229454, "learning_rate": 2.494320545543007e-07, "loss": 0.77414441, "num_input_tokens_seen": 151673340, "step": 7020, "time_per_iteration": 2.493598461151123 }, { "auxiliary_loss_clip": 0.01171742, "auxiliary_loss_mlp": 0.01024294, "balance_loss_clip": 1.04783416, "balance_loss_mlp": 1.01740408, "epoch": 0.8442253351770577, "flos": 21835271810880.0, "grad_norm": 1.8985625042461225, "language_loss": 0.66550404, "learning_rate": 2.490554694009308e-07, "loss": 0.68746442, "num_input_tokens_seen": 151694205, "step": 7021, "time_per_iteration": 2.46242356300354 }, { "auxiliary_loss_clip": 0.01156427, "auxiliary_loss_mlp": 0.01027498, "balance_loss_clip": 1.04428196, "balance_loss_mlp": 1.02053308, "epoch": 0.8443455780676967, "flos": 34346369896320.0, "grad_norm": 1.544618599714685, "language_loss": 0.78043658, "learning_rate": 2.4867914986445426e-07, "loss": 0.80227584, "num_input_tokens_seen": 151716595, "step": 7022, "time_per_iteration": 2.614370107650757 }, { "auxiliary_loss_clip": 0.01144223, "auxiliary_loss_mlp": 0.01023395, "balance_loss_clip": 1.04355264, "balance_loss_mlp": 1.01703191, "epoch": 0.8444658209583359, "flos": 48214605366240.0, "grad_norm": 4.02777141003019, "language_loss": 0.71144688, "learning_rate": 2.483030960019581e-07, "loss": 0.73312306, "num_input_tokens_seen": 151740525, "step": 7023, "time_per_iteration": 2.7583329677581787 }, { "auxiliary_loss_clip": 0.01018707, "auxiliary_loss_mlp": 0.01001361, "balance_loss_clip": 1.0065856, "balance_loss_mlp": 1.00002563, "epoch": 0.8445860638489749, "flos": 68484783007200.0, "grad_norm": 0.7341186920196853, "language_loss": 0.55506635, "learning_rate": 2.479273078704891e-07, "loss": 0.57526708, "num_input_tokens_seen": 151793890, "step": 7024, "time_per_iteration": 3.7676403522491455 }, { "auxiliary_loss_clip": 0.01013373, "auxiliary_loss_mlp": 0.01001646, "balance_loss_clip": 1.00812697, "balance_loss_mlp": 1.0005846, "epoch": 0.844706306739614, "flos": 62833341616800.0, "grad_norm": 0.7797035487447935, "language_loss": 0.64729393, "learning_rate": 2.475517855270552e-07, "loss": 0.66744411, "num_input_tokens_seen": 151853970, "step": 7025, "time_per_iteration": 3.1611404418945312 }, { "auxiliary_loss_clip": 0.01168968, "auxiliary_loss_mlp": 0.01028497, "balance_loss_clip": 1.04923308, "balance_loss_mlp": 1.02160931, "epoch": 0.8448265496302532, "flos": 14976117953280.0, "grad_norm": 2.2463839021266283, "language_loss": 0.72590715, "learning_rate": 2.4717652902862143e-07, "loss": 0.74788177, "num_input_tokens_seen": 151872945, "step": 7026, "time_per_iteration": 2.428927421569824 }, { "auxiliary_loss_clip": 0.01141163, "auxiliary_loss_mlp": 0.01025274, "balance_loss_clip": 1.04557073, "balance_loss_mlp": 1.01809812, "epoch": 0.8449467925208922, "flos": 23441269009920.0, "grad_norm": 1.7374177776605781, "language_loss": 0.81326336, "learning_rate": 2.4680153843211495e-07, "loss": 0.8349278, "num_input_tokens_seen": 151892875, "step": 7027, "time_per_iteration": 2.5486276149749756 }, { "auxiliary_loss_clip": 0.01140867, "auxiliary_loss_mlp": 0.01024209, "balance_loss_clip": 1.04855156, "balance_loss_mlp": 1.01699662, "epoch": 0.8450670354115313, "flos": 22748049392160.0, "grad_norm": 1.560181060602079, "language_loss": 0.72378337, "learning_rate": 2.464268137944212e-07, "loss": 0.74543405, "num_input_tokens_seen": 151914170, "step": 7028, "time_per_iteration": 3.6948511600494385 }, { "auxiliary_loss_clip": 0.01100344, "auxiliary_loss_mlp": 0.01030416, "balance_loss_clip": 1.04045761, "balance_loss_mlp": 1.02311468, "epoch": 0.8451872783021703, "flos": 29825575953600.0, "grad_norm": 2.0578224617845984, "language_loss": 0.78527009, "learning_rate": 2.46052355172385e-07, "loss": 0.80657768, "num_input_tokens_seen": 151932210, "step": 7029, "time_per_iteration": 2.6779050827026367 }, { "auxiliary_loss_clip": 0.01171651, "auxiliary_loss_mlp": 0.01025471, "balance_loss_clip": 1.04853845, "balance_loss_mlp": 1.01802063, "epoch": 0.8453075211928095, "flos": 21870033039840.0, "grad_norm": 1.8102065597856656, "language_loss": 0.7461791, "learning_rate": 2.456781626228128e-07, "loss": 0.76815033, "num_input_tokens_seen": 151951715, "step": 7030, "time_per_iteration": 2.4550981521606445 }, { "auxiliary_loss_clip": 0.010206, "auxiliary_loss_mlp": 0.00752496, "balance_loss_clip": 1.00639188, "balance_loss_mlp": 0.99995846, "epoch": 0.8454277640834486, "flos": 58751879102880.0, "grad_norm": 0.9259464150106157, "language_loss": 0.66321814, "learning_rate": 2.453042362024675e-07, "loss": 0.68094909, "num_input_tokens_seen": 152004960, "step": 7031, "time_per_iteration": 3.878434896469116 }, { "auxiliary_loss_clip": 0.01166938, "auxiliary_loss_mlp": 0.01025974, "balance_loss_clip": 1.04673719, "balance_loss_mlp": 1.01923847, "epoch": 0.8455480069740876, "flos": 27090080649600.0, "grad_norm": 1.4928159447889289, "language_loss": 0.73064339, "learning_rate": 2.449305759680751e-07, "loss": 0.75257254, "num_input_tokens_seen": 152026285, "step": 7032, "time_per_iteration": 2.519012689590454 }, { "auxiliary_loss_clip": 0.01126899, "auxiliary_loss_mlp": 0.01020465, "balance_loss_clip": 1.04699707, "balance_loss_mlp": 1.01378345, "epoch": 0.8456682498647268, "flos": 27198674371680.0, "grad_norm": 1.4234448914022804, "language_loss": 0.75292563, "learning_rate": 2.445571819763188e-07, "loss": 0.77439934, "num_input_tokens_seen": 152048585, "step": 7033, "time_per_iteration": 2.5932302474975586 }, { "auxiliary_loss_clip": 0.01168199, "auxiliary_loss_mlp": 0.01029951, "balance_loss_clip": 1.04876709, "balance_loss_mlp": 1.02276874, "epoch": 0.8457884927553658, "flos": 20631905295840.0, "grad_norm": 1.5808495065715136, "language_loss": 0.58152777, "learning_rate": 2.4418405428384227e-07, "loss": 0.60350931, "num_input_tokens_seen": 152068795, "step": 7034, "time_per_iteration": 2.4753828048706055 }, { "auxiliary_loss_clip": 0.01166618, "auxiliary_loss_mlp": 0.0076204, "balance_loss_clip": 1.04687953, "balance_loss_mlp": 1.00062275, "epoch": 0.8459087356460049, "flos": 15299025762720.0, "grad_norm": 1.851970490451691, "language_loss": 0.71560955, "learning_rate": 2.4381119294724864e-07, "loss": 0.73489618, "num_input_tokens_seen": 152086240, "step": 7035, "time_per_iteration": 2.4364678859710693 }, { "auxiliary_loss_clip": 0.0116931, "auxiliary_loss_mlp": 0.01024704, "balance_loss_clip": 1.04869056, "balance_loss_mlp": 1.01821613, "epoch": 0.846028978536644, "flos": 18843158628960.0, "grad_norm": 1.9732526224333047, "language_loss": 0.54094434, "learning_rate": 2.434385980231004e-07, "loss": 0.56288451, "num_input_tokens_seen": 152105080, "step": 7036, "time_per_iteration": 2.451925277709961 }, { "auxiliary_loss_clip": 0.01154717, "auxiliary_loss_mlp": 0.01022973, "balance_loss_clip": 1.04779232, "balance_loss_mlp": 1.01615405, "epoch": 0.8461492214272831, "flos": 52661746400640.0, "grad_norm": 1.5929275170696329, "language_loss": 0.65539038, "learning_rate": 2.4306626956792043e-07, "loss": 0.6771673, "num_input_tokens_seen": 152130025, "step": 7037, "time_per_iteration": 2.764892816543579 }, { "auxiliary_loss_clip": 0.01153286, "auxiliary_loss_mlp": 0.01023774, "balance_loss_clip": 1.04474699, "balance_loss_mlp": 1.01666927, "epoch": 0.8462694643179222, "flos": 18588408432960.0, "grad_norm": 1.6825946723407514, "language_loss": 0.75509274, "learning_rate": 2.4269420763819017e-07, "loss": 0.77686328, "num_input_tokens_seen": 152148070, "step": 7038, "time_per_iteration": 2.468301296234131 }, { "auxiliary_loss_clip": 0.01149661, "auxiliary_loss_mlp": 0.01027549, "balance_loss_clip": 1.04540658, "balance_loss_mlp": 1.02067089, "epoch": 0.8463897072085613, "flos": 24387083806080.0, "grad_norm": 2.538124332842769, "language_loss": 0.83906603, "learning_rate": 2.4232241229035223e-07, "loss": 0.86083817, "num_input_tokens_seen": 152165825, "step": 7039, "time_per_iteration": 2.4765443801879883 }, { "auxiliary_loss_clip": 0.01051846, "auxiliary_loss_mlp": 0.01000304, "balance_loss_clip": 1.00620818, "balance_loss_mlp": 0.99907058, "epoch": 0.8465099500992004, "flos": 68702150036160.0, "grad_norm": 0.7694503989999179, "language_loss": 0.56737578, "learning_rate": 2.419508835808064e-07, "loss": 0.58789718, "num_input_tokens_seen": 152222380, "step": 7040, "time_per_iteration": 3.0364580154418945 }, { "auxiliary_loss_clip": 0.01140574, "auxiliary_loss_mlp": 0.01025319, "balance_loss_clip": 1.04567027, "balance_loss_mlp": 1.01831269, "epoch": 0.8466301929898394, "flos": 13735727440800.0, "grad_norm": 2.401616716332486, "language_loss": 0.63121575, "learning_rate": 2.415796215659134e-07, "loss": 0.65287471, "num_input_tokens_seen": 152239085, "step": 7041, "time_per_iteration": 3.241436719894409 }, { "auxiliary_loss_clip": 0.01128627, "auxiliary_loss_mlp": 0.01024569, "balance_loss_clip": 1.03967142, "balance_loss_mlp": 1.01750636, "epoch": 0.8467504358804786, "flos": 19241263776000.0, "grad_norm": 1.9911799234325271, "language_loss": 0.76314551, "learning_rate": 2.412086263019939e-07, "loss": 0.78467745, "num_input_tokens_seen": 152257110, "step": 7042, "time_per_iteration": 2.549579620361328 }, { "auxiliary_loss_clip": 0.01163775, "auxiliary_loss_mlp": 0.01027114, "balance_loss_clip": 1.04758763, "balance_loss_mlp": 1.02065921, "epoch": 0.8468706787711177, "flos": 21324119238720.0, "grad_norm": 1.7228892641555993, "language_loss": 0.79792881, "learning_rate": 2.408378978453276e-07, "loss": 0.81983769, "num_input_tokens_seen": 152277230, "step": 7043, "time_per_iteration": 2.462325096130371 }, { "auxiliary_loss_clip": 0.01052481, "auxiliary_loss_mlp": 0.01002286, "balance_loss_clip": 1.00666404, "balance_loss_mlp": 1.00103414, "epoch": 0.8469909216617567, "flos": 64877449068000.0, "grad_norm": 0.82279592501833, "language_loss": 0.64028811, "learning_rate": 2.404674362521533e-07, "loss": 0.66083586, "num_input_tokens_seen": 152335725, "step": 7044, "time_per_iteration": 2.986875295639038 }, { "auxiliary_loss_clip": 0.01152072, "auxiliary_loss_mlp": 0.01025152, "balance_loss_clip": 1.04762232, "balance_loss_mlp": 1.01858926, "epoch": 0.8471111645523959, "flos": 19280586458880.0, "grad_norm": 2.4313976118514526, "language_loss": 0.74947834, "learning_rate": 2.4009724157866997e-07, "loss": 0.77125049, "num_input_tokens_seen": 152352785, "step": 7045, "time_per_iteration": 2.4688668251037598 }, { "auxiliary_loss_clip": 0.01166852, "auxiliary_loss_mlp": 0.01025162, "balance_loss_clip": 1.04826045, "balance_loss_mlp": 1.01840568, "epoch": 0.8472314074430349, "flos": 22015830344160.0, "grad_norm": 2.426571957679556, "language_loss": 0.7666682, "learning_rate": 2.3972731388103564e-07, "loss": 0.78858835, "num_input_tokens_seen": 152371265, "step": 7046, "time_per_iteration": 2.44388484954834 }, { "auxiliary_loss_clip": 0.01003061, "auxiliary_loss_mlp": 0.01002658, "balance_loss_clip": 1.00671875, "balance_loss_mlp": 1.0014714, "epoch": 0.847351650333674, "flos": 57882590571840.0, "grad_norm": 0.8108656877654212, "language_loss": 0.62429368, "learning_rate": 2.393576532153687e-07, "loss": 0.64435089, "num_input_tokens_seen": 152435050, "step": 7047, "time_per_iteration": 3.3405401706695557 }, { "auxiliary_loss_clip": 0.01048467, "auxiliary_loss_mlp": 0.01000001, "balance_loss_clip": 1.00597322, "balance_loss_mlp": 0.99882078, "epoch": 0.8474718932243132, "flos": 41284247341920.0, "grad_norm": 1.454864138773207, "language_loss": 0.57851058, "learning_rate": 2.389882596377453e-07, "loss": 0.59899521, "num_input_tokens_seen": 152489315, "step": 7048, "time_per_iteration": 3.6980485916137695 }, { "auxiliary_loss_clip": 0.01166857, "auxiliary_loss_mlp": 0.01023259, "balance_loss_clip": 1.04698741, "balance_loss_mlp": 1.01618695, "epoch": 0.8475921361149522, "flos": 38180912111520.0, "grad_norm": 1.7848179848338657, "language_loss": 0.76332289, "learning_rate": 2.386191332042031e-07, "loss": 0.78522408, "num_input_tokens_seen": 152511210, "step": 7049, "time_per_iteration": 2.6544761657714844 }, { "auxiliary_loss_clip": 0.0117194, "auxiliary_loss_mlp": 0.01027655, "balance_loss_clip": 1.04911137, "balance_loss_mlp": 1.02059829, "epoch": 0.8477123790055913, "flos": 25375058724960.0, "grad_norm": 1.6816569508661747, "language_loss": 0.72470737, "learning_rate": 2.3825027397073794e-07, "loss": 0.74670327, "num_input_tokens_seen": 152531685, "step": 7050, "time_per_iteration": 3.4835293292999268 }, { "auxiliary_loss_clip": 0.01151355, "auxiliary_loss_mlp": 0.01029965, "balance_loss_clip": 1.04752374, "balance_loss_mlp": 1.02312589, "epoch": 0.8478326218962304, "flos": 30225189612960.0, "grad_norm": 2.38388189165703, "language_loss": 0.66949707, "learning_rate": 2.3788168199330515e-07, "loss": 0.69131029, "num_input_tokens_seen": 152553245, "step": 7051, "time_per_iteration": 2.610912799835205 }, { "auxiliary_loss_clip": 0.01122866, "auxiliary_loss_mlp": 0.01023862, "balance_loss_clip": 1.03968787, "balance_loss_mlp": 1.01734698, "epoch": 0.8479528647868695, "flos": 38213805658560.0, "grad_norm": 1.5906576741900091, "language_loss": 0.72454786, "learning_rate": 2.3751335732782074e-07, "loss": 0.74601513, "num_input_tokens_seen": 152574505, "step": 7052, "time_per_iteration": 2.695289373397827 }, { "auxiliary_loss_clip": 0.01153103, "auxiliary_loss_mlp": 0.01024958, "balance_loss_clip": 1.04837751, "balance_loss_mlp": 1.01808882, "epoch": 0.8480731076775085, "flos": 20957794212960.0, "grad_norm": 2.7375451351181086, "language_loss": 0.79026854, "learning_rate": 2.371453000301582e-07, "loss": 0.81204921, "num_input_tokens_seen": 152593190, "step": 7053, "time_per_iteration": 3.2517833709716797 }, { "auxiliary_loss_clip": 0.01121583, "auxiliary_loss_mlp": 0.0102322, "balance_loss_clip": 1.0433532, "balance_loss_mlp": 1.01697922, "epoch": 0.8481933505681477, "flos": 32596514908800.0, "grad_norm": 2.0904779694794327, "language_loss": 0.7443862, "learning_rate": 2.3677751015615222e-07, "loss": 0.76583421, "num_input_tokens_seen": 152615265, "step": 7054, "time_per_iteration": 2.6487209796905518 }, { "auxiliary_loss_clip": 0.01127599, "auxiliary_loss_mlp": 0.01027433, "balance_loss_clip": 1.04079497, "balance_loss_mlp": 1.0210048, "epoch": 0.8483135934587868, "flos": 20741181440160.0, "grad_norm": 1.79210586866907, "language_loss": 0.85327995, "learning_rate": 2.3640998776159593e-07, "loss": 0.87483031, "num_input_tokens_seen": 152632770, "step": 7055, "time_per_iteration": 2.5272974967956543 }, { "auxiliary_loss_clip": 0.01140506, "auxiliary_loss_mlp": 0.01025669, "balance_loss_clip": 1.04626739, "balance_loss_mlp": 1.01980042, "epoch": 0.8484338363494258, "flos": 21653061097440.0, "grad_norm": 1.850010923185436, "language_loss": 0.81010437, "learning_rate": 2.3604273290224253e-07, "loss": 0.83176613, "num_input_tokens_seen": 152653485, "step": 7056, "time_per_iteration": 3.25948429107666 }, { "auxiliary_loss_clip": 0.01143433, "auxiliary_loss_mlp": 0.01030162, "balance_loss_clip": 1.04644656, "balance_loss_mlp": 1.02311051, "epoch": 0.848554079240065, "flos": 15013968040800.0, "grad_norm": 2.3924171226548507, "language_loss": 0.74855709, "learning_rate": 2.356757456338039e-07, "loss": 0.77029306, "num_input_tokens_seen": 152670970, "step": 7057, "time_per_iteration": 2.500509023666382 }, { "auxiliary_loss_clip": 0.01038616, "auxiliary_loss_mlp": 0.0100139, "balance_loss_clip": 1.0084343, "balance_loss_mlp": 1.00013852, "epoch": 0.848674322130704, "flos": 68060464867680.0, "grad_norm": 0.7499337662702885, "language_loss": 0.59116268, "learning_rate": 2.3530902601195147e-07, "loss": 0.61156273, "num_input_tokens_seen": 152739460, "step": 7058, "time_per_iteration": 3.236363172531128 }, { "auxiliary_loss_clip": 0.01155097, "auxiliary_loss_mlp": 0.01027564, "balance_loss_clip": 1.0481956, "balance_loss_mlp": 1.01955318, "epoch": 0.8487945650213431, "flos": 18475792011360.0, "grad_norm": 2.2386395437252724, "language_loss": 0.78602397, "learning_rate": 2.34942574092317e-07, "loss": 0.8078506, "num_input_tokens_seen": 152754710, "step": 7059, "time_per_iteration": 2.457106351852417 }, { "auxiliary_loss_clip": 0.01158674, "auxiliary_loss_mlp": 0.01025752, "balance_loss_clip": 1.04729176, "balance_loss_mlp": 1.01845074, "epoch": 0.8489148079119821, "flos": 23473192799040.0, "grad_norm": 1.8031325974923436, "language_loss": 0.76765889, "learning_rate": 2.3457638993049045e-07, "loss": 0.78950316, "num_input_tokens_seen": 152772700, "step": 7060, "time_per_iteration": 2.498648166656494 }, { "auxiliary_loss_clip": 0.01101932, "auxiliary_loss_mlp": 0.01023312, "balance_loss_clip": 1.04560423, "balance_loss_mlp": 1.01602542, "epoch": 0.8490350508026213, "flos": 19937608169280.0, "grad_norm": 1.9131472798668399, "language_loss": 0.64221555, "learning_rate": 2.3421047358202252e-07, "loss": 0.66346794, "num_input_tokens_seen": 152791550, "step": 7061, "time_per_iteration": 2.6109559535980225 }, { "auxiliary_loss_clip": 0.01154671, "auxiliary_loss_mlp": 0.01026399, "balance_loss_clip": 1.04598379, "balance_loss_mlp": 1.01970887, "epoch": 0.8491552936932604, "flos": 24279962679360.0, "grad_norm": 2.201110512072994, "language_loss": 0.82969427, "learning_rate": 2.3384482510242144e-07, "loss": 0.85150498, "num_input_tokens_seen": 152809410, "step": 7062, "time_per_iteration": 2.4936928749084473 }, { "auxiliary_loss_clip": 0.01172273, "auxiliary_loss_mlp": 0.01025965, "balance_loss_clip": 1.04846931, "balance_loss_mlp": 1.01906562, "epoch": 0.8492755365838994, "flos": 22522529213280.0, "grad_norm": 1.929571141735816, "language_loss": 0.76930201, "learning_rate": 2.3347944454715575e-07, "loss": 0.79128438, "num_input_tokens_seen": 152825800, "step": 7063, "time_per_iteration": 2.4389758110046387 }, { "auxiliary_loss_clip": 0.01170652, "auxiliary_loss_mlp": 0.01027427, "balance_loss_clip": 1.04818439, "balance_loss_mlp": 1.02011991, "epoch": 0.8493957794745386, "flos": 26980445335680.0, "grad_norm": 1.7854957510735099, "language_loss": 0.67305672, "learning_rate": 2.331143319716542e-07, "loss": 0.69503754, "num_input_tokens_seen": 152845330, "step": 7064, "time_per_iteration": 2.497288703918457 }, { "auxiliary_loss_clip": 0.01128353, "auxiliary_loss_mlp": 0.01026586, "balance_loss_clip": 1.04564452, "balance_loss_mlp": 1.01953793, "epoch": 0.8495160223651776, "flos": 29861989362720.0, "grad_norm": 2.679000292972447, "language_loss": 0.66058016, "learning_rate": 2.3274948743130363e-07, "loss": 0.68212962, "num_input_tokens_seen": 152865165, "step": 7065, "time_per_iteration": 2.608207941055298 }, { "auxiliary_loss_clip": 0.01165847, "auxiliary_loss_mlp": 0.01025843, "balance_loss_clip": 1.04487276, "balance_loss_mlp": 1.01905107, "epoch": 0.8496362652558167, "flos": 23075446821600.0, "grad_norm": 1.70883014815383, "language_loss": 0.79476929, "learning_rate": 2.3238491098145085e-07, "loss": 0.81668615, "num_input_tokens_seen": 152884695, "step": 7066, "time_per_iteration": 2.4680347442626953 }, { "auxiliary_loss_clip": 0.01154763, "auxiliary_loss_mlp": 0.01025044, "balance_loss_clip": 1.04708576, "balance_loss_mlp": 1.01787329, "epoch": 0.8497565081464559, "flos": 14609110505280.0, "grad_norm": 2.315435917772889, "language_loss": 0.73332107, "learning_rate": 2.3202060267740141e-07, "loss": 0.75511909, "num_input_tokens_seen": 152902220, "step": 7067, "time_per_iteration": 3.225515842437744 }, { "auxiliary_loss_clip": 0.01104348, "auxiliary_loss_mlp": 0.01028222, "balance_loss_clip": 1.03771734, "balance_loss_mlp": 1.02144837, "epoch": 0.8498767510370949, "flos": 21136449147360.0, "grad_norm": 2.217538919451258, "language_loss": 0.77227807, "learning_rate": 2.3165656257442044e-07, "loss": 0.79360378, "num_input_tokens_seen": 152920740, "step": 7068, "time_per_iteration": 2.5729799270629883 }, { "auxiliary_loss_clip": 0.01149954, "auxiliary_loss_mlp": 0.01027856, "balance_loss_clip": 1.04621625, "balance_loss_mlp": 1.0212698, "epoch": 0.849996993927734, "flos": 23654541505440.0, "grad_norm": 2.0293368312076323, "language_loss": 0.89936811, "learning_rate": 2.31292790727734e-07, "loss": 0.92114621, "num_input_tokens_seen": 152938305, "step": 7069, "time_per_iteration": 2.4925668239593506 }, { "auxiliary_loss_clip": 0.01165259, "auxiliary_loss_mlp": 0.01023948, "balance_loss_clip": 1.04608023, "balance_loss_mlp": 1.01761246, "epoch": 0.8501172368183731, "flos": 20558072802720.0, "grad_norm": 2.258199418658607, "language_loss": 0.80490279, "learning_rate": 2.3092928719252392e-07, "loss": 0.82679492, "num_input_tokens_seen": 152956705, "step": 7070, "time_per_iteration": 2.4548914432525635 }, { "auxiliary_loss_clip": 0.01152539, "auxiliary_loss_mlp": 0.01023285, "balance_loss_clip": 1.04548073, "balance_loss_mlp": 1.01645482, "epoch": 0.8502374797090122, "flos": 22272627806880.0, "grad_norm": 1.9420860264519866, "language_loss": 0.78764719, "learning_rate": 2.3056605202393475e-07, "loss": 0.80940545, "num_input_tokens_seen": 152974265, "step": 7071, "time_per_iteration": 2.4990124702453613 }, { "auxiliary_loss_clip": 0.01148201, "auxiliary_loss_mlp": 0.00762452, "balance_loss_clip": 1.04285181, "balance_loss_mlp": 1.00063968, "epoch": 0.8503577225996513, "flos": 23659821298560.0, "grad_norm": 1.7708494021887742, "language_loss": 0.66647291, "learning_rate": 2.3020308527706888e-07, "loss": 0.68557942, "num_input_tokens_seen": 152993680, "step": 7072, "time_per_iteration": 2.54150652885437 }, { "auxiliary_loss_clip": 0.01142597, "auxiliary_loss_mlp": 0.01032776, "balance_loss_clip": 1.04229581, "balance_loss_mlp": 1.02648771, "epoch": 0.8504779654902904, "flos": 26758516852800.0, "grad_norm": 1.6838903103860472, "language_loss": 0.89132309, "learning_rate": 2.2984038700698715e-07, "loss": 0.91307682, "num_input_tokens_seen": 153012990, "step": 7073, "time_per_iteration": 2.5689377784729004 }, { "auxiliary_loss_clip": 0.01151963, "auxiliary_loss_mlp": 0.01028326, "balance_loss_clip": 1.04774654, "balance_loss_mlp": 1.02143848, "epoch": 0.8505982083809295, "flos": 26468251171680.0, "grad_norm": 1.6327365322756593, "language_loss": 0.79069138, "learning_rate": 2.2947795726871222e-07, "loss": 0.81249422, "num_input_tokens_seen": 153034015, "step": 7074, "time_per_iteration": 2.5358924865722656 }, { "auxiliary_loss_clip": 0.01155988, "auxiliary_loss_mlp": 0.00761835, "balance_loss_clip": 1.05198145, "balance_loss_mlp": 1.00051808, "epoch": 0.8507184512715685, "flos": 20303394440640.0, "grad_norm": 1.6647383766965425, "language_loss": 0.85617971, "learning_rate": 2.2911579611722253e-07, "loss": 0.87535793, "num_input_tokens_seen": 153053160, "step": 7075, "time_per_iteration": 2.475095748901367 }, { "auxiliary_loss_clip": 0.01131686, "auxiliary_loss_mlp": 0.01022954, "balance_loss_clip": 1.04366207, "balance_loss_mlp": 1.01630807, "epoch": 0.8508386941622077, "flos": 19025189757600.0, "grad_norm": 1.7863361691408821, "language_loss": 0.87216127, "learning_rate": 2.2875390360745905e-07, "loss": 0.89370769, "num_input_tokens_seen": 153072565, "step": 7076, "time_per_iteration": 3.2478883266448975 }, { "auxiliary_loss_clip": 0.01131872, "auxiliary_loss_mlp": 0.01027262, "balance_loss_clip": 1.04462552, "balance_loss_mlp": 1.02028847, "epoch": 0.8509589370528468, "flos": 16433408574240.0, "grad_norm": 1.68130640208696, "language_loss": 0.77668953, "learning_rate": 2.2839227979432008e-07, "loss": 0.79828089, "num_input_tokens_seen": 153090215, "step": 7077, "time_per_iteration": 2.4980931282043457 }, { "auxiliary_loss_clip": 0.01143541, "auxiliary_loss_mlp": 0.01018236, "balance_loss_clip": 1.04580677, "balance_loss_mlp": 1.01131642, "epoch": 0.8510791799434858, "flos": 18259717992960.0, "grad_norm": 1.8863683714988955, "language_loss": 0.85066235, "learning_rate": 2.2803092473266373e-07, "loss": 0.87228012, "num_input_tokens_seen": 153107740, "step": 7078, "time_per_iteration": 2.5022873878479004 }, { "auxiliary_loss_clip": 0.01169914, "auxiliary_loss_mlp": 0.01030145, "balance_loss_clip": 1.04932952, "balance_loss_mlp": 1.02307248, "epoch": 0.851199422834125, "flos": 23441376760800.0, "grad_norm": 5.412897048374033, "language_loss": 0.86789358, "learning_rate": 2.2766983847730724e-07, "loss": 0.88989413, "num_input_tokens_seen": 153127410, "step": 7079, "time_per_iteration": 3.234795331954956 }, { "auxiliary_loss_clip": 0.01136992, "auxiliary_loss_mlp": 0.01031439, "balance_loss_clip": 1.04420161, "balance_loss_mlp": 1.0243578, "epoch": 0.851319665724764, "flos": 16289407117920.0, "grad_norm": 2.012300679421904, "language_loss": 0.66459215, "learning_rate": 2.2730902108302663e-07, "loss": 0.68627656, "num_input_tokens_seen": 153144325, "step": 7080, "time_per_iteration": 2.53295636177063 }, { "auxiliary_loss_clip": 0.01132516, "auxiliary_loss_mlp": 0.01031086, "balance_loss_clip": 1.04159009, "balance_loss_mlp": 1.02365959, "epoch": 0.8514399086154031, "flos": 18989351019840.0, "grad_norm": 1.5411586730366107, "language_loss": 0.684044, "learning_rate": 2.269484726045583e-07, "loss": 0.70568001, "num_input_tokens_seen": 153163240, "step": 7081, "time_per_iteration": 2.500181198120117 }, { "auxiliary_loss_clip": 0.01131161, "auxiliary_loss_mlp": 0.01029076, "balance_loss_clip": 1.0448668, "balance_loss_mlp": 1.022439, "epoch": 0.8515601515060423, "flos": 24571198118400.0, "grad_norm": 2.705416961809216, "language_loss": 0.78938174, "learning_rate": 2.2658819309659672e-07, "loss": 0.81098413, "num_input_tokens_seen": 153183440, "step": 7082, "time_per_iteration": 2.5971765518188477 }, { "auxiliary_loss_clip": 0.01138877, "auxiliary_loss_mlp": 0.01023249, "balance_loss_clip": 1.04725146, "balance_loss_mlp": 1.01662135, "epoch": 0.8516803943966813, "flos": 19529446273440.0, "grad_norm": 1.8388515312631155, "language_loss": 0.84725296, "learning_rate": 2.2622818261379706e-07, "loss": 0.86887419, "num_input_tokens_seen": 153200460, "step": 7083, "time_per_iteration": 3.246032953262329 }, { "auxiliary_loss_clip": 0.01140046, "auxiliary_loss_mlp": 0.01033507, "balance_loss_clip": 1.04563665, "balance_loss_mlp": 1.02626538, "epoch": 0.8518006372873204, "flos": 20265795771840.0, "grad_norm": 1.8292129512284818, "language_loss": 0.74687016, "learning_rate": 2.2586844121077142e-07, "loss": 0.76860571, "num_input_tokens_seen": 153218970, "step": 7084, "time_per_iteration": 2.5380172729492188 }, { "auxiliary_loss_clip": 0.01108141, "auxiliary_loss_mlp": 0.01023473, "balance_loss_clip": 1.03964138, "balance_loss_mlp": 1.01637363, "epoch": 0.8519208801779595, "flos": 24133231534080.0, "grad_norm": 1.7318220277672465, "language_loss": 0.71978718, "learning_rate": 2.2550896894209215e-07, "loss": 0.74110329, "num_input_tokens_seen": 153238485, "step": 7085, "time_per_iteration": 2.603015661239624 }, { "auxiliary_loss_clip": 0.01008886, "auxiliary_loss_mlp": 0.01003287, "balance_loss_clip": 1.00687623, "balance_loss_mlp": 1.00189257, "epoch": 0.8520411230685986, "flos": 63035234824320.0, "grad_norm": 0.6962720196280695, "language_loss": 0.566535, "learning_rate": 2.2514976586229184e-07, "loss": 0.58665669, "num_input_tokens_seen": 153306430, "step": 7086, "time_per_iteration": 3.4813992977142334 }, { "auxiliary_loss_clip": 0.0105068, "auxiliary_loss_mlp": 0.01001852, "balance_loss_clip": 1.00543439, "balance_loss_mlp": 1.00059462, "epoch": 0.8521613659592376, "flos": 65836876392000.0, "grad_norm": 0.7554389554304445, "language_loss": 0.54729104, "learning_rate": 2.247908320258609e-07, "loss": 0.56781638, "num_input_tokens_seen": 153366520, "step": 7087, "time_per_iteration": 3.361710786819458 }, { "auxiliary_loss_clip": 0.01106124, "auxiliary_loss_mlp": 0.01026727, "balance_loss_clip": 1.04317284, "balance_loss_mlp": 1.01940799, "epoch": 0.8522816088498768, "flos": 23112327151200.0, "grad_norm": 3.04114186794946, "language_loss": 0.79508066, "learning_rate": 2.2443216748724914e-07, "loss": 0.81640911, "num_input_tokens_seen": 153387230, "step": 7088, "time_per_iteration": 2.5824625492095947 }, { "auxiliary_loss_clip": 0.0115617, "auxiliary_loss_mlp": 0.00761965, "balance_loss_clip": 1.0475136, "balance_loss_mlp": 1.00049257, "epoch": 0.8524018517405159, "flos": 31758144492000.0, "grad_norm": 1.79918811719086, "language_loss": 0.74264753, "learning_rate": 2.2407377230086588e-07, "loss": 0.76182884, "num_input_tokens_seen": 153409585, "step": 7089, "time_per_iteration": 2.5864031314849854 }, { "auxiliary_loss_clip": 0.01127262, "auxiliary_loss_mlp": 0.01029166, "balance_loss_clip": 1.04748368, "balance_loss_mlp": 1.02181947, "epoch": 0.8525220946311549, "flos": 18690321600480.0, "grad_norm": 2.0669188344426295, "language_loss": 0.83405566, "learning_rate": 2.23715646521079e-07, "loss": 0.85561991, "num_input_tokens_seen": 153427105, "step": 7090, "time_per_iteration": 2.5616607666015625 }, { "auxiliary_loss_clip": 0.01157853, "auxiliary_loss_mlp": 0.00762949, "balance_loss_clip": 1.0465548, "balance_loss_mlp": 1.0005722, "epoch": 0.852642337521794, "flos": 21793219439040.0, "grad_norm": 1.9306140807591348, "language_loss": 0.84175915, "learning_rate": 2.2335779020221724e-07, "loss": 0.86096716, "num_input_tokens_seen": 153443725, "step": 7091, "time_per_iteration": 2.501946210861206 }, { "auxiliary_loss_clip": 0.01061144, "auxiliary_loss_mlp": 0.01003571, "balance_loss_clip": 1.02260196, "balance_loss_mlp": 1.00238514, "epoch": 0.8527625804124331, "flos": 69040142968800.0, "grad_norm": 0.8017623363312776, "language_loss": 0.56496882, "learning_rate": 2.2300020339856497e-07, "loss": 0.58561599, "num_input_tokens_seen": 153506410, "step": 7092, "time_per_iteration": 3.189984083175659 }, { "auxiliary_loss_clip": 0.01138018, "auxiliary_loss_mlp": 0.01021694, "balance_loss_clip": 1.04598844, "balance_loss_mlp": 1.01499748, "epoch": 0.8528828233030722, "flos": 26979403743840.0, "grad_norm": 2.630182991517171, "language_loss": 0.78510928, "learning_rate": 2.2264288616436966e-07, "loss": 0.80670643, "num_input_tokens_seen": 153526665, "step": 7093, "time_per_iteration": 2.5566320419311523 }, { "auxiliary_loss_clip": 0.01134659, "auxiliary_loss_mlp": 0.01025549, "balance_loss_clip": 1.0448401, "balance_loss_mlp": 1.01825321, "epoch": 0.8530030661937112, "flos": 17487601590720.0, "grad_norm": 2.557531443337997, "language_loss": 0.72323483, "learning_rate": 2.222858385538351e-07, "loss": 0.74483693, "num_input_tokens_seen": 153543465, "step": 7094, "time_per_iteration": 3.2592034339904785 }, { "auxiliary_loss_clip": 0.01150676, "auxiliary_loss_mlp": 0.01020562, "balance_loss_clip": 1.04463887, "balance_loss_mlp": 1.01415443, "epoch": 0.8531233090843504, "flos": 22160801558400.0, "grad_norm": 2.1417506579832564, "language_loss": 0.68069911, "learning_rate": 2.2192906062112527e-07, "loss": 0.70241147, "num_input_tokens_seen": 153563340, "step": 7095, "time_per_iteration": 2.5046420097351074 }, { "auxiliary_loss_clip": 0.01167231, "auxiliary_loss_mlp": 0.01024952, "balance_loss_clip": 1.04640555, "balance_loss_mlp": 1.01807404, "epoch": 0.8532435519749895, "flos": 37635393396960.0, "grad_norm": 1.5174751658399204, "language_loss": 0.70187867, "learning_rate": 2.2157255242036377e-07, "loss": 0.72380054, "num_input_tokens_seen": 153587005, "step": 7096, "time_per_iteration": 2.734325885772705 }, { "auxiliary_loss_clip": 0.01123228, "auxiliary_loss_mlp": 0.01025797, "balance_loss_clip": 1.04388833, "balance_loss_mlp": 1.01945519, "epoch": 0.8533637948656285, "flos": 21398167233600.0, "grad_norm": 1.5758679745169626, "language_loss": 0.74438405, "learning_rate": 2.2121631400563135e-07, "loss": 0.76587427, "num_input_tokens_seen": 153606835, "step": 7097, "time_per_iteration": 2.539001941680908 }, { "auxiliary_loss_clip": 0.0104816, "auxiliary_loss_mlp": 0.01002867, "balance_loss_clip": 1.00803208, "balance_loss_mlp": 1.00159764, "epoch": 0.8534840377562677, "flos": 53345131159680.0, "grad_norm": 0.7591960613144582, "language_loss": 0.52959335, "learning_rate": 2.208603454309701e-07, "loss": 0.55010366, "num_input_tokens_seen": 153664925, "step": 7098, "time_per_iteration": 3.0761425495147705 }, { "auxiliary_loss_clip": 0.01110697, "auxiliary_loss_mlp": 0.0103169, "balance_loss_clip": 1.04247487, "balance_loss_mlp": 1.02438283, "epoch": 0.8536042806469067, "flos": 20814151926240.0, "grad_norm": 1.923884782737813, "language_loss": 0.71260297, "learning_rate": 2.2050464675037994e-07, "loss": 0.73402679, "num_input_tokens_seen": 153683550, "step": 7099, "time_per_iteration": 2.5787477493286133 }, { "auxiliary_loss_clip": 0.01140143, "auxiliary_loss_mlp": 0.01024487, "balance_loss_clip": 1.04542458, "balance_loss_mlp": 1.01748371, "epoch": 0.8537245235375458, "flos": 24681372186720.0, "grad_norm": 1.9795782050176474, "language_loss": 0.73001528, "learning_rate": 2.2014921801782016e-07, "loss": 0.75166154, "num_input_tokens_seen": 153703040, "step": 7100, "time_per_iteration": 2.552551507949829 }, { "auxiliary_loss_clip": 0.0113996, "auxiliary_loss_mlp": 0.01030089, "balance_loss_clip": 1.04128098, "balance_loss_mlp": 1.02307343, "epoch": 0.853844766428185, "flos": 24384821037600.0, "grad_norm": 1.7682634982994079, "language_loss": 0.73746228, "learning_rate": 2.1979405928720872e-07, "loss": 0.75916278, "num_input_tokens_seen": 153722695, "step": 7101, "time_per_iteration": 2.542112112045288 }, { "auxiliary_loss_clip": 0.01143336, "auxiliary_loss_mlp": 0.01023227, "balance_loss_clip": 1.04513574, "balance_loss_mlp": 1.01681948, "epoch": 0.853965009318824, "flos": 20955711029280.0, "grad_norm": 2.1015132799202894, "language_loss": 0.7894128, "learning_rate": 2.1943917061242257e-07, "loss": 0.81107843, "num_input_tokens_seen": 153742550, "step": 7102, "time_per_iteration": 3.287184000015259 }, { "auxiliary_loss_clip": 0.01160878, "auxiliary_loss_mlp": 0.00762923, "balance_loss_clip": 1.04718041, "balance_loss_mlp": 1.00060141, "epoch": 0.8540852522094631, "flos": 24201820151040.0, "grad_norm": 1.6427420608842374, "language_loss": 0.66377556, "learning_rate": 2.1908455204729903e-07, "loss": 0.68301356, "num_input_tokens_seen": 153761700, "step": 7103, "time_per_iteration": 2.527827501296997 }, { "auxiliary_loss_clip": 0.01138029, "auxiliary_loss_mlp": 0.01023415, "balance_loss_clip": 1.04232407, "balance_loss_mlp": 1.01654863, "epoch": 0.8542054951001022, "flos": 25082925361920.0, "grad_norm": 1.9967033309493032, "language_loss": 0.78253299, "learning_rate": 2.1873020364563265e-07, "loss": 0.80414742, "num_input_tokens_seen": 153780765, "step": 7104, "time_per_iteration": 2.5532925128936768 }, { "auxiliary_loss_clip": 0.01150933, "auxiliary_loss_mlp": 0.01023532, "balance_loss_clip": 1.04654837, "balance_loss_mlp": 1.01718366, "epoch": 0.8543257379907413, "flos": 24316555673280.0, "grad_norm": 2.2534202387890607, "language_loss": 0.76594615, "learning_rate": 2.183761254611789e-07, "loss": 0.78769082, "num_input_tokens_seen": 153801090, "step": 7105, "time_per_iteration": 3.3095502853393555 }, { "auxiliary_loss_clip": 0.01154004, "auxiliary_loss_mlp": 0.01026104, "balance_loss_clip": 1.04828811, "balance_loss_mlp": 1.01916957, "epoch": 0.8544459808813804, "flos": 55286636632800.0, "grad_norm": 1.9758360618881292, "language_loss": 0.70367908, "learning_rate": 2.1802231754764987e-07, "loss": 0.7254802, "num_input_tokens_seen": 153826530, "step": 7106, "time_per_iteration": 2.7744011878967285 }, { "auxiliary_loss_clip": 0.01143948, "auxiliary_loss_mlp": 0.01028088, "balance_loss_clip": 1.04506898, "balance_loss_mlp": 1.02110779, "epoch": 0.8545662237720195, "flos": 25776252730560.0, "grad_norm": 1.90953254652695, "language_loss": 0.7613008, "learning_rate": 2.17668779958718e-07, "loss": 0.78302109, "num_input_tokens_seen": 153849110, "step": 7107, "time_per_iteration": 2.6250195503234863 }, { "auxiliary_loss_clip": 0.01168783, "auxiliary_loss_mlp": 0.01025772, "balance_loss_clip": 1.0489341, "balance_loss_mlp": 1.01838386, "epoch": 0.8546864666626586, "flos": 11108323021440.0, "grad_norm": 2.3840257598599326, "language_loss": 0.80252534, "learning_rate": 2.1731551274801553e-07, "loss": 0.82447088, "num_input_tokens_seen": 153865550, "step": 7108, "time_per_iteration": 2.4740259647369385 }, { "auxiliary_loss_clip": 0.01143958, "auxiliary_loss_mlp": 0.01025175, "balance_loss_clip": 1.046664, "balance_loss_mlp": 1.0183022, "epoch": 0.8548067095532976, "flos": 25520173607040.0, "grad_norm": 2.0423290779745793, "language_loss": 0.61611867, "learning_rate": 2.169625159691324e-07, "loss": 0.63780999, "num_input_tokens_seen": 153885425, "step": 7109, "time_per_iteration": 3.3288917541503906 }, { "auxiliary_loss_clip": 0.01118079, "auxiliary_loss_mlp": 0.01027463, "balance_loss_clip": 1.04044247, "balance_loss_mlp": 1.02019691, "epoch": 0.8549269524439368, "flos": 24717857429760.0, "grad_norm": 3.123001292326626, "language_loss": 0.74334264, "learning_rate": 2.1660978967561784e-07, "loss": 0.76479805, "num_input_tokens_seen": 153904760, "step": 7110, "time_per_iteration": 2.6528432369232178 }, { "auxiliary_loss_clip": 0.01164919, "auxiliary_loss_mlp": 0.01023172, "balance_loss_clip": 1.04551554, "balance_loss_mlp": 1.01622224, "epoch": 0.8550471953345758, "flos": 19825602336000.0, "grad_norm": 2.248580762301479, "language_loss": 0.78965104, "learning_rate": 2.1625733392098035e-07, "loss": 0.8115319, "num_input_tokens_seen": 153920370, "step": 7111, "time_per_iteration": 2.4630959033966064 }, { "auxiliary_loss_clip": 0.01164593, "auxiliary_loss_mlp": 0.01020572, "balance_loss_clip": 1.04533529, "balance_loss_mlp": 1.01383972, "epoch": 0.8551674382252149, "flos": 22820445206880.0, "grad_norm": 1.686220331561728, "language_loss": 0.796574, "learning_rate": 2.159051487586867e-07, "loss": 0.81842566, "num_input_tokens_seen": 153940500, "step": 7112, "time_per_iteration": 2.5039925575256348 }, { "auxiliary_loss_clip": 0.01144026, "auxiliary_loss_mlp": 0.01025466, "balance_loss_clip": 1.04673052, "balance_loss_mlp": 1.01835489, "epoch": 0.8552876811158541, "flos": 20631258790560.0, "grad_norm": 2.9350905973613717, "language_loss": 0.72660971, "learning_rate": 2.155532342421642e-07, "loss": 0.74830461, "num_input_tokens_seen": 153958500, "step": 7113, "time_per_iteration": 2.536362648010254 }, { "auxiliary_loss_clip": 0.01157541, "auxiliary_loss_mlp": 0.01026794, "balance_loss_clip": 1.04701221, "balance_loss_mlp": 1.0196116, "epoch": 0.8554079240064931, "flos": 23112363068160.0, "grad_norm": 1.9463313476422122, "language_loss": 0.78446352, "learning_rate": 2.1520159042479636e-07, "loss": 0.8063069, "num_input_tokens_seen": 153976790, "step": 7114, "time_per_iteration": 2.5466694831848145 }, { "auxiliary_loss_clip": 0.01154583, "auxiliary_loss_mlp": 0.01022964, "balance_loss_clip": 1.0489645, "balance_loss_mlp": 1.01621425, "epoch": 0.8555281668971322, "flos": 22128051679200.0, "grad_norm": 2.11439748187232, "language_loss": 0.70783669, "learning_rate": 2.148502173599287e-07, "loss": 0.72961223, "num_input_tokens_seen": 153994930, "step": 7115, "time_per_iteration": 2.5418713092803955 }, { "auxiliary_loss_clip": 0.0113604, "auxiliary_loss_mlp": 0.01026571, "balance_loss_clip": 1.04493999, "balance_loss_mlp": 1.01966929, "epoch": 0.8556484097877713, "flos": 31139044703040.0, "grad_norm": 1.7139426177914943, "language_loss": 0.65879971, "learning_rate": 2.1449911510086372e-07, "loss": 0.68042582, "num_input_tokens_seen": 154014400, "step": 7116, "time_per_iteration": 2.6297430992126465 }, { "auxiliary_loss_clip": 0.0115153, "auxiliary_loss_mlp": 0.01024547, "balance_loss_clip": 1.04553699, "balance_loss_mlp": 1.01806235, "epoch": 0.8557686526784104, "flos": 24316555673280.0, "grad_norm": 1.9149407420950988, "language_loss": 0.77202368, "learning_rate": 2.141482837008628e-07, "loss": 0.79378444, "num_input_tokens_seen": 154034940, "step": 7117, "time_per_iteration": 2.5355732440948486 }, { "auxiliary_loss_clip": 0.01144756, "auxiliary_loss_mlp": 0.01023104, "balance_loss_clip": 1.04310286, "balance_loss_mlp": 1.01623476, "epoch": 0.8558888955690495, "flos": 17712726683040.0, "grad_norm": 1.9931547067552855, "language_loss": 0.72223896, "learning_rate": 2.1379772321314826e-07, "loss": 0.74391758, "num_input_tokens_seen": 154052985, "step": 7118, "time_per_iteration": 2.4870617389678955 }, { "auxiliary_loss_clip": 0.01090513, "auxiliary_loss_mlp": 0.01032593, "balance_loss_clip": 1.04099584, "balance_loss_mlp": 1.02592039, "epoch": 0.8560091384596886, "flos": 19171705401120.0, "grad_norm": 1.8761780919043234, "language_loss": 0.81551355, "learning_rate": 2.1344743369089802e-07, "loss": 0.83674461, "num_input_tokens_seen": 154068765, "step": 7119, "time_per_iteration": 3.322471857070923 }, { "auxiliary_loss_clip": 0.0114233, "auxiliary_loss_mlp": 0.01025379, "balance_loss_clip": 1.04807639, "balance_loss_mlp": 1.01879287, "epoch": 0.8561293813503277, "flos": 23914858830240.0, "grad_norm": 1.6218119410672742, "language_loss": 0.81829154, "learning_rate": 2.130974151872522e-07, "loss": 0.83996868, "num_input_tokens_seen": 154089100, "step": 7120, "time_per_iteration": 2.5996835231781006 }, { "auxiliary_loss_clip": 0.01125245, "auxiliary_loss_mlp": 0.01027253, "balance_loss_clip": 1.04601431, "balance_loss_mlp": 1.01978481, "epoch": 0.8562496242409667, "flos": 22529209767840.0, "grad_norm": 1.6749091666255196, "language_loss": 0.78015494, "learning_rate": 2.1274766775530773e-07, "loss": 0.80167985, "num_input_tokens_seen": 154108965, "step": 7121, "time_per_iteration": 2.543980598449707 }, { "auxiliary_loss_clip": 0.01173208, "auxiliary_loss_mlp": 0.01024736, "balance_loss_clip": 1.04859376, "balance_loss_mlp": 1.01736248, "epoch": 0.8563698671316058, "flos": 14712747686880.0, "grad_norm": 2.022477010334618, "language_loss": 0.79607302, "learning_rate": 2.1239819144812077e-07, "loss": 0.81805247, "num_input_tokens_seen": 154123425, "step": 7122, "time_per_iteration": 2.4205973148345947 }, { "auxiliary_loss_clip": 0.01116937, "auxiliary_loss_mlp": 0.01024481, "balance_loss_clip": 1.03922188, "balance_loss_mlp": 1.01773047, "epoch": 0.856490110022245, "flos": 39167773604640.0, "grad_norm": 1.8064362494938502, "language_loss": 0.69764251, "learning_rate": 2.1204898631870716e-07, "loss": 0.71905673, "num_input_tokens_seen": 154148315, "step": 7123, "time_per_iteration": 2.7170095443725586 }, { "auxiliary_loss_clip": 0.01140143, "auxiliary_loss_mlp": 0.01026956, "balance_loss_clip": 1.04582036, "balance_loss_mlp": 1.02009201, "epoch": 0.856610352912884, "flos": 29059349932800.0, "grad_norm": 2.2040402353475437, "language_loss": 0.76096737, "learning_rate": 2.1170005242004006e-07, "loss": 0.78263843, "num_input_tokens_seen": 154169665, "step": 7124, "time_per_iteration": 2.577150821685791 }, { "auxiliary_loss_clip": 0.01139209, "auxiliary_loss_mlp": 0.010242, "balance_loss_clip": 1.04361963, "balance_loss_mlp": 1.01760161, "epoch": 0.8567305958035231, "flos": 23878337670240.0, "grad_norm": 2.16352209681534, "language_loss": 0.78219587, "learning_rate": 2.1135138980505384e-07, "loss": 0.80382991, "num_input_tokens_seen": 154190335, "step": 7125, "time_per_iteration": 2.5465588569641113 }, { "auxiliary_loss_clip": 0.01135068, "auxiliary_loss_mlp": 0.01024401, "balance_loss_clip": 1.0442903, "balance_loss_mlp": 1.01772869, "epoch": 0.8568508386941622, "flos": 22200124241280.0, "grad_norm": 1.7458057705083478, "language_loss": 0.72605574, "learning_rate": 2.110029985266395e-07, "loss": 0.74765038, "num_input_tokens_seen": 154210040, "step": 7126, "time_per_iteration": 2.52384352684021 }, { "auxiliary_loss_clip": 0.01139835, "auxiliary_loss_mlp": 0.01022823, "balance_loss_clip": 1.04402828, "balance_loss_mlp": 1.01600432, "epoch": 0.8569710815848013, "flos": 17307509977920.0, "grad_norm": 1.6874972925152039, "language_loss": 0.73830593, "learning_rate": 2.1065487863764787e-07, "loss": 0.75993252, "num_input_tokens_seen": 154228385, "step": 7127, "time_per_iteration": 2.4945430755615234 }, { "auxiliary_loss_clip": 0.0110177, "auxiliary_loss_mlp": 0.0102328, "balance_loss_clip": 1.03706634, "balance_loss_mlp": 1.01645494, "epoch": 0.8570913244754403, "flos": 23732288947200.0, "grad_norm": 1.4068276034086702, "language_loss": 0.85690242, "learning_rate": 2.1030703019088846e-07, "loss": 0.87815297, "num_input_tokens_seen": 154249015, "step": 7128, "time_per_iteration": 3.368152618408203 }, { "auxiliary_loss_clip": 0.01147976, "auxiliary_loss_mlp": 0.01022194, "balance_loss_clip": 1.04543066, "balance_loss_mlp": 1.01530409, "epoch": 0.8572115673660795, "flos": 20048751995520.0, "grad_norm": 2.4326603854537283, "language_loss": 0.70867616, "learning_rate": 2.099594532391291e-07, "loss": 0.73037785, "num_input_tokens_seen": 154267700, "step": 7129, "time_per_iteration": 2.4951586723327637 }, { "auxiliary_loss_clip": 0.01144687, "auxiliary_loss_mlp": 0.01020932, "balance_loss_clip": 1.04368913, "balance_loss_mlp": 1.01432467, "epoch": 0.8573318102567186, "flos": 27160393280640.0, "grad_norm": 1.8476299301969055, "language_loss": 0.7904045, "learning_rate": 2.0961214783509806e-07, "loss": 0.81206065, "num_input_tokens_seen": 154290580, "step": 7130, "time_per_iteration": 2.5309221744537354 }, { "auxiliary_loss_clip": 0.0114486, "auxiliary_loss_mlp": 0.01024847, "balance_loss_clip": 1.04534769, "balance_loss_mlp": 1.01837111, "epoch": 0.8574520531473576, "flos": 24936589303200.0, "grad_norm": 1.7157274144631545, "language_loss": 0.74703962, "learning_rate": 2.0926511403148051e-07, "loss": 0.76873666, "num_input_tokens_seen": 154309545, "step": 7131, "time_per_iteration": 2.5585694313049316 }, { "auxiliary_loss_clip": 0.01130832, "auxiliary_loss_mlp": 0.01019971, "balance_loss_clip": 1.04686153, "balance_loss_mlp": 1.01322031, "epoch": 0.8575722960379968, "flos": 18771157900800.0, "grad_norm": 1.7809500487516718, "language_loss": 0.75526702, "learning_rate": 2.0891835188092143e-07, "loss": 0.77677512, "num_input_tokens_seen": 154326545, "step": 7132, "time_per_iteration": 3.2954132556915283 }, { "auxiliary_loss_clip": 0.01127984, "auxiliary_loss_mlp": 0.01024706, "balance_loss_clip": 1.04357886, "balance_loss_mlp": 1.01761615, "epoch": 0.8576925389286358, "flos": 22200303826080.0, "grad_norm": 1.8744157290908487, "language_loss": 0.81762886, "learning_rate": 2.0857186143602434e-07, "loss": 0.83915579, "num_input_tokens_seen": 154345190, "step": 7133, "time_per_iteration": 2.5752856731414795 }, { "auxiliary_loss_clip": 0.01116167, "auxiliary_loss_mlp": 0.01024088, "balance_loss_clip": 1.04090238, "balance_loss_mlp": 1.01730824, "epoch": 0.8578127818192749, "flos": 22894349533920.0, "grad_norm": 2.247093424402192, "language_loss": 0.6761688, "learning_rate": 2.0822564274935094e-07, "loss": 0.69757134, "num_input_tokens_seen": 154364615, "step": 7134, "time_per_iteration": 2.558241128921509 }, { "auxiliary_loss_clip": 0.01143202, "auxiliary_loss_mlp": 0.01023587, "balance_loss_clip": 1.04856455, "balance_loss_mlp": 1.01617861, "epoch": 0.8579330247099141, "flos": 34824844423200.0, "grad_norm": 3.2254393752030697, "language_loss": 0.66950637, "learning_rate": 2.078796958734239e-07, "loss": 0.69117427, "num_input_tokens_seen": 154387335, "step": 7135, "time_per_iteration": 3.373382806777954 }, { "auxiliary_loss_clip": 0.0115496, "auxiliary_loss_mlp": 0.01027719, "balance_loss_clip": 1.04835856, "balance_loss_mlp": 1.02048004, "epoch": 0.8580532676005531, "flos": 19755684791520.0, "grad_norm": 2.2040818916372795, "language_loss": 0.74867463, "learning_rate": 2.0753402086072124e-07, "loss": 0.77050143, "num_input_tokens_seen": 154405965, "step": 7136, "time_per_iteration": 2.4607410430908203 }, { "auxiliary_loss_clip": 0.01094774, "auxiliary_loss_mlp": 0.01029423, "balance_loss_clip": 1.04215372, "balance_loss_mlp": 1.02205849, "epoch": 0.8581735104911922, "flos": 22739321570880.0, "grad_norm": 2.090806176511421, "language_loss": 0.75233877, "learning_rate": 2.071886177636828e-07, "loss": 0.77358073, "num_input_tokens_seen": 154422750, "step": 7137, "time_per_iteration": 2.6218156814575195 }, { "auxiliary_loss_clip": 0.01150684, "auxiliary_loss_mlp": 0.01024901, "balance_loss_clip": 1.04682469, "balance_loss_mlp": 1.01851737, "epoch": 0.8582937533818313, "flos": 23149135646880.0, "grad_norm": 3.3722486063191845, "language_loss": 0.83063143, "learning_rate": 2.0684348663470575e-07, "loss": 0.85238731, "num_input_tokens_seen": 154442930, "step": 7138, "time_per_iteration": 2.532449960708618 }, { "auxiliary_loss_clip": 0.0113751, "auxiliary_loss_mlp": 0.01025522, "balance_loss_clip": 1.04127502, "balance_loss_mlp": 1.01840854, "epoch": 0.8584139962724704, "flos": 19498671827040.0, "grad_norm": 1.698327617722744, "language_loss": 0.61379415, "learning_rate": 2.0649862752614555e-07, "loss": 0.63542449, "num_input_tokens_seen": 154461640, "step": 7139, "time_per_iteration": 2.5567784309387207 }, { "auxiliary_loss_clip": 0.01041, "auxiliary_loss_mlp": 0.01001139, "balance_loss_clip": 1.00594735, "balance_loss_mlp": 0.99979812, "epoch": 0.8585342391631094, "flos": 71276589716160.0, "grad_norm": 0.7567336709285335, "language_loss": 0.57111812, "learning_rate": 2.0615404049031838e-07, "loss": 0.5915395, "num_input_tokens_seen": 154518610, "step": 7140, "time_per_iteration": 3.117111921310425 }, { "auxiliary_loss_clip": 0.01154121, "auxiliary_loss_mlp": 0.0102595, "balance_loss_clip": 1.04712093, "balance_loss_mlp": 1.01851416, "epoch": 0.8586544820537486, "flos": 10815435402240.0, "grad_norm": 2.4709507266161053, "language_loss": 0.78242397, "learning_rate": 2.0580972557949616e-07, "loss": 0.80422461, "num_input_tokens_seen": 154533700, "step": 7141, "time_per_iteration": 2.4516074657440186 }, { "auxiliary_loss_clip": 0.01050736, "auxiliary_loss_mlp": 0.01001133, "balance_loss_clip": 1.00541115, "balance_loss_mlp": 0.99981594, "epoch": 0.8587747249443877, "flos": 64811195053440.0, "grad_norm": 0.7958868289718716, "language_loss": 0.54257619, "learning_rate": 2.054656828459125e-07, "loss": 0.56309491, "num_input_tokens_seen": 154597810, "step": 7142, "time_per_iteration": 3.1157031059265137 }, { "auxiliary_loss_clip": 0.01109478, "auxiliary_loss_mlp": 0.01026964, "balance_loss_clip": 1.04214478, "balance_loss_mlp": 1.0194242, "epoch": 0.8588949678350267, "flos": 26834612114400.0, "grad_norm": 1.7626687766563336, "language_loss": 0.77068436, "learning_rate": 2.051219123417578e-07, "loss": 0.79204881, "num_input_tokens_seen": 154617870, "step": 7143, "time_per_iteration": 2.619971990585327 }, { "auxiliary_loss_clip": 0.01169516, "auxiliary_loss_mlp": 0.0102053, "balance_loss_clip": 1.04670644, "balance_loss_mlp": 1.01315129, "epoch": 0.8590152107256659, "flos": 26104260748320.0, "grad_norm": 3.3002495419114792, "language_loss": 0.60366154, "learning_rate": 2.0477841411918196e-07, "loss": 0.62556195, "num_input_tokens_seen": 154637395, "step": 7144, "time_per_iteration": 2.562749147415161 }, { "auxiliary_loss_clip": 0.01149509, "auxiliary_loss_mlp": 0.01021558, "balance_loss_clip": 1.04535151, "balance_loss_mlp": 1.01486158, "epoch": 0.859135453616305, "flos": 26140889659200.0, "grad_norm": 2.138746256264639, "language_loss": 0.7451604, "learning_rate": 2.0443518823029326e-07, "loss": 0.76687104, "num_input_tokens_seen": 154657935, "step": 7145, "time_per_iteration": 2.60473895072937 }, { "auxiliary_loss_clip": 0.01119924, "auxiliary_loss_mlp": 0.01025543, "balance_loss_clip": 1.0428524, "balance_loss_mlp": 1.01819634, "epoch": 0.859255696506944, "flos": 12969321835200.0, "grad_norm": 1.9546274200694191, "language_loss": 0.76605308, "learning_rate": 2.0409223472715854e-07, "loss": 0.78750777, "num_input_tokens_seen": 154675080, "step": 7146, "time_per_iteration": 3.2536463737487793 }, { "auxiliary_loss_clip": 0.01125667, "auxiliary_loss_mlp": 0.00761664, "balance_loss_clip": 1.04348564, "balance_loss_mlp": 1.00058889, "epoch": 0.8593759393975832, "flos": 18475756094400.0, "grad_norm": 1.8477368264145049, "language_loss": 0.74759275, "learning_rate": 2.0374955366180434e-07, "loss": 0.76646602, "num_input_tokens_seen": 154692720, "step": 7147, "time_per_iteration": 2.5389389991760254 }, { "auxiliary_loss_clip": 0.01129603, "auxiliary_loss_mlp": 0.0102541, "balance_loss_clip": 1.04220057, "balance_loss_mlp": 1.01853752, "epoch": 0.8594961822882222, "flos": 22200160158240.0, "grad_norm": 1.88048933948675, "language_loss": 0.7240293, "learning_rate": 2.034071450862147e-07, "loss": 0.74557936, "num_input_tokens_seen": 154710190, "step": 7148, "time_per_iteration": 2.5660178661346436 }, { "auxiliary_loss_clip": 0.01141434, "auxiliary_loss_mlp": 0.01026774, "balance_loss_clip": 1.04222119, "balance_loss_mlp": 1.01870334, "epoch": 0.8596164251788613, "flos": 23294753366400.0, "grad_norm": 1.740811149957807, "language_loss": 0.77023101, "learning_rate": 2.030650090523327e-07, "loss": 0.79191303, "num_input_tokens_seen": 154729380, "step": 7149, "time_per_iteration": 2.5292272567749023 }, { "auxiliary_loss_clip": 0.0112245, "auxiliary_loss_mlp": 0.01028516, "balance_loss_clip": 1.04213321, "balance_loss_mlp": 1.02143168, "epoch": 0.8597366680695004, "flos": 31649909939520.0, "grad_norm": 1.5844463651518212, "language_loss": 0.59277815, "learning_rate": 2.0272314561205995e-07, "loss": 0.61428773, "num_input_tokens_seen": 154749775, "step": 7150, "time_per_iteration": 2.658419370651245 }, { "auxiliary_loss_clip": 0.01117378, "auxiliary_loss_mlp": 0.01020737, "balance_loss_clip": 1.03972709, "balance_loss_mlp": 1.0142014, "epoch": 0.8598569109601395, "flos": 21287741746560.0, "grad_norm": 1.7840162988207011, "language_loss": 0.72820824, "learning_rate": 2.023815548172567e-07, "loss": 0.74958938, "num_input_tokens_seen": 154769845, "step": 7151, "time_per_iteration": 2.5562567710876465 }, { "auxiliary_loss_clip": 0.01153451, "auxiliary_loss_mlp": 0.01023259, "balance_loss_clip": 1.04558635, "balance_loss_mlp": 1.01621079, "epoch": 0.8599771538507786, "flos": 25447813709280.0, "grad_norm": 1.8191615533553287, "language_loss": 0.65940893, "learning_rate": 2.0204023671974267e-07, "loss": 0.68117607, "num_input_tokens_seen": 154789230, "step": 7152, "time_per_iteration": 2.5408129692077637 }, { "auxiliary_loss_clip": 0.0114925, "auxiliary_loss_mlp": 0.01027703, "balance_loss_clip": 1.04496813, "balance_loss_mlp": 1.02081537, "epoch": 0.8600973967414177, "flos": 16723961591040.0, "grad_norm": 2.0906298392069433, "language_loss": 0.81032282, "learning_rate": 2.0169919137129532e-07, "loss": 0.83209234, "num_input_tokens_seen": 154807670, "step": 7153, "time_per_iteration": 2.451441764831543 }, { "auxiliary_loss_clip": 0.01156458, "auxiliary_loss_mlp": 0.01023462, "balance_loss_clip": 1.04789972, "balance_loss_mlp": 1.01613677, "epoch": 0.8602176396320568, "flos": 25227932493120.0, "grad_norm": 2.3583538759022438, "language_loss": 0.71108001, "learning_rate": 2.013584188236508e-07, "loss": 0.73287922, "num_input_tokens_seen": 154825575, "step": 7154, "time_per_iteration": 3.2883036136627197 }, { "auxiliary_loss_clip": 0.01170149, "auxiliary_loss_mlp": 0.01027058, "balance_loss_clip": 1.04775047, "balance_loss_mlp": 1.02052283, "epoch": 0.8603378825226958, "flos": 20412239581440.0, "grad_norm": 1.9990450678467426, "language_loss": 0.7932902, "learning_rate": 2.0101791912850396e-07, "loss": 0.8152622, "num_input_tokens_seen": 154845115, "step": 7155, "time_per_iteration": 2.455810308456421 }, { "auxiliary_loss_clip": 0.01141179, "auxiliary_loss_mlp": 0.0103418, "balance_loss_clip": 1.0466125, "balance_loss_mlp": 1.02694988, "epoch": 0.8604581254133349, "flos": 34930205618880.0, "grad_norm": 1.758725862589489, "language_loss": 0.63902283, "learning_rate": 2.006776923375082e-07, "loss": 0.66077638, "num_input_tokens_seen": 154866770, "step": 7156, "time_per_iteration": 2.6220638751983643 }, { "auxiliary_loss_clip": 0.01167347, "auxiliary_loss_mlp": 0.01022276, "balance_loss_clip": 1.04772472, "balance_loss_mlp": 1.0155977, "epoch": 0.860578368303974, "flos": 22596541291200.0, "grad_norm": 1.6790161072924392, "language_loss": 0.71140277, "learning_rate": 2.003377385022764e-07, "loss": 0.73329908, "num_input_tokens_seen": 154885595, "step": 7157, "time_per_iteration": 2.480966567993164 }, { "auxiliary_loss_clip": 0.01139597, "auxiliary_loss_mlp": 0.01026221, "balance_loss_clip": 1.04356527, "balance_loss_mlp": 1.01918507, "epoch": 0.8606986111946131, "flos": 21324334740480.0, "grad_norm": 1.742586980996071, "language_loss": 0.77458525, "learning_rate": 1.9999805767437826e-07, "loss": 0.79624343, "num_input_tokens_seen": 154904485, "step": 7158, "time_per_iteration": 3.3214166164398193 }, { "auxiliary_loss_clip": 0.01129859, "auxiliary_loss_mlp": 0.01023615, "balance_loss_clip": 1.04084182, "balance_loss_mlp": 1.0165782, "epoch": 0.8608188540852522, "flos": 28877211053280.0, "grad_norm": 1.7068392692749654, "language_loss": 0.71879894, "learning_rate": 1.9965864990534386e-07, "loss": 0.74033368, "num_input_tokens_seen": 154925010, "step": 7159, "time_per_iteration": 2.5772883892059326 }, { "auxiliary_loss_clip": 0.01116674, "auxiliary_loss_mlp": 0.01024269, "balance_loss_clip": 1.03930712, "balance_loss_mlp": 1.01770616, "epoch": 0.8609390969758913, "flos": 29716192058400.0, "grad_norm": 1.715694333589384, "language_loss": 0.77496493, "learning_rate": 1.9931951524666092e-07, "loss": 0.79637438, "num_input_tokens_seen": 154946100, "step": 7160, "time_per_iteration": 2.6102817058563232 }, { "auxiliary_loss_clip": 0.01156034, "auxiliary_loss_mlp": 0.00761748, "balance_loss_clip": 1.04616189, "balance_loss_mlp": 1.00055981, "epoch": 0.8610593398665304, "flos": 21249352904640.0, "grad_norm": 2.3912659417335824, "language_loss": 0.81035709, "learning_rate": 1.9898065374977534e-07, "loss": 0.82953489, "num_input_tokens_seen": 154966305, "step": 7161, "time_per_iteration": 3.4540812969207764 }, { "auxiliary_loss_clip": 0.01118811, "auxiliary_loss_mlp": 0.0102126, "balance_loss_clip": 1.04230499, "balance_loss_mlp": 1.01547563, "epoch": 0.8611795827571694, "flos": 14830105147200.0, "grad_norm": 2.0697443523182617, "language_loss": 0.73146325, "learning_rate": 1.9864206546609342e-07, "loss": 0.752864, "num_input_tokens_seen": 154985145, "step": 7162, "time_per_iteration": 2.560499429702759 }, { "auxiliary_loss_clip": 0.01166872, "auxiliary_loss_mlp": 0.01027361, "balance_loss_clip": 1.04631209, "balance_loss_mlp": 1.02039075, "epoch": 0.8612998256478086, "flos": 24243261934560.0, "grad_norm": 1.7364692657527396, "language_loss": 0.84286422, "learning_rate": 1.983037504469771e-07, "loss": 0.86480653, "num_input_tokens_seen": 155003855, "step": 7163, "time_per_iteration": 2.466031312942505 }, { "auxiliary_loss_clip": 0.01157685, "auxiliary_loss_mlp": 0.01032064, "balance_loss_clip": 1.04846311, "balance_loss_mlp": 1.02493262, "epoch": 0.8614200685384477, "flos": 21252657264960.0, "grad_norm": 1.7383491797549477, "language_loss": 0.6660434, "learning_rate": 1.9796570874374984e-07, "loss": 0.6879409, "num_input_tokens_seen": 155023960, "step": 7164, "time_per_iteration": 2.5186824798583984 }, { "auxiliary_loss_clip": 0.01141208, "auxiliary_loss_mlp": 0.01024669, "balance_loss_clip": 1.04370356, "balance_loss_mlp": 1.01822615, "epoch": 0.8615403114290867, "flos": 20007741215520.0, "grad_norm": 1.6895172746543932, "language_loss": 0.77609479, "learning_rate": 1.976279404076917e-07, "loss": 0.79775357, "num_input_tokens_seen": 155043360, "step": 7165, "time_per_iteration": 2.5415782928466797 }, { "auxiliary_loss_clip": 0.01126416, "auxiliary_loss_mlp": 0.01028328, "balance_loss_clip": 1.04624653, "balance_loss_mlp": 1.02124107, "epoch": 0.8616605543197259, "flos": 29789378046240.0, "grad_norm": 1.8018671057752829, "language_loss": 0.75823879, "learning_rate": 1.9729044549004193e-07, "loss": 0.77978617, "num_input_tokens_seen": 155064745, "step": 7166, "time_per_iteration": 2.6470768451690674 }, { "auxiliary_loss_clip": 0.01149826, "auxiliary_loss_mlp": 0.01027803, "balance_loss_clip": 1.04546094, "balance_loss_mlp": 1.02068973, "epoch": 0.8617807972103649, "flos": 28911613112640.0, "grad_norm": 1.7243043880527504, "language_loss": 0.70156991, "learning_rate": 1.9695322404199822e-07, "loss": 0.72334617, "num_input_tokens_seen": 155086790, "step": 7167, "time_per_iteration": 2.580658435821533 }, { "auxiliary_loss_clip": 0.01143125, "auxiliary_loss_mlp": 0.01026265, "balance_loss_clip": 1.04782689, "balance_loss_mlp": 1.01870728, "epoch": 0.861901040101004, "flos": 27673808621280.0, "grad_norm": 2.181053314681039, "language_loss": 0.82358372, "learning_rate": 1.9661627611471654e-07, "loss": 0.84527767, "num_input_tokens_seen": 155106585, "step": 7168, "time_per_iteration": 2.5806620121002197 }, { "auxiliary_loss_clip": 0.01142322, "auxiliary_loss_mlp": 0.01023967, "balance_loss_clip": 1.04485512, "balance_loss_mlp": 1.01633179, "epoch": 0.8620212829916432, "flos": 49748063082720.0, "grad_norm": 1.8210514456912383, "language_loss": 0.70339996, "learning_rate": 1.9627960175931246e-07, "loss": 0.72506285, "num_input_tokens_seen": 155131285, "step": 7169, "time_per_iteration": 2.7626073360443115 }, { "auxiliary_loss_clip": 0.01153971, "auxiliary_loss_mlp": 0.01025002, "balance_loss_clip": 1.04803634, "balance_loss_mlp": 1.01801014, "epoch": 0.8621415258822822, "flos": 21138675998880.0, "grad_norm": 1.9081853533673314, "language_loss": 0.74013549, "learning_rate": 1.9594320102685847e-07, "loss": 0.76192522, "num_input_tokens_seen": 155150555, "step": 7170, "time_per_iteration": 2.498262882232666 }, { "auxiliary_loss_clip": 0.01128931, "auxiliary_loss_mlp": 0.00761322, "balance_loss_clip": 1.04203272, "balance_loss_mlp": 1.00051641, "epoch": 0.8622617687729213, "flos": 21689043503040.0, "grad_norm": 2.2249842246362252, "language_loss": 0.63918167, "learning_rate": 1.956070739683864e-07, "loss": 0.65808415, "num_input_tokens_seen": 155169890, "step": 7171, "time_per_iteration": 2.530454397201538 }, { "auxiliary_loss_clip": 0.01109782, "auxiliary_loss_mlp": 0.01025239, "balance_loss_clip": 1.03993702, "balance_loss_mlp": 1.01892686, "epoch": 0.8623820116635604, "flos": 26250596807040.0, "grad_norm": 1.5327397301571695, "language_loss": 0.73696119, "learning_rate": 1.9527122063488678e-07, "loss": 0.75831139, "num_input_tokens_seen": 155191005, "step": 7172, "time_per_iteration": 3.397733688354492 }, { "auxiliary_loss_clip": 0.01135746, "auxiliary_loss_mlp": 0.01020094, "balance_loss_clip": 1.03966188, "balance_loss_mlp": 1.01342726, "epoch": 0.8625022545541995, "flos": 19647557989920.0, "grad_norm": 1.903361107224093, "language_loss": 0.80557442, "learning_rate": 1.9493564107730755e-07, "loss": 0.82713282, "num_input_tokens_seen": 155211005, "step": 7173, "time_per_iteration": 2.5450186729431152 }, { "auxiliary_loss_clip": 0.01132441, "auxiliary_loss_mlp": 0.01021231, "balance_loss_clip": 1.0412271, "balance_loss_mlp": 1.01508617, "epoch": 0.8626224974448385, "flos": 21908385964800.0, "grad_norm": 1.9050709811909787, "language_loss": 0.60751605, "learning_rate": 1.9460033534655684e-07, "loss": 0.62905276, "num_input_tokens_seen": 155230365, "step": 7174, "time_per_iteration": 2.5640954971313477 }, { "auxiliary_loss_clip": 0.01134658, "auxiliary_loss_mlp": 0.01020292, "balance_loss_clip": 1.04109263, "balance_loss_mlp": 1.01349103, "epoch": 0.8627427403354777, "flos": 23331202692480.0, "grad_norm": 1.5632318539460124, "language_loss": 0.8417573, "learning_rate": 1.9426530349349978e-07, "loss": 0.86330676, "num_input_tokens_seen": 155250815, "step": 7175, "time_per_iteration": 2.560020685195923 }, { "auxiliary_loss_clip": 0.01153, "auxiliary_loss_mlp": 0.00761857, "balance_loss_clip": 1.04632127, "balance_loss_mlp": 1.00051117, "epoch": 0.8628629832261168, "flos": 16362880441440.0, "grad_norm": 2.099701298371576, "language_loss": 0.65068853, "learning_rate": 1.9393054556896038e-07, "loss": 0.66983712, "num_input_tokens_seen": 155268515, "step": 7176, "time_per_iteration": 2.5185647010803223 }, { "auxiliary_loss_clip": 0.01118387, "auxiliary_loss_mlp": 0.01029228, "balance_loss_clip": 1.04176068, "balance_loss_mlp": 1.02214122, "epoch": 0.8629832261167558, "flos": 28103945308320.0, "grad_norm": 2.361642606250402, "language_loss": 0.69083595, "learning_rate": 1.9359606162372133e-07, "loss": 0.7123121, "num_input_tokens_seen": 155290120, "step": 7177, "time_per_iteration": 2.596820592880249 }, { "auxiliary_loss_clip": 0.01168603, "auxiliary_loss_mlp": 0.01022421, "balance_loss_clip": 1.04875159, "balance_loss_mlp": 1.01590025, "epoch": 0.863103469007395, "flos": 20230064784960.0, "grad_norm": 1.6219096230760484, "language_loss": 0.70734459, "learning_rate": 1.9326185170852293e-07, "loss": 0.72925484, "num_input_tokens_seen": 155309085, "step": 7178, "time_per_iteration": 2.4725382328033447 }, { "auxiliary_loss_clip": 0.01152692, "auxiliary_loss_mlp": 0.01025643, "balance_loss_clip": 1.04650497, "balance_loss_mlp": 1.01852036, "epoch": 0.863223711898034, "flos": 24498550884960.0, "grad_norm": 2.514652803614225, "language_loss": 0.71842599, "learning_rate": 1.9292791587406598e-07, "loss": 0.74020934, "num_input_tokens_seen": 155327945, "step": 7179, "time_per_iteration": 2.522479772567749 }, { "auxiliary_loss_clip": 0.01152614, "auxiliary_loss_mlp": 0.00762126, "balance_loss_clip": 1.04510999, "balance_loss_mlp": 1.00058067, "epoch": 0.8633439547886731, "flos": 17675379432960.0, "grad_norm": 2.106225424676761, "language_loss": 0.86763173, "learning_rate": 1.9259425417100661e-07, "loss": 0.88677919, "num_input_tokens_seen": 155344060, "step": 7180, "time_per_iteration": 3.2324047088623047 }, { "auxiliary_loss_clip": 0.01091407, "auxiliary_loss_mlp": 0.01021177, "balance_loss_clip": 1.03275681, "balance_loss_mlp": 1.0141139, "epoch": 0.8634641976793123, "flos": 12895058338560.0, "grad_norm": 2.001639394295849, "language_loss": 0.75148451, "learning_rate": 1.9226086664996234e-07, "loss": 0.77261031, "num_input_tokens_seen": 155362305, "step": 7181, "time_per_iteration": 2.6226601600646973 }, { "auxiliary_loss_clip": 0.01138183, "auxiliary_loss_mlp": 0.01024644, "balance_loss_clip": 1.04617929, "balance_loss_mlp": 1.0177958, "epoch": 0.8635844405699513, "flos": 23878984175520.0, "grad_norm": 1.756070073576528, "language_loss": 0.738469, "learning_rate": 1.9192775336150712e-07, "loss": 0.76009727, "num_input_tokens_seen": 155382605, "step": 7182, "time_per_iteration": 2.530933141708374 }, { "auxiliary_loss_clip": 0.01048252, "auxiliary_loss_mlp": 0.01002557, "balance_loss_clip": 1.00715876, "balance_loss_mlp": 1.00129938, "epoch": 0.8637046834605904, "flos": 60453396250560.0, "grad_norm": 0.766364369383558, "language_loss": 0.56287926, "learning_rate": 1.915949143561739e-07, "loss": 0.58338737, "num_input_tokens_seen": 155437280, "step": 7183, "time_per_iteration": 3.0727579593658447 }, { "auxiliary_loss_clip": 0.01155122, "auxiliary_loss_mlp": 0.01025206, "balance_loss_clip": 1.04796314, "balance_loss_mlp": 1.01829815, "epoch": 0.8638249263512295, "flos": 20558755224960.0, "grad_norm": 1.7900782100013182, "language_loss": 0.7802242, "learning_rate": 1.9126234968445498e-07, "loss": 0.80202746, "num_input_tokens_seen": 155456970, "step": 7184, "time_per_iteration": 2.496673107147217 }, { "auxiliary_loss_clip": 0.01166097, "auxiliary_loss_mlp": 0.01023899, "balance_loss_clip": 1.04607356, "balance_loss_mlp": 1.01690161, "epoch": 0.8639451692418686, "flos": 26615772490080.0, "grad_norm": 1.5792748579407896, "language_loss": 0.67703307, "learning_rate": 1.9093005939679884e-07, "loss": 0.69893312, "num_input_tokens_seen": 155478925, "step": 7185, "time_per_iteration": 3.286086320877075 }, { "auxiliary_loss_clip": 0.01156069, "auxiliary_loss_mlp": 0.01026397, "balance_loss_clip": 1.04930186, "balance_loss_mlp": 1.01916957, "epoch": 0.8640654121325076, "flos": 15122454012000.0, "grad_norm": 1.8886518788524198, "language_loss": 0.76489222, "learning_rate": 1.9059804354361452e-07, "loss": 0.78671694, "num_input_tokens_seen": 155496700, "step": 7186, "time_per_iteration": 2.455949544906616 }, { "auxiliary_loss_clip": 0.01132609, "auxiliary_loss_mlp": 0.01029788, "balance_loss_clip": 1.04163074, "balance_loss_mlp": 1.02261746, "epoch": 0.8641856550231467, "flos": 31869072816480.0, "grad_norm": 1.56667765593276, "language_loss": 0.70132315, "learning_rate": 1.902663021752684e-07, "loss": 0.72294712, "num_input_tokens_seen": 155518130, "step": 7187, "time_per_iteration": 3.3255538940429688 }, { "auxiliary_loss_clip": 0.0116906, "auxiliary_loss_mlp": 0.01026022, "balance_loss_clip": 1.04810977, "balance_loss_mlp": 1.01900077, "epoch": 0.8643058979137859, "flos": 14976548956800.0, "grad_norm": 2.329827289448361, "language_loss": 0.82429302, "learning_rate": 1.8993483534208556e-07, "loss": 0.84624386, "num_input_tokens_seen": 155537040, "step": 7188, "time_per_iteration": 2.4524009227752686 }, { "auxiliary_loss_clip": 0.01135787, "auxiliary_loss_mlp": 0.01024509, "balance_loss_clip": 1.046013, "balance_loss_mlp": 1.01751184, "epoch": 0.8644261408044249, "flos": 13115729727840.0, "grad_norm": 2.3609028340638054, "language_loss": 0.75075537, "learning_rate": 1.8960364309434884e-07, "loss": 0.7723583, "num_input_tokens_seen": 155554535, "step": 7189, "time_per_iteration": 2.4776556491851807 }, { "auxiliary_loss_clip": 0.01092379, "auxiliary_loss_mlp": 0.00761516, "balance_loss_clip": 1.04035819, "balance_loss_mlp": 1.00056791, "epoch": 0.864546383695064, "flos": 20850924504960.0, "grad_norm": 1.6339121017844578, "language_loss": 0.78198951, "learning_rate": 1.8927272548229967e-07, "loss": 0.80052847, "num_input_tokens_seen": 155574225, "step": 7190, "time_per_iteration": 2.6313462257385254 }, { "auxiliary_loss_clip": 0.01112553, "auxiliary_loss_mlp": 0.01033655, "balance_loss_clip": 1.04170156, "balance_loss_mlp": 1.02682161, "epoch": 0.8646666265857031, "flos": 21324586159200.0, "grad_norm": 1.5850830104701188, "language_loss": 0.83245504, "learning_rate": 1.8894208255613876e-07, "loss": 0.85391712, "num_input_tokens_seen": 155593540, "step": 7191, "time_per_iteration": 2.595240354537964 }, { "auxiliary_loss_clip": 0.01167081, "auxiliary_loss_mlp": 0.0102245, "balance_loss_clip": 1.04801941, "balance_loss_mlp": 1.01570284, "epoch": 0.8647868694763422, "flos": 19750836001920.0, "grad_norm": 2.07766948160541, "language_loss": 0.78001273, "learning_rate": 1.8861171436602397e-07, "loss": 0.80190808, "num_input_tokens_seen": 155610655, "step": 7192, "time_per_iteration": 2.457690954208374 }, { "auxiliary_loss_clip": 0.01157058, "auxiliary_loss_mlp": 0.01024499, "balance_loss_clip": 1.04811954, "balance_loss_mlp": 1.01774311, "epoch": 0.8649071123669813, "flos": 26176764313920.0, "grad_norm": 2.5148413707411605, "language_loss": 0.80534983, "learning_rate": 1.882816209620719e-07, "loss": 0.82716531, "num_input_tokens_seen": 155627365, "step": 7193, "time_per_iteration": 2.5359132289886475 }, { "auxiliary_loss_clip": 0.011463, "auxiliary_loss_mlp": 0.01028644, "balance_loss_clip": 1.04876566, "balance_loss_mlp": 1.0211544, "epoch": 0.8650273552576204, "flos": 20302891603200.0, "grad_norm": 1.8777034445491125, "language_loss": 0.77207708, "learning_rate": 1.8795180239435738e-07, "loss": 0.79382646, "num_input_tokens_seen": 155646220, "step": 7194, "time_per_iteration": 2.5246379375457764 }, { "auxiliary_loss_clip": 0.01145185, "auxiliary_loss_mlp": 0.01031025, "balance_loss_clip": 1.04557252, "balance_loss_mlp": 1.02408707, "epoch": 0.8651475981482595, "flos": 23951092654560.0, "grad_norm": 10.451026224761518, "language_loss": 0.75871211, "learning_rate": 1.8762225871291348e-07, "loss": 0.78047419, "num_input_tokens_seen": 155662095, "step": 7195, "time_per_iteration": 2.5317463874816895 }, { "auxiliary_loss_clip": 0.01166532, "auxiliary_loss_mlp": 0.00761394, "balance_loss_clip": 1.04695821, "balance_loss_mlp": 1.00054812, "epoch": 0.8652678410388985, "flos": 21684625716960.0, "grad_norm": 2.361975452177534, "language_loss": 0.80735046, "learning_rate": 1.8729298996773201e-07, "loss": 0.8266297, "num_input_tokens_seen": 155680845, "step": 7196, "time_per_iteration": 2.4960484504699707 }, { "auxiliary_loss_clip": 0.01045477, "auxiliary_loss_mlp": 0.01001367, "balance_loss_clip": 1.00536084, "balance_loss_mlp": 1.00015068, "epoch": 0.8653880839295377, "flos": 65224672659360.0, "grad_norm": 0.8341129484499132, "language_loss": 0.60933971, "learning_rate": 1.8696399620876301e-07, "loss": 0.62980819, "num_input_tokens_seen": 155737875, "step": 7197, "time_per_iteration": 3.0513718128204346 }, { "auxiliary_loss_clip": 0.01121489, "auxiliary_loss_mlp": 0.01025975, "balance_loss_clip": 1.03898954, "balance_loss_mlp": 1.01873004, "epoch": 0.8655083268201768, "flos": 17749176009120.0, "grad_norm": 2.7392475524830697, "language_loss": 0.79187131, "learning_rate": 1.866352774859141e-07, "loss": 0.81334591, "num_input_tokens_seen": 155753100, "step": 7198, "time_per_iteration": 3.301894187927246 }, { "auxiliary_loss_clip": 0.01127506, "auxiliary_loss_mlp": 0.0102056, "balance_loss_clip": 1.04115951, "balance_loss_mlp": 1.01419163, "epoch": 0.8656285697108158, "flos": 20703977857920.0, "grad_norm": 2.228454985832608, "language_loss": 0.69278342, "learning_rate": 1.8630683384905188e-07, "loss": 0.71426404, "num_input_tokens_seen": 155772430, "step": 7199, "time_per_iteration": 2.5369808673858643 }, { "auxiliary_loss_clip": 0.01165429, "auxiliary_loss_mlp": 0.00761542, "balance_loss_clip": 1.04628086, "balance_loss_mlp": 1.00054717, "epoch": 0.865748812601455, "flos": 18653836357440.0, "grad_norm": 2.0704749667783693, "language_loss": 0.88472736, "learning_rate": 1.8597866534800045e-07, "loss": 0.90399712, "num_input_tokens_seen": 155787545, "step": 7200, "time_per_iteration": 2.4501357078552246 }, { "auxiliary_loss_clip": 0.01158422, "auxiliary_loss_mlp": 0.00761998, "balance_loss_clip": 1.04719877, "balance_loss_mlp": 1.0005604, "epoch": 0.865869055492094, "flos": 70652562915360.0, "grad_norm": 1.7904097672034887, "language_loss": 0.74676973, "learning_rate": 1.8565077203254398e-07, "loss": 0.76597393, "num_input_tokens_seen": 155813005, "step": 7201, "time_per_iteration": 2.8781509399414062 }, { "auxiliary_loss_clip": 0.01129536, "auxiliary_loss_mlp": 0.01030925, "balance_loss_clip": 1.04820251, "balance_loss_mlp": 1.02364182, "epoch": 0.8659892983827331, "flos": 17383964409120.0, "grad_norm": 2.3113891063415406, "language_loss": 0.72869527, "learning_rate": 1.8532315395242203e-07, "loss": 0.75029987, "num_input_tokens_seen": 155829455, "step": 7202, "time_per_iteration": 2.5276217460632324 }, { "auxiliary_loss_clip": 0.01129025, "auxiliary_loss_mlp": 0.01025848, "balance_loss_clip": 1.04466081, "balance_loss_mlp": 1.01918423, "epoch": 0.8661095412733723, "flos": 17895224732160.0, "grad_norm": 2.0135105444361217, "language_loss": 0.72362328, "learning_rate": 1.849958111573353e-07, "loss": 0.74517202, "num_input_tokens_seen": 155848060, "step": 7203, "time_per_iteration": 2.521639108657837 }, { "auxiliary_loss_clip": 0.01163827, "auxiliary_loss_mlp": 0.0102512, "balance_loss_clip": 1.04526591, "balance_loss_mlp": 1.01801538, "epoch": 0.8662297841640113, "flos": 18224166590880.0, "grad_norm": 1.852592220478655, "language_loss": 0.63965881, "learning_rate": 1.8466874369694074e-07, "loss": 0.66154832, "num_input_tokens_seen": 155865755, "step": 7204, "time_per_iteration": 2.4490034580230713 }, { "auxiliary_loss_clip": 0.01124128, "auxiliary_loss_mlp": 0.0102542, "balance_loss_clip": 1.03892732, "balance_loss_mlp": 1.01850545, "epoch": 0.8663500270546504, "flos": 16362162102240.0, "grad_norm": 3.512038172710416, "language_loss": 0.70081007, "learning_rate": 1.843419516208542e-07, "loss": 0.7223056, "num_input_tokens_seen": 155882680, "step": 7205, "time_per_iteration": 2.5165910720825195 }, { "auxiliary_loss_clip": 0.01157088, "auxiliary_loss_mlp": 0.01026775, "balance_loss_clip": 1.04845667, "balance_loss_mlp": 1.01860332, "epoch": 0.8664702699452895, "flos": 17894434559040.0, "grad_norm": 2.2743313526387894, "language_loss": 0.79329175, "learning_rate": 1.8401543497865047e-07, "loss": 0.81513041, "num_input_tokens_seen": 155900680, "step": 7206, "time_per_iteration": 3.2050271034240723 }, { "auxiliary_loss_clip": 0.01155983, "auxiliary_loss_mlp": 0.00761499, "balance_loss_clip": 1.04520202, "balance_loss_mlp": 1.00057304, "epoch": 0.8665905128359286, "flos": 30736378102080.0, "grad_norm": 2.705734212576686, "language_loss": 0.64129519, "learning_rate": 1.836891938198608e-07, "loss": 0.66047001, "num_input_tokens_seen": 155921105, "step": 7207, "time_per_iteration": 2.578279495239258 }, { "auxiliary_loss_clip": 0.01138627, "auxiliary_loss_mlp": 0.01029039, "balance_loss_clip": 1.0450089, "balance_loss_mlp": 1.02208304, "epoch": 0.8667107557265676, "flos": 18656422378560.0, "grad_norm": 2.5677625968192763, "language_loss": 0.70936435, "learning_rate": 1.8336322819397677e-07, "loss": 0.73104095, "num_input_tokens_seen": 155938640, "step": 7208, "time_per_iteration": 2.488236904144287 }, { "auxiliary_loss_clip": 0.01129733, "auxiliary_loss_mlp": 0.01025928, "balance_loss_clip": 1.04071188, "balance_loss_mlp": 1.01865017, "epoch": 0.8668309986172068, "flos": 20083728726240.0, "grad_norm": 2.029940478747775, "language_loss": 0.62418902, "learning_rate": 1.8303753815044654e-07, "loss": 0.64574558, "num_input_tokens_seen": 155957945, "step": 7209, "time_per_iteration": 2.5612080097198486 }, { "auxiliary_loss_clip": 0.011494, "auxiliary_loss_mlp": 0.01031264, "balance_loss_clip": 1.04442668, "balance_loss_mlp": 1.02322936, "epoch": 0.8669512415078459, "flos": 21615103259040.0, "grad_norm": 2.837080238039327, "language_loss": 0.70499754, "learning_rate": 1.827121237386773e-07, "loss": 0.72680426, "num_input_tokens_seen": 155975390, "step": 7210, "time_per_iteration": 3.2925963401794434 }, { "auxiliary_loss_clip": 0.01143417, "auxiliary_loss_mlp": 0.01022101, "balance_loss_clip": 1.0455271, "balance_loss_mlp": 1.01473367, "epoch": 0.8670714843984849, "flos": 17703603775200.0, "grad_norm": 2.696741060303244, "language_loss": 0.7479865, "learning_rate": 1.8238698500803374e-07, "loss": 0.76964164, "num_input_tokens_seen": 155988155, "step": 7211, "time_per_iteration": 2.4704017639160156 }, { "auxiliary_loss_clip": 0.01052391, "auxiliary_loss_mlp": 0.01000958, "balance_loss_clip": 1.00594449, "balance_loss_mlp": 0.99966472, "epoch": 0.8671917272891241, "flos": 60705494979840.0, "grad_norm": 0.7178797382164612, "language_loss": 0.56349748, "learning_rate": 1.820621220078391e-07, "loss": 0.58403099, "num_input_tokens_seen": 156052065, "step": 7212, "time_per_iteration": 3.152468204498291 }, { "auxiliary_loss_clip": 0.01166337, "auxiliary_loss_mlp": 0.01027695, "balance_loss_clip": 1.04587626, "balance_loss_mlp": 1.02027988, "epoch": 0.8673119701797631, "flos": 20451885516960.0, "grad_norm": 1.9703476751068376, "language_loss": 0.67718017, "learning_rate": 1.8173753478737553e-07, "loss": 0.69912052, "num_input_tokens_seen": 156072500, "step": 7213, "time_per_iteration": 3.2364912033081055 }, { "auxiliary_loss_clip": 0.01168527, "auxiliary_loss_mlp": 0.01026161, "balance_loss_clip": 1.04770875, "balance_loss_mlp": 1.01871395, "epoch": 0.8674322130704022, "flos": 19647414322080.0, "grad_norm": 2.5145884596263293, "language_loss": 0.7982111, "learning_rate": 1.8141322339588205e-07, "loss": 0.82015795, "num_input_tokens_seen": 156089840, "step": 7214, "time_per_iteration": 2.4247424602508545 }, { "auxiliary_loss_clip": 0.01167189, "auxiliary_loss_mlp": 0.01024756, "balance_loss_clip": 1.04825497, "balance_loss_mlp": 1.01825571, "epoch": 0.8675524559610414, "flos": 26025004794240.0, "grad_norm": 3.571493157821137, "language_loss": 0.70131242, "learning_rate": 1.810891878825569e-07, "loss": 0.72323185, "num_input_tokens_seen": 156109815, "step": 7215, "time_per_iteration": 2.500195026397705 }, { "auxiliary_loss_clip": 0.0113846, "auxiliary_loss_mlp": 0.01024024, "balance_loss_clip": 1.0430088, "balance_loss_mlp": 1.01693726, "epoch": 0.8676726988516804, "flos": 15049447608960.0, "grad_norm": 2.11706217957283, "language_loss": 0.72300208, "learning_rate": 1.8076542829655561e-07, "loss": 0.744627, "num_input_tokens_seen": 156128620, "step": 7216, "time_per_iteration": 2.4780380725860596 }, { "auxiliary_loss_clip": 0.01140403, "auxiliary_loss_mlp": 0.01021257, "balance_loss_clip": 1.04691672, "balance_loss_mlp": 1.01389575, "epoch": 0.8677929417423195, "flos": 16288114107360.0, "grad_norm": 2.7074622145453984, "language_loss": 0.79210305, "learning_rate": 1.8044194468699203e-07, "loss": 0.81371963, "num_input_tokens_seen": 156145930, "step": 7217, "time_per_iteration": 2.499378204345703 }, { "auxiliary_loss_clip": 0.01137402, "auxiliary_loss_mlp": 0.01032501, "balance_loss_clip": 1.0464865, "balance_loss_mlp": 1.02560186, "epoch": 0.8679131846329585, "flos": 18844164303840.0, "grad_norm": 2.2049874619374092, "language_loss": 0.75605714, "learning_rate": 1.8011873710293912e-07, "loss": 0.77775621, "num_input_tokens_seen": 156164435, "step": 7218, "time_per_iteration": 2.503309726715088 }, { "auxiliary_loss_clip": 0.01152238, "auxiliary_loss_mlp": 0.01028174, "balance_loss_clip": 1.04764855, "balance_loss_mlp": 1.02090216, "epoch": 0.8680334275235977, "flos": 33620723652000.0, "grad_norm": 1.7628723140819844, "language_loss": 0.69646466, "learning_rate": 1.7979580559342677e-07, "loss": 0.71826875, "num_input_tokens_seen": 156185165, "step": 7219, "time_per_iteration": 2.5962345600128174 }, { "auxiliary_loss_clip": 0.01139356, "auxiliary_loss_mlp": 0.01027289, "balance_loss_clip": 1.04430985, "balance_loss_mlp": 1.02028227, "epoch": 0.8681536704142367, "flos": 24681156684960.0, "grad_norm": 1.6227097579297325, "language_loss": 0.66968554, "learning_rate": 1.7947315020744358e-07, "loss": 0.69135201, "num_input_tokens_seen": 156206260, "step": 7220, "time_per_iteration": 2.564234972000122 }, { "auxiliary_loss_clip": 0.01136134, "auxiliary_loss_mlp": 0.01023388, "balance_loss_clip": 1.04126763, "balance_loss_mlp": 1.01705849, "epoch": 0.8682739133048758, "flos": 20011045575840.0, "grad_norm": 1.72969974631673, "language_loss": 0.80272162, "learning_rate": 1.7915077099393594e-07, "loss": 0.8243168, "num_input_tokens_seen": 156222860, "step": 7221, "time_per_iteration": 2.500401020050049 }, { "auxiliary_loss_clip": 0.01156197, "auxiliary_loss_mlp": 0.01022526, "balance_loss_clip": 1.04614115, "balance_loss_mlp": 1.01459289, "epoch": 0.868394156195515, "flos": 16654762385760.0, "grad_norm": 2.234716793738921, "language_loss": 0.73166513, "learning_rate": 1.788286680018083e-07, "loss": 0.75345242, "num_input_tokens_seen": 156241570, "step": 7222, "time_per_iteration": 2.4770607948303223 }, { "auxiliary_loss_clip": 0.01144524, "auxiliary_loss_mlp": 0.01023922, "balance_loss_clip": 1.04695451, "balance_loss_mlp": 1.01697469, "epoch": 0.868514399086154, "flos": 28001385635520.0, "grad_norm": 1.5662614805711172, "language_loss": 0.72267848, "learning_rate": 1.7850684127992443e-07, "loss": 0.74436283, "num_input_tokens_seen": 156261315, "step": 7223, "time_per_iteration": 2.5724854469299316 }, { "auxiliary_loss_clip": 0.01126698, "auxiliary_loss_mlp": 0.01031214, "balance_loss_clip": 1.04533291, "balance_loss_mlp": 1.02486062, "epoch": 0.8686346419767931, "flos": 20084590733280.0, "grad_norm": 1.5529966432404927, "language_loss": 0.70424318, "learning_rate": 1.7818529087710378e-07, "loss": 0.72582233, "num_input_tokens_seen": 156281670, "step": 7224, "time_per_iteration": 3.3205180168151855 }, { "auxiliary_loss_clip": 0.01152072, "auxiliary_loss_mlp": 0.00761674, "balance_loss_clip": 1.04664838, "balance_loss_mlp": 1.00056696, "epoch": 0.8687548848674322, "flos": 18223520085600.0, "grad_norm": 2.3432501685666134, "language_loss": 0.84305739, "learning_rate": 1.7786401684212637e-07, "loss": 0.8621949, "num_input_tokens_seen": 156300500, "step": 7225, "time_per_iteration": 2.4880168437957764 }, { "auxiliary_loss_clip": 0.01026997, "auxiliary_loss_mlp": 0.00999678, "balance_loss_clip": 1.00969768, "balance_loss_mlp": 0.99852735, "epoch": 0.8688751277580713, "flos": 70457895405120.0, "grad_norm": 0.7365978331369539, "language_loss": 0.55951464, "learning_rate": 1.7754301922372883e-07, "loss": 0.57978129, "num_input_tokens_seen": 156350145, "step": 7226, "time_per_iteration": 3.0169122219085693 }, { "auxiliary_loss_clip": 0.01104724, "auxiliary_loss_mlp": 0.01026785, "balance_loss_clip": 1.04049182, "balance_loss_mlp": 1.01945913, "epoch": 0.8689953706487104, "flos": 26906792427360.0, "grad_norm": 1.9752025476471717, "language_loss": 0.80950367, "learning_rate": 1.7722229807060617e-07, "loss": 0.83081877, "num_input_tokens_seen": 156368725, "step": 7227, "time_per_iteration": 2.6593079566955566 }, { "auxiliary_loss_clip": 0.01113405, "auxiliary_loss_mlp": 0.01028968, "balance_loss_clip": 1.03811765, "balance_loss_mlp": 1.02243853, "epoch": 0.8691156135393495, "flos": 34637389833600.0, "grad_norm": 2.0651876526029755, "language_loss": 0.8130554, "learning_rate": 1.7690185343141172e-07, "loss": 0.83447909, "num_input_tokens_seen": 156388640, "step": 7228, "time_per_iteration": 2.653931140899658 }, { "auxiliary_loss_clip": 0.01140749, "auxiliary_loss_mlp": 0.01027788, "balance_loss_clip": 1.04436707, "balance_loss_mlp": 1.02126145, "epoch": 0.8692358564299886, "flos": 18989817940320.0, "grad_norm": 1.9852033118050203, "language_loss": 0.69445503, "learning_rate": 1.7658168535475615e-07, "loss": 0.71614039, "num_input_tokens_seen": 156406425, "step": 7229, "time_per_iteration": 2.5014123916625977 }, { "auxiliary_loss_clip": 0.01146026, "auxiliary_loss_mlp": 0.01025281, "balance_loss_clip": 1.04795861, "balance_loss_mlp": 1.01781571, "epoch": 0.8693560993206276, "flos": 30370843249440.0, "grad_norm": 1.5304857874867117, "language_loss": 0.64281678, "learning_rate": 1.7626179388920948e-07, "loss": 0.66452986, "num_input_tokens_seen": 156427705, "step": 7230, "time_per_iteration": 2.5910391807556152 }, { "auxiliary_loss_clip": 0.01141264, "auxiliary_loss_mlp": 0.00762076, "balance_loss_clip": 1.04645145, "balance_loss_mlp": 1.00051069, "epoch": 0.8694763422112668, "flos": 27200434302720.0, "grad_norm": 1.6555930335761107, "language_loss": 0.80500364, "learning_rate": 1.7594217908329866e-07, "loss": 0.82403708, "num_input_tokens_seen": 156449890, "step": 7231, "time_per_iteration": 2.581162691116333 }, { "auxiliary_loss_clip": 0.01130937, "auxiliary_loss_mlp": 0.01025889, "balance_loss_clip": 1.04410362, "balance_loss_mlp": 1.01912093, "epoch": 0.8695965851019059, "flos": 26139165645120.0, "grad_norm": 2.0978522722940913, "language_loss": 0.73899031, "learning_rate": 1.7562284098550895e-07, "loss": 0.76055849, "num_input_tokens_seen": 156469600, "step": 7232, "time_per_iteration": 3.3063461780548096 }, { "auxiliary_loss_clip": 0.01036826, "auxiliary_loss_mlp": 0.01000754, "balance_loss_clip": 1.00885248, "balance_loss_mlp": 0.9995265, "epoch": 0.8697168279925449, "flos": 67332627688800.0, "grad_norm": 0.8314911130794878, "language_loss": 0.62251985, "learning_rate": 1.753037796442838e-07, "loss": 0.6428957, "num_input_tokens_seen": 156529040, "step": 7233, "time_per_iteration": 3.0928659439086914 }, { "auxiliary_loss_clip": 0.01167006, "auxiliary_loss_mlp": 0.01030548, "balance_loss_clip": 1.0464977, "balance_loss_mlp": 1.02337718, "epoch": 0.8698370708831841, "flos": 19718696711040.0, "grad_norm": 2.1775304449861257, "language_loss": 0.75324661, "learning_rate": 1.74984995108024e-07, "loss": 0.77522218, "num_input_tokens_seen": 156546970, "step": 7234, "time_per_iteration": 2.4585886001586914 }, { "auxiliary_loss_clip": 0.01156399, "auxiliary_loss_mlp": 0.01024723, "balance_loss_clip": 1.04755771, "balance_loss_mlp": 1.01753485, "epoch": 0.8699573137738231, "flos": 12859973856960.0, "grad_norm": 2.064836709628076, "language_loss": 0.83197463, "learning_rate": 1.7466648742508981e-07, "loss": 0.85378587, "num_input_tokens_seen": 156563155, "step": 7235, "time_per_iteration": 2.473106861114502 }, { "auxiliary_loss_clip": 0.01139109, "auxiliary_loss_mlp": 0.01021986, "balance_loss_clip": 1.0462116, "balance_loss_mlp": 1.01479506, "epoch": 0.8700775566644622, "flos": 17420736987840.0, "grad_norm": 1.9163801885431837, "language_loss": 0.84475148, "learning_rate": 1.7434825664379837e-07, "loss": 0.86636245, "num_input_tokens_seen": 156581660, "step": 7236, "time_per_iteration": 2.5267155170440674 }, { "auxiliary_loss_clip": 0.01155162, "auxiliary_loss_mlp": 0.01024231, "balance_loss_clip": 1.04662037, "balance_loss_mlp": 1.0168395, "epoch": 0.8701977995551013, "flos": 13735224603360.0, "grad_norm": 2.6932584796629504, "language_loss": 0.85683197, "learning_rate": 1.740303028124246e-07, "loss": 0.87862587, "num_input_tokens_seen": 156597720, "step": 7237, "time_per_iteration": 3.282217264175415 }, { "auxiliary_loss_clip": 0.01088946, "auxiliary_loss_mlp": 0.01025569, "balance_loss_clip": 1.0382669, "balance_loss_mlp": 1.0183506, "epoch": 0.8703180424457404, "flos": 30555711817920.0, "grad_norm": 1.814261494105417, "language_loss": 0.75812691, "learning_rate": 1.7371262597920212e-07, "loss": 0.77927214, "num_input_tokens_seen": 156619780, "step": 7238, "time_per_iteration": 2.7458248138427734 }, { "auxiliary_loss_clip": 0.01113222, "auxiliary_loss_mlp": 0.0102714, "balance_loss_clip": 1.04447401, "balance_loss_mlp": 1.02030337, "epoch": 0.8704382853363795, "flos": 19608989563200.0, "grad_norm": 1.5436552539264086, "language_loss": 0.75933218, "learning_rate": 1.7339522619232195e-07, "loss": 0.78073579, "num_input_tokens_seen": 156638160, "step": 7239, "time_per_iteration": 3.352928876876831 }, { "auxiliary_loss_clip": 0.01147657, "auxiliary_loss_mlp": 0.01024663, "balance_loss_clip": 1.04330528, "balance_loss_mlp": 1.01750469, "epoch": 0.8705585282270186, "flos": 26613904808160.0, "grad_norm": 1.8347138701204906, "language_loss": 0.75969195, "learning_rate": 1.730781034999338e-07, "loss": 0.78141522, "num_input_tokens_seen": 156659740, "step": 7240, "time_per_iteration": 2.6566176414489746 }, { "auxiliary_loss_clip": 0.0116485, "auxiliary_loss_mlp": 0.01024575, "balance_loss_clip": 1.0489819, "balance_loss_mlp": 1.01764333, "epoch": 0.8706787711176577, "flos": 34090470357600.0, "grad_norm": 1.9048694785705815, "language_loss": 0.73409927, "learning_rate": 1.7276125795014497e-07, "loss": 0.75599349, "num_input_tokens_seen": 156678190, "step": 7241, "time_per_iteration": 2.5916779041290283 }, { "auxiliary_loss_clip": 0.01142882, "auxiliary_loss_mlp": 0.01029243, "balance_loss_clip": 1.04306674, "balance_loss_mlp": 1.02178061, "epoch": 0.8707990140082967, "flos": 14611516941600.0, "grad_norm": 1.9559470638790513, "language_loss": 0.67265362, "learning_rate": 1.7244468959102054e-07, "loss": 0.69437486, "num_input_tokens_seen": 156695245, "step": 7242, "time_per_iteration": 2.592329263687134 }, { "auxiliary_loss_clip": 0.01154719, "auxiliary_loss_mlp": 0.01023992, "balance_loss_clip": 1.04768896, "balance_loss_mlp": 1.01694059, "epoch": 0.8709192568989359, "flos": 20084159729760.0, "grad_norm": 2.199047366828468, "language_loss": 0.84740031, "learning_rate": 1.7212839847058348e-07, "loss": 0.86918736, "num_input_tokens_seen": 156710375, "step": 7243, "time_per_iteration": 2.4900126457214355 }, { "auxiliary_loss_clip": 0.01102451, "auxiliary_loss_mlp": 0.01022898, "balance_loss_clip": 1.03989553, "balance_loss_mlp": 1.01586485, "epoch": 0.871039499789575, "flos": 16727086366560.0, "grad_norm": 2.041857369313385, "language_loss": 0.73768872, "learning_rate": 1.718123846368147e-07, "loss": 0.75894225, "num_input_tokens_seen": 156729420, "step": 7244, "time_per_iteration": 2.6705548763275146 }, { "auxiliary_loss_clip": 0.01138772, "auxiliary_loss_mlp": 0.00761188, "balance_loss_clip": 1.04618227, "balance_loss_mlp": 1.00052238, "epoch": 0.871159742680214, "flos": 21068794371360.0, "grad_norm": 1.6855230131845518, "language_loss": 0.71389955, "learning_rate": 1.714966481376543e-07, "loss": 0.73289919, "num_input_tokens_seen": 156746100, "step": 7245, "time_per_iteration": 2.526132106781006 }, { "auxiliary_loss_clip": 0.01151606, "auxiliary_loss_mlp": 0.01028603, "balance_loss_clip": 1.04508972, "balance_loss_mlp": 1.02119136, "epoch": 0.8712799855708532, "flos": 28256530918080.0, "grad_norm": 1.8961925701962574, "language_loss": 0.83162904, "learning_rate": 1.7118118902099797e-07, "loss": 0.85343111, "num_input_tokens_seen": 156764185, "step": 7246, "time_per_iteration": 2.5627174377441406 }, { "auxiliary_loss_clip": 0.01155651, "auxiliary_loss_mlp": 0.01023449, "balance_loss_clip": 1.04793549, "balance_loss_mlp": 1.01685357, "epoch": 0.8714002284614922, "flos": 22236681318240.0, "grad_norm": 1.6851755191735047, "language_loss": 0.80725861, "learning_rate": 1.7086600733470146e-07, "loss": 0.82904965, "num_input_tokens_seen": 156784855, "step": 7247, "time_per_iteration": 2.50443696975708 }, { "auxiliary_loss_clip": 0.01149476, "auxiliary_loss_mlp": 0.01022423, "balance_loss_clip": 1.04502368, "balance_loss_mlp": 1.01631093, "epoch": 0.8715204713521313, "flos": 21431922787680.0, "grad_norm": 1.879649650938127, "language_loss": 0.77143729, "learning_rate": 1.7055110312657738e-07, "loss": 0.79315627, "num_input_tokens_seen": 156804350, "step": 7248, "time_per_iteration": 2.5356199741363525 }, { "auxiliary_loss_clip": 0.01136182, "auxiliary_loss_mlp": 0.01029712, "balance_loss_clip": 1.04518986, "balance_loss_mlp": 1.02227294, "epoch": 0.8716407142427703, "flos": 23440442919840.0, "grad_norm": 2.823738545383259, "language_loss": 0.74336725, "learning_rate": 1.702364764443962e-07, "loss": 0.76502621, "num_input_tokens_seen": 156823425, "step": 7249, "time_per_iteration": 2.548161745071411 }, { "auxiliary_loss_clip": 0.01093457, "auxiliary_loss_mlp": 0.0102864, "balance_loss_clip": 1.03901148, "balance_loss_mlp": 1.02110577, "epoch": 0.8717609571334095, "flos": 27958686758400.0, "grad_norm": 2.1026732899306952, "language_loss": 0.72378111, "learning_rate": 1.6992212733588685e-07, "loss": 0.74500209, "num_input_tokens_seen": 156843090, "step": 7250, "time_per_iteration": 3.3711628913879395 }, { "auxiliary_loss_clip": 0.01134486, "auxiliary_loss_mlp": 0.0102804, "balance_loss_clip": 1.04272509, "balance_loss_mlp": 1.02129006, "epoch": 0.8718812000240486, "flos": 25479486079680.0, "grad_norm": 1.8006621066089923, "language_loss": 0.74724251, "learning_rate": 1.6960805584873538e-07, "loss": 0.76886785, "num_input_tokens_seen": 156861090, "step": 7251, "time_per_iteration": 2.676206111907959 }, { "auxiliary_loss_clip": 0.01112491, "auxiliary_loss_mlp": 0.01027058, "balance_loss_clip": 1.03891242, "balance_loss_mlp": 1.0204004, "epoch": 0.8720014429146876, "flos": 23403059752800.0, "grad_norm": 1.5312978733642801, "language_loss": 0.78239942, "learning_rate": 1.6929426203058684e-07, "loss": 0.80379492, "num_input_tokens_seen": 156881515, "step": 7252, "time_per_iteration": 2.616729736328125 }, { "auxiliary_loss_clip": 0.0117042, "auxiliary_loss_mlp": 0.00762806, "balance_loss_clip": 1.04607248, "balance_loss_mlp": 1.00056922, "epoch": 0.8721216858053268, "flos": 24352825414560.0, "grad_norm": 2.1102386580947385, "language_loss": 0.80272734, "learning_rate": 1.689807459290431e-07, "loss": 0.82205963, "num_input_tokens_seen": 156900170, "step": 7253, "time_per_iteration": 2.512561798095703 }, { "auxiliary_loss_clip": 0.01142298, "auxiliary_loss_mlp": 0.01026816, "balance_loss_clip": 1.04735947, "balance_loss_mlp": 1.02011371, "epoch": 0.8722419286959658, "flos": 33869691217440.0, "grad_norm": 1.903195916234081, "language_loss": 0.70849633, "learning_rate": 1.6866750759166437e-07, "loss": 0.73018754, "num_input_tokens_seen": 156920150, "step": 7254, "time_per_iteration": 2.6610116958618164 }, { "auxiliary_loss_clip": 0.01121138, "auxiliary_loss_mlp": 0.01025364, "balance_loss_clip": 1.0405618, "balance_loss_mlp": 1.01811588, "epoch": 0.8723621715866049, "flos": 18369389223840.0, "grad_norm": 2.375054938424995, "language_loss": 0.77056199, "learning_rate": 1.6835454706596865e-07, "loss": 0.792027, "num_input_tokens_seen": 156937980, "step": 7255, "time_per_iteration": 2.5393896102905273 }, { "auxiliary_loss_clip": 0.01168554, "auxiliary_loss_mlp": 0.01030259, "balance_loss_clip": 1.04841197, "balance_loss_mlp": 1.02336323, "epoch": 0.8724824144772441, "flos": 22013352073920.0, "grad_norm": 1.5747160546606476, "language_loss": 0.73492908, "learning_rate": 1.680418643994317e-07, "loss": 0.75691724, "num_input_tokens_seen": 156956550, "step": 7256, "time_per_iteration": 2.528153896331787 }, { "auxiliary_loss_clip": 0.01059998, "auxiliary_loss_mlp": 0.01000751, "balance_loss_clip": 1.00580025, "balance_loss_mlp": 0.99944544, "epoch": 0.8726026573678831, "flos": 66698730112320.0, "grad_norm": 0.8935860757831144, "language_loss": 0.64612055, "learning_rate": 1.6772945963948738e-07, "loss": 0.66672802, "num_input_tokens_seen": 157014715, "step": 7257, "time_per_iteration": 3.093094825744629 }, { "auxiliary_loss_clip": 0.01138959, "auxiliary_loss_mlp": 0.01021758, "balance_loss_clip": 1.04685616, "balance_loss_mlp": 1.01489472, "epoch": 0.8727229002585222, "flos": 13370910927360.0, "grad_norm": 2.768276590946834, "language_loss": 0.77669442, "learning_rate": 1.6741733283352733e-07, "loss": 0.79830158, "num_input_tokens_seen": 157032320, "step": 7258, "time_per_iteration": 3.2691686153411865 }, { "auxiliary_loss_clip": 0.01112706, "auxiliary_loss_mlp": 0.01032082, "balance_loss_clip": 1.04298985, "balance_loss_mlp": 1.02496481, "epoch": 0.8728431431491613, "flos": 21796990719840.0, "grad_norm": 1.4755037254099403, "language_loss": 0.83587039, "learning_rate": 1.6710548402890102e-07, "loss": 0.85731828, "num_input_tokens_seen": 157052845, "step": 7259, "time_per_iteration": 2.745516300201416 }, { "auxiliary_loss_clip": 0.01172441, "auxiliary_loss_mlp": 0.01024735, "balance_loss_clip": 1.04826593, "balance_loss_mlp": 1.01733208, "epoch": 0.8729633860398004, "flos": 36173828657760.0, "grad_norm": 2.137449508866979, "language_loss": 0.66920578, "learning_rate": 1.6679391327291527e-07, "loss": 0.69117761, "num_input_tokens_seen": 157074050, "step": 7260, "time_per_iteration": 2.5863375663757324 }, { "auxiliary_loss_clip": 0.01140262, "auxiliary_loss_mlp": 0.01028823, "balance_loss_clip": 1.04272914, "balance_loss_mlp": 1.02181935, "epoch": 0.8730836289304394, "flos": 16359683832000.0, "grad_norm": 2.8199923763200028, "language_loss": 0.67721784, "learning_rate": 1.6648262061283492e-07, "loss": 0.69890881, "num_input_tokens_seen": 157089350, "step": 7261, "time_per_iteration": 2.5355758666992188 }, { "auxiliary_loss_clip": 0.01125775, "auxiliary_loss_mlp": 0.01027823, "balance_loss_clip": 1.04066324, "balance_loss_mlp": 1.02080452, "epoch": 0.8732038718210786, "flos": 21215130430080.0, "grad_norm": 2.0240604066400314, "language_loss": 0.73225933, "learning_rate": 1.6617160609588353e-07, "loss": 0.75379527, "num_input_tokens_seen": 157108525, "step": 7262, "time_per_iteration": 3.3394775390625 }, { "auxiliary_loss_clip": 0.01139204, "auxiliary_loss_mlp": 0.01022988, "balance_loss_clip": 1.04454374, "balance_loss_mlp": 1.0160265, "epoch": 0.8733241147117177, "flos": 16610698664160.0, "grad_norm": 2.1504084029406365, "language_loss": 0.71899551, "learning_rate": 1.6586086976924163e-07, "loss": 0.74061739, "num_input_tokens_seen": 157124025, "step": 7263, "time_per_iteration": 2.4837899208068848 }, { "auxiliary_loss_clip": 0.01152852, "auxiliary_loss_mlp": 0.01023022, "balance_loss_clip": 1.04462731, "balance_loss_mlp": 1.01657248, "epoch": 0.8734443576023567, "flos": 20193938711520.0, "grad_norm": 1.816800153010401, "language_loss": 0.78107667, "learning_rate": 1.6555041168004747e-07, "loss": 0.80283546, "num_input_tokens_seen": 157143345, "step": 7264, "time_per_iteration": 2.5034537315368652 }, { "auxiliary_loss_clip": 0.01138245, "auxiliary_loss_mlp": 0.01026315, "balance_loss_clip": 1.0468148, "balance_loss_mlp": 1.02027965, "epoch": 0.8735646004929959, "flos": 18041165704320.0, "grad_norm": 1.7628765175257468, "language_loss": 0.688622, "learning_rate": 1.6524023187539715e-07, "loss": 0.7102676, "num_input_tokens_seen": 157161630, "step": 7265, "time_per_iteration": 3.254758358001709 }, { "auxiliary_loss_clip": 0.01139243, "auxiliary_loss_mlp": 0.01027852, "balance_loss_clip": 1.04372454, "balance_loss_mlp": 1.0207566, "epoch": 0.873684843383635, "flos": 20262347743680.0, "grad_norm": 1.7156688606376749, "language_loss": 0.74719143, "learning_rate": 1.649303304023446e-07, "loss": 0.76886237, "num_input_tokens_seen": 157181385, "step": 7266, "time_per_iteration": 2.5353407859802246 }, { "auxiliary_loss_clip": 0.01122035, "auxiliary_loss_mlp": 0.01026569, "balance_loss_clip": 1.04501843, "balance_loss_mlp": 1.02039385, "epoch": 0.873805086274274, "flos": 16947290835360.0, "grad_norm": 1.6153705191890642, "language_loss": 0.78849977, "learning_rate": 1.6462070730790246e-07, "loss": 0.80998576, "num_input_tokens_seen": 157200545, "step": 7267, "time_per_iteration": 2.5333974361419678 }, { "auxiliary_loss_clip": 0.01130054, "auxiliary_loss_mlp": 0.01024351, "balance_loss_clip": 1.03881609, "balance_loss_mlp": 1.01743937, "epoch": 0.8739253291649132, "flos": 18041273455200.0, "grad_norm": 3.0015797554999684, "language_loss": 0.78246272, "learning_rate": 1.6431136263903912e-07, "loss": 0.8040067, "num_input_tokens_seen": 157219545, "step": 7268, "time_per_iteration": 2.4909756183624268 }, { "auxiliary_loss_clip": 0.01156262, "auxiliary_loss_mlp": 0.00761806, "balance_loss_clip": 1.04400587, "balance_loss_mlp": 1.00053978, "epoch": 0.8740455720555522, "flos": 21325088996640.0, "grad_norm": 2.430729833015466, "language_loss": 0.73581487, "learning_rate": 1.6400229644268282e-07, "loss": 0.75499558, "num_input_tokens_seen": 157237900, "step": 7269, "time_per_iteration": 2.4927978515625 }, { "auxiliary_loss_clip": 0.01124441, "auxiliary_loss_mlp": 0.01026808, "balance_loss_clip": 1.04750395, "balance_loss_mlp": 1.01955402, "epoch": 0.8741658149461913, "flos": 15158687836320.0, "grad_norm": 1.897129863589245, "language_loss": 0.81045598, "learning_rate": 1.6369350876571852e-07, "loss": 0.83196849, "num_input_tokens_seen": 157256055, "step": 7270, "time_per_iteration": 2.534932851791382 }, { "auxiliary_loss_clip": 0.01107508, "auxiliary_loss_mlp": 0.01026176, "balance_loss_clip": 1.04059052, "balance_loss_mlp": 1.01958108, "epoch": 0.8742860578368304, "flos": 23039859502560.0, "grad_norm": 2.162313310256634, "language_loss": 0.81501919, "learning_rate": 1.6338499965498874e-07, "loss": 0.83635604, "num_input_tokens_seen": 157274785, "step": 7271, "time_per_iteration": 2.594282627105713 }, { "auxiliary_loss_clip": 0.01121545, "auxiliary_loss_mlp": 0.01026227, "balance_loss_clip": 1.04250467, "balance_loss_mlp": 1.01874661, "epoch": 0.8744063007274695, "flos": 28145351174880.0, "grad_norm": 1.4741534368874631, "language_loss": 0.7747618, "learning_rate": 1.630767691572943e-07, "loss": 0.7962395, "num_input_tokens_seen": 157294805, "step": 7272, "time_per_iteration": 2.624504804611206 }, { "auxiliary_loss_clip": 0.01042614, "auxiliary_loss_mlp": 0.0100006, "balance_loss_clip": 1.00610888, "balance_loss_mlp": 0.99867707, "epoch": 0.8745265436181086, "flos": 64034086193760.0, "grad_norm": 0.7472235471316352, "language_loss": 0.53527421, "learning_rate": 1.6276881731939306e-07, "loss": 0.5557009, "num_input_tokens_seen": 157356695, "step": 7273, "time_per_iteration": 3.1867923736572266 }, { "auxiliary_loss_clip": 0.01151123, "auxiliary_loss_mlp": 0.0102747, "balance_loss_clip": 1.04689407, "balance_loss_mlp": 1.02095509, "epoch": 0.8746467865087477, "flos": 28658622847680.0, "grad_norm": 1.9904306065201773, "language_loss": 0.75444281, "learning_rate": 1.6246114418800193e-07, "loss": 0.77622879, "num_input_tokens_seen": 157376975, "step": 7274, "time_per_iteration": 2.5550503730773926 }, { "auxiliary_loss_clip": 0.01148784, "auxiliary_loss_mlp": 0.01030099, "balance_loss_clip": 1.04484725, "balance_loss_mlp": 1.02309513, "epoch": 0.8747670293993868, "flos": 23985853883520.0, "grad_norm": 1.7500800696414474, "language_loss": 0.76580572, "learning_rate": 1.6215374980979423e-07, "loss": 0.78759456, "num_input_tokens_seen": 157397385, "step": 7275, "time_per_iteration": 2.533125400543213 }, { "auxiliary_loss_clip": 0.01151077, "auxiliary_loss_mlp": 0.01025858, "balance_loss_clip": 1.0479393, "balance_loss_mlp": 1.01915205, "epoch": 0.8748872722900258, "flos": 45221630177280.0, "grad_norm": 1.9929946712077775, "language_loss": 0.68708682, "learning_rate": 1.6184663423140133e-07, "loss": 0.70885611, "num_input_tokens_seen": 157417685, "step": 7276, "time_per_iteration": 3.385866641998291 }, { "auxiliary_loss_clip": 0.01114045, "auxiliary_loss_mlp": 0.01028228, "balance_loss_clip": 1.04170108, "balance_loss_mlp": 1.02131724, "epoch": 0.875007515180665, "flos": 19754284030080.0, "grad_norm": 1.9344234556318103, "language_loss": 0.63957936, "learning_rate": 1.615397974994126e-07, "loss": 0.66100204, "num_input_tokens_seen": 157435490, "step": 7277, "time_per_iteration": 2.5832278728485107 }, { "auxiliary_loss_clip": 0.01166851, "auxiliary_loss_mlp": 0.01027366, "balance_loss_clip": 1.04826581, "balance_loss_mlp": 1.02077949, "epoch": 0.875127758071304, "flos": 22710738059040.0, "grad_norm": 1.4883517411422211, "language_loss": 0.80996078, "learning_rate": 1.6123323966037438e-07, "loss": 0.83190292, "num_input_tokens_seen": 157454010, "step": 7278, "time_per_iteration": 2.4667584896087646 }, { "auxiliary_loss_clip": 0.01166824, "auxiliary_loss_mlp": 0.01019399, "balance_loss_clip": 1.04831541, "balance_loss_mlp": 1.01277089, "epoch": 0.8752480009619431, "flos": 23403849925920.0, "grad_norm": 1.857785241166778, "language_loss": 0.78370857, "learning_rate": 1.6092696076079216e-07, "loss": 0.80557078, "num_input_tokens_seen": 157472385, "step": 7279, "time_per_iteration": 2.4634008407592773 }, { "auxiliary_loss_clip": 0.01112194, "auxiliary_loss_mlp": 0.01019118, "balance_loss_clip": 1.04089236, "balance_loss_mlp": 1.01258576, "epoch": 0.8753682438525822, "flos": 26213105889120.0, "grad_norm": 1.6704426798678038, "language_loss": 0.73624539, "learning_rate": 1.6062096084712785e-07, "loss": 0.75755847, "num_input_tokens_seen": 157493735, "step": 7280, "time_per_iteration": 2.5851922035217285 }, { "auxiliary_loss_clip": 0.01128161, "auxiliary_loss_mlp": 0.00762034, "balance_loss_clip": 1.04028797, "balance_loss_mlp": 1.00058496, "epoch": 0.8754884867432213, "flos": 23326748989440.0, "grad_norm": 1.8736488331393004, "language_loss": 0.70596206, "learning_rate": 1.6031523996580098e-07, "loss": 0.72486401, "num_input_tokens_seen": 157511295, "step": 7281, "time_per_iteration": 2.533522605895996 }, { "auxiliary_loss_clip": 0.01132502, "auxiliary_loss_mlp": 0.01029321, "balance_loss_clip": 1.04290497, "balance_loss_mlp": 1.02191567, "epoch": 0.8756087296338604, "flos": 12495229177440.0, "grad_norm": 10.59126886853085, "language_loss": 0.66358906, "learning_rate": 1.6000979816318981e-07, "loss": 0.68520731, "num_input_tokens_seen": 157529760, "step": 7282, "time_per_iteration": 2.5200252532958984 }, { "auxiliary_loss_clip": 0.01150353, "auxiliary_loss_mlp": 0.0102706, "balance_loss_clip": 1.04711723, "balance_loss_mlp": 1.02012825, "epoch": 0.8757289725244994, "flos": 18952901693760.0, "grad_norm": 2.200773838734316, "language_loss": 0.75255823, "learning_rate": 1.5970463548562886e-07, "loss": 0.7743324, "num_input_tokens_seen": 157548915, "step": 7283, "time_per_iteration": 2.4600255489349365 }, { "auxiliary_loss_clip": 0.01137735, "auxiliary_loss_mlp": 0.01021435, "balance_loss_clip": 1.04521573, "balance_loss_mlp": 1.01478302, "epoch": 0.8758492154151386, "flos": 25265962165440.0, "grad_norm": 1.629820043881026, "language_loss": 0.71016008, "learning_rate": 1.5939975197941192e-07, "loss": 0.7317518, "num_input_tokens_seen": 157570570, "step": 7284, "time_per_iteration": 3.3161351680755615 }, { "auxiliary_loss_clip": 0.01041721, "auxiliary_loss_mlp": 0.00999819, "balance_loss_clip": 1.00576937, "balance_loss_mlp": 0.99841195, "epoch": 0.8759694583057777, "flos": 65571674360640.0, "grad_norm": 0.8294665373638815, "language_loss": 0.53391153, "learning_rate": 1.5909514769078892e-07, "loss": 0.55432689, "num_input_tokens_seen": 157635675, "step": 7285, "time_per_iteration": 3.1857662200927734 }, { "auxiliary_loss_clip": 0.01120988, "auxiliary_loss_mlp": 0.01022289, "balance_loss_clip": 1.04742432, "balance_loss_mlp": 1.01555932, "epoch": 0.8760897011964167, "flos": 25446197446080.0, "grad_norm": 1.8543166793357393, "language_loss": 0.77396774, "learning_rate": 1.5879082266596867e-07, "loss": 0.7954005, "num_input_tokens_seen": 157657015, "step": 7286, "time_per_iteration": 2.5815236568450928 }, { "auxiliary_loss_clip": 0.01134161, "auxiliary_loss_mlp": 0.0103157, "balance_loss_clip": 1.0405848, "balance_loss_mlp": 1.02484655, "epoch": 0.8762099440870559, "flos": 28984835017440.0, "grad_norm": 2.0926726926253596, "language_loss": 0.71966732, "learning_rate": 1.5848677695111645e-07, "loss": 0.74132466, "num_input_tokens_seen": 157678615, "step": 7287, "time_per_iteration": 2.5666067600250244 }, { "auxiliary_loss_clip": 0.01136791, "auxiliary_loss_mlp": 0.01028181, "balance_loss_clip": 1.04723299, "balance_loss_mlp": 1.02064109, "epoch": 0.8763301869776949, "flos": 21609464296320.0, "grad_norm": 3.1385478414021395, "language_loss": 0.69732457, "learning_rate": 1.5818301059235562e-07, "loss": 0.71897429, "num_input_tokens_seen": 157693790, "step": 7288, "time_per_iteration": 2.5157196521759033 }, { "auxiliary_loss_clip": 0.01139329, "auxiliary_loss_mlp": 0.01031745, "balance_loss_clip": 1.04506373, "balance_loss_mlp": 1.02449131, "epoch": 0.876450429868334, "flos": 24644420023200.0, "grad_norm": 1.4776203329644533, "language_loss": 0.81602812, "learning_rate": 1.578795236357684e-07, "loss": 0.83773887, "num_input_tokens_seen": 157715255, "step": 7289, "time_per_iteration": 3.303252696990967 }, { "auxiliary_loss_clip": 0.01142465, "auxiliary_loss_mlp": 0.01027878, "balance_loss_clip": 1.04714167, "balance_loss_mlp": 1.02138066, "epoch": 0.8765706727589732, "flos": 20260049058240.0, "grad_norm": 2.2194402082882414, "language_loss": 0.85624433, "learning_rate": 1.5757631612739218e-07, "loss": 0.87794775, "num_input_tokens_seen": 157728800, "step": 7290, "time_per_iteration": 3.2843334674835205 }, { "auxiliary_loss_clip": 0.01059844, "auxiliary_loss_mlp": 0.01001433, "balance_loss_clip": 1.0057627, "balance_loss_mlp": 1.00016975, "epoch": 0.8766909156496122, "flos": 71371175823840.0, "grad_norm": 0.7811439977355136, "language_loss": 0.61510134, "learning_rate": 1.572733881132242e-07, "loss": 0.63571417, "num_input_tokens_seen": 157789445, "step": 7291, "time_per_iteration": 3.1111228466033936 }, { "auxiliary_loss_clip": 0.01026528, "auxiliary_loss_mlp": 0.01003538, "balance_loss_clip": 1.00770688, "balance_loss_mlp": 1.0023396, "epoch": 0.8768111585402513, "flos": 69523502202240.0, "grad_norm": 0.784442116857599, "language_loss": 0.58612692, "learning_rate": 1.5697073963921814e-07, "loss": 0.60642755, "num_input_tokens_seen": 157848685, "step": 7292, "time_per_iteration": 3.0775842666625977 }, { "auxiliary_loss_clip": 0.0115497, "auxiliary_loss_mlp": 0.01028747, "balance_loss_clip": 1.04711378, "balance_loss_mlp": 1.02114201, "epoch": 0.8769314014308904, "flos": 18838561258080.0, "grad_norm": 2.2525512876079827, "language_loss": 0.84918821, "learning_rate": 1.566683707512857e-07, "loss": 0.87102538, "num_input_tokens_seen": 157866360, "step": 7293, "time_per_iteration": 2.4664652347564697 }, { "auxiliary_loss_clip": 0.01135734, "auxiliary_loss_mlp": 0.01027169, "balance_loss_clip": 1.0440228, "balance_loss_mlp": 1.01980209, "epoch": 0.8770516443215295, "flos": 14976405288960.0, "grad_norm": 2.306937947498895, "language_loss": 0.79394233, "learning_rate": 1.5636628149529553e-07, "loss": 0.81557143, "num_input_tokens_seen": 157884150, "step": 7294, "time_per_iteration": 2.499753475189209 }, { "auxiliary_loss_clip": 0.01138835, "auxiliary_loss_mlp": 0.01022662, "balance_loss_clip": 1.04429662, "balance_loss_mlp": 1.01617718, "epoch": 0.8771718872121685, "flos": 31649658520800.0, "grad_norm": 2.1070463833383077, "language_loss": 0.79484022, "learning_rate": 1.560644719170743e-07, "loss": 0.81645525, "num_input_tokens_seen": 157905020, "step": 7295, "time_per_iteration": 2.577704668045044 }, { "auxiliary_loss_clip": 0.01123123, "auxiliary_loss_mlp": 0.01031078, "balance_loss_clip": 1.04087007, "balance_loss_mlp": 1.02329695, "epoch": 0.8772921301028077, "flos": 36095470627680.0, "grad_norm": 2.3434297431837434, "language_loss": 0.72094178, "learning_rate": 1.5576294206240692e-07, "loss": 0.74248379, "num_input_tokens_seen": 157924545, "step": 7296, "time_per_iteration": 2.704747438430786 }, { "auxiliary_loss_clip": 0.01137763, "auxiliary_loss_mlp": 0.01030885, "balance_loss_clip": 1.0451386, "balance_loss_mlp": 1.02374184, "epoch": 0.8774123729934468, "flos": 57116968751040.0, "grad_norm": 2.0265916293437454, "language_loss": 0.67564195, "learning_rate": 1.5546169197703507e-07, "loss": 0.69732845, "num_input_tokens_seen": 157950820, "step": 7297, "time_per_iteration": 2.879986047744751 }, { "auxiliary_loss_clip": 0.01144527, "auxiliary_loss_mlp": 0.01022982, "balance_loss_clip": 1.04341769, "balance_loss_mlp": 1.01576996, "epoch": 0.8775326158840858, "flos": 23914499660640.0, "grad_norm": 3.12539294653634, "language_loss": 0.77548504, "learning_rate": 1.5516072170665774e-07, "loss": 0.79716015, "num_input_tokens_seen": 157968790, "step": 7298, "time_per_iteration": 2.529053211212158 }, { "auxiliary_loss_clip": 0.01154819, "auxiliary_loss_mlp": 0.01025873, "balance_loss_clip": 1.04588628, "balance_loss_mlp": 1.01897061, "epoch": 0.877652858774725, "flos": 17123287914720.0, "grad_norm": 2.030183424593269, "language_loss": 0.86653477, "learning_rate": 1.5486003129693214e-07, "loss": 0.88834167, "num_input_tokens_seen": 157986155, "step": 7299, "time_per_iteration": 2.4974865913391113 }, { "auxiliary_loss_clip": 0.01157891, "auxiliary_loss_mlp": 0.01021719, "balance_loss_clip": 1.04757118, "balance_loss_mlp": 1.01509678, "epoch": 0.877773101665364, "flos": 16508965081440.0, "grad_norm": 1.9777786879168249, "language_loss": 0.78352684, "learning_rate": 1.545596207934725e-07, "loss": 0.80532295, "num_input_tokens_seen": 158004640, "step": 7300, "time_per_iteration": 2.4593186378479004 }, { "auxiliary_loss_clip": 0.01133758, "auxiliary_loss_mlp": 0.01019953, "balance_loss_clip": 1.04329228, "balance_loss_mlp": 1.01362324, "epoch": 0.8778933445560031, "flos": 22053213511200.0, "grad_norm": 1.7052359622851612, "language_loss": 0.77835107, "learning_rate": 1.5425949024185147e-07, "loss": 0.79988825, "num_input_tokens_seen": 158024665, "step": 7301, "time_per_iteration": 2.5699350833892822 }, { "auxiliary_loss_clip": 0.01139489, "auxiliary_loss_mlp": 0.01022223, "balance_loss_clip": 1.04208255, "balance_loss_mlp": 1.0152905, "epoch": 0.8780135874466423, "flos": 22564761169920.0, "grad_norm": 5.336576252296814, "language_loss": 0.67834318, "learning_rate": 1.5395963968759818e-07, "loss": 0.69996029, "num_input_tokens_seen": 158044940, "step": 7302, "time_per_iteration": 3.2723772525787354 }, { "auxiliary_loss_clip": 0.01140246, "auxiliary_loss_mlp": 0.0102781, "balance_loss_clip": 1.04428291, "balance_loss_mlp": 1.02103853, "epoch": 0.8781338303372813, "flos": 61532006411520.0, "grad_norm": 1.7032128339048271, "language_loss": 0.64567524, "learning_rate": 1.536600691761998e-07, "loss": 0.66735584, "num_input_tokens_seen": 158070770, "step": 7303, "time_per_iteration": 2.9012694358825684 }, { "auxiliary_loss_clip": 0.01128094, "auxiliary_loss_mlp": 0.01023231, "balance_loss_clip": 1.04559779, "balance_loss_mlp": 1.01650143, "epoch": 0.8782540732279204, "flos": 22674755653440.0, "grad_norm": 1.913531164692404, "language_loss": 0.71810663, "learning_rate": 1.5336077875310084e-07, "loss": 0.73961985, "num_input_tokens_seen": 158089995, "step": 7304, "time_per_iteration": 2.5517680644989014 }, { "auxiliary_loss_clip": 0.01116987, "auxiliary_loss_mlp": 0.01026286, "balance_loss_clip": 1.0421319, "balance_loss_mlp": 1.01916885, "epoch": 0.8783743161185595, "flos": 16070351991840.0, "grad_norm": 2.15683775076155, "language_loss": 0.74284023, "learning_rate": 1.5306176846370321e-07, "loss": 0.76427293, "num_input_tokens_seen": 158108140, "step": 7305, "time_per_iteration": 2.5606396198272705 }, { "auxiliary_loss_clip": 0.01146923, "auxiliary_loss_mlp": 0.01032931, "balance_loss_clip": 1.04397142, "balance_loss_mlp": 1.02623463, "epoch": 0.8784945590091986, "flos": 26067883256160.0, "grad_norm": 2.1743364331356334, "language_loss": 0.73830283, "learning_rate": 1.5276303835336712e-07, "loss": 0.76010138, "num_input_tokens_seen": 158128680, "step": 7306, "time_per_iteration": 2.5463433265686035 }, { "auxiliary_loss_clip": 0.0105154, "auxiliary_loss_mlp": 0.01001422, "balance_loss_clip": 1.00568354, "balance_loss_mlp": 1.00010514, "epoch": 0.8786148018998376, "flos": 62720653361280.0, "grad_norm": 0.766103934475095, "language_loss": 0.53534555, "learning_rate": 1.524645884674094e-07, "loss": 0.55587518, "num_input_tokens_seen": 158185610, "step": 7307, "time_per_iteration": 3.078855514526367 }, { "auxiliary_loss_clip": 0.01168326, "auxiliary_loss_mlp": 0.00763212, "balance_loss_clip": 1.04665601, "balance_loss_mlp": 1.00061476, "epoch": 0.8787350447904768, "flos": 21652737844800.0, "grad_norm": 1.822663759573638, "language_loss": 0.79164374, "learning_rate": 1.521664188511047e-07, "loss": 0.8109591, "num_input_tokens_seen": 158205635, "step": 7308, "time_per_iteration": 2.4674999713897705 }, { "auxiliary_loss_clip": 0.01141266, "auxiliary_loss_mlp": 0.00762111, "balance_loss_clip": 1.04835176, "balance_loss_mlp": 1.00059342, "epoch": 0.8788552876811159, "flos": 25478480404800.0, "grad_norm": 1.883401085256416, "language_loss": 0.80265498, "learning_rate": 1.518685295496851e-07, "loss": 0.82168877, "num_input_tokens_seen": 158223495, "step": 7309, "time_per_iteration": 2.57497239112854 }, { "auxiliary_loss_clip": 0.01152769, "auxiliary_loss_mlp": 0.01021898, "balance_loss_clip": 1.04387558, "balance_loss_mlp": 1.01527572, "epoch": 0.8789755305717549, "flos": 22310226475680.0, "grad_norm": 1.6799265650415618, "language_loss": 0.85399461, "learning_rate": 1.5157092060833975e-07, "loss": 0.87574124, "num_input_tokens_seen": 158243145, "step": 7310, "time_per_iteration": 3.310140609741211 }, { "auxiliary_loss_clip": 0.01131064, "auxiliary_loss_mlp": 0.0102198, "balance_loss_clip": 1.04147041, "balance_loss_mlp": 1.01495576, "epoch": 0.879095773462394, "flos": 29310975353280.0, "grad_norm": 1.676250849041821, "language_loss": 0.66089511, "learning_rate": 1.5127359207221658e-07, "loss": 0.68242562, "num_input_tokens_seen": 158262625, "step": 7311, "time_per_iteration": 2.5732553005218506 }, { "auxiliary_loss_clip": 0.01084104, "auxiliary_loss_mlp": 0.01024682, "balance_loss_clip": 1.03418326, "balance_loss_mlp": 1.01710284, "epoch": 0.8792160163530331, "flos": 16690026452160.0, "grad_norm": 1.9007843662532995, "language_loss": 0.73577368, "learning_rate": 1.5097654398641923e-07, "loss": 0.75686157, "num_input_tokens_seen": 158280530, "step": 7312, "time_per_iteration": 2.5958712100982666 }, { "auxiliary_loss_clip": 0.01154561, "auxiliary_loss_mlp": 0.01026549, "balance_loss_clip": 1.04695451, "balance_loss_mlp": 1.02023971, "epoch": 0.8793362592436722, "flos": 24499305141120.0, "grad_norm": 1.3100254172055306, "language_loss": 0.73106843, "learning_rate": 1.5067977639601014e-07, "loss": 0.75287956, "num_input_tokens_seen": 158303290, "step": 7313, "time_per_iteration": 2.520559549331665 }, { "auxiliary_loss_clip": 0.01138644, "auxiliary_loss_mlp": 0.01023993, "balance_loss_clip": 1.04624021, "balance_loss_mlp": 1.01746655, "epoch": 0.8794565021343113, "flos": 14538402787680.0, "grad_norm": 4.482821671207022, "language_loss": 0.70924592, "learning_rate": 1.5038328934600864e-07, "loss": 0.73087227, "num_input_tokens_seen": 158319925, "step": 7314, "time_per_iteration": 3.2649848461151123 }, { "auxiliary_loss_clip": 0.01139364, "auxiliary_loss_mlp": 0.01024948, "balance_loss_clip": 1.04607725, "balance_loss_mlp": 1.01843333, "epoch": 0.8795767450249504, "flos": 39530291432640.0, "grad_norm": 1.9547999574896995, "language_loss": 0.70022321, "learning_rate": 1.5008708288139161e-07, "loss": 0.72186637, "num_input_tokens_seen": 158342285, "step": 7315, "time_per_iteration": 2.6571576595306396 }, { "auxiliary_loss_clip": 0.0115655, "auxiliary_loss_mlp": 0.01026135, "balance_loss_clip": 1.04908705, "balance_loss_mlp": 1.01879811, "epoch": 0.8796969879155895, "flos": 22960675382400.0, "grad_norm": 1.8911161135332324, "language_loss": 0.73300242, "learning_rate": 1.497911570470931e-07, "loss": 0.75482929, "num_input_tokens_seen": 158362290, "step": 7316, "time_per_iteration": 3.2437775135040283 }, { "auxiliary_loss_clip": 0.01115488, "auxiliary_loss_mlp": 0.0102131, "balance_loss_clip": 1.0424186, "balance_loss_mlp": 1.01415706, "epoch": 0.8798172308062285, "flos": 28362430868160.0, "grad_norm": 1.7002565287398148, "language_loss": 0.85628247, "learning_rate": 1.494955118880048e-07, "loss": 0.87765044, "num_input_tokens_seen": 158383275, "step": 7317, "time_per_iteration": 2.642019271850586 }, { "auxiliary_loss_clip": 0.01155224, "auxiliary_loss_mlp": 0.01022691, "balance_loss_clip": 1.04628158, "balance_loss_mlp": 1.01603913, "epoch": 0.8799374736968677, "flos": 23988978659040.0, "grad_norm": 1.6449176290233367, "language_loss": 0.72942978, "learning_rate": 1.4920014744897634e-07, "loss": 0.7512089, "num_input_tokens_seen": 158402690, "step": 7318, "time_per_iteration": 2.50215220451355 }, { "auxiliary_loss_clip": 0.01132247, "auxiliary_loss_mlp": 0.01028388, "balance_loss_clip": 1.04515648, "balance_loss_mlp": 1.02149153, "epoch": 0.8800577165875068, "flos": 25630275841440.0, "grad_norm": 1.7343485666437786, "language_loss": 0.86086059, "learning_rate": 1.4890506377481392e-07, "loss": 0.88246691, "num_input_tokens_seen": 158421780, "step": 7319, "time_per_iteration": 2.5469765663146973 }, { "auxiliary_loss_clip": 0.01091726, "auxiliary_loss_mlp": 0.01026112, "balance_loss_clip": 1.04196048, "balance_loss_mlp": 1.01972532, "epoch": 0.8801779594781458, "flos": 23440335168960.0, "grad_norm": 1.4905371375446117, "language_loss": 0.63739908, "learning_rate": 1.486102609102815e-07, "loss": 0.6585775, "num_input_tokens_seen": 158442330, "step": 7320, "time_per_iteration": 2.640899181365967 }, { "auxiliary_loss_clip": 0.01134697, "auxiliary_loss_mlp": 0.01029766, "balance_loss_clip": 1.04508877, "balance_loss_mlp": 1.022596, "epoch": 0.880298202368785, "flos": 11508583186080.0, "grad_norm": 2.475174107569565, "language_loss": 0.85826051, "learning_rate": 1.483157389001004e-07, "loss": 0.8799051, "num_input_tokens_seen": 158459890, "step": 7321, "time_per_iteration": 2.485372543334961 }, { "auxiliary_loss_clip": 0.0113828, "auxiliary_loss_mlp": 0.01031115, "balance_loss_clip": 1.04230607, "balance_loss_mlp": 1.02387881, "epoch": 0.880418445259424, "flos": 22671451293120.0, "grad_norm": 2.591186982283075, "language_loss": 0.78418875, "learning_rate": 1.4802149778894933e-07, "loss": 0.80588269, "num_input_tokens_seen": 158478680, "step": 7322, "time_per_iteration": 2.522501230239868 }, { "auxiliary_loss_clip": 0.0114086, "auxiliary_loss_mlp": 0.01023541, "balance_loss_clip": 1.04016185, "balance_loss_mlp": 1.01712787, "epoch": 0.8805386881500631, "flos": 20522162231040.0, "grad_norm": 1.679402592499138, "language_loss": 0.87878746, "learning_rate": 1.4772753762146484e-07, "loss": 0.90043151, "num_input_tokens_seen": 158497935, "step": 7323, "time_per_iteration": 2.4920406341552734 }, { "auxiliary_loss_clip": 0.01147924, "auxiliary_loss_mlp": 0.01025118, "balance_loss_clip": 1.04451191, "balance_loss_mlp": 1.01763475, "epoch": 0.8806589310407023, "flos": 36538896589920.0, "grad_norm": 1.652183851427415, "language_loss": 0.70647663, "learning_rate": 1.474338584422401e-07, "loss": 0.72820711, "num_input_tokens_seen": 158523145, "step": 7324, "time_per_iteration": 2.64780855178833 }, { "auxiliary_loss_clip": 0.01151163, "auxiliary_loss_mlp": 0.01024949, "balance_loss_clip": 1.04797614, "balance_loss_mlp": 1.0180701, "epoch": 0.8807791739313413, "flos": 23440191501120.0, "grad_norm": 1.6097112614405304, "language_loss": 0.75967717, "learning_rate": 1.4714046029582595e-07, "loss": 0.78143823, "num_input_tokens_seen": 158542210, "step": 7325, "time_per_iteration": 2.514148473739624 }, { "auxiliary_loss_clip": 0.0112752, "auxiliary_loss_mlp": 0.01023367, "balance_loss_clip": 1.04227877, "balance_loss_mlp": 1.01626205, "epoch": 0.8808994168219804, "flos": 25956847180800.0, "grad_norm": 2.1459383296000034, "language_loss": 0.75471371, "learning_rate": 1.46847343226731e-07, "loss": 0.77622259, "num_input_tokens_seen": 158563250, "step": 7326, "time_per_iteration": 2.589104652404785 }, { "auxiliary_loss_clip": 0.01157455, "auxiliary_loss_mlp": 0.01027518, "balance_loss_clip": 1.04663038, "balance_loss_mlp": 1.02092886, "epoch": 0.8810196597126195, "flos": 17092082464800.0, "grad_norm": 2.0155322157697397, "language_loss": 0.69372576, "learning_rate": 1.465545072794203e-07, "loss": 0.71557552, "num_input_tokens_seen": 158581125, "step": 7327, "time_per_iteration": 2.4695026874542236 }, { "auxiliary_loss_clip": 0.01107984, "auxiliary_loss_mlp": 0.01025036, "balance_loss_clip": 1.04532194, "balance_loss_mlp": 1.01861954, "epoch": 0.8811399026032586, "flos": 23002835505120.0, "grad_norm": 1.7203013268275091, "language_loss": 0.75845456, "learning_rate": 1.4626195249831774e-07, "loss": 0.77978474, "num_input_tokens_seen": 158602025, "step": 7328, "time_per_iteration": 3.372101306915283 }, { "auxiliary_loss_clip": 0.01150635, "auxiliary_loss_mlp": 0.01021227, "balance_loss_clip": 1.04409695, "balance_loss_mlp": 1.01440549, "epoch": 0.8812601454938976, "flos": 14463816038400.0, "grad_norm": 2.559827904089695, "language_loss": 0.71957695, "learning_rate": 1.4596967892780244e-07, "loss": 0.74129558, "num_input_tokens_seen": 158618355, "step": 7329, "time_per_iteration": 2.4649384021759033 }, { "auxiliary_loss_clip": 0.01165371, "auxiliary_loss_mlp": 0.01027072, "balance_loss_clip": 1.04663634, "balance_loss_mlp": 1.02070594, "epoch": 0.8813803883845368, "flos": 22493227362240.0, "grad_norm": 1.7619592278545035, "language_loss": 0.74511898, "learning_rate": 1.4567768661221314e-07, "loss": 0.76704347, "num_input_tokens_seen": 158638925, "step": 7330, "time_per_iteration": 2.4626200199127197 }, { "auxiliary_loss_clip": 0.01158411, "auxiliary_loss_mlp": 0.00761625, "balance_loss_clip": 1.04775608, "balance_loss_mlp": 1.00062943, "epoch": 0.8815006312751759, "flos": 21506904623520.0, "grad_norm": 2.2503097631721976, "language_loss": 0.74481899, "learning_rate": 1.4538597559584442e-07, "loss": 0.76401925, "num_input_tokens_seen": 158656715, "step": 7331, "time_per_iteration": 2.5013577938079834 }, { "auxiliary_loss_clip": 0.01133011, "auxiliary_loss_mlp": 0.01025559, "balance_loss_clip": 1.0424943, "balance_loss_mlp": 1.01840091, "epoch": 0.8816208741658149, "flos": 22784570552160.0, "grad_norm": 2.0767140020592354, "language_loss": 0.79118007, "learning_rate": 1.4509454592294823e-07, "loss": 0.81276578, "num_input_tokens_seen": 158677200, "step": 7332, "time_per_iteration": 2.5502610206604004 }, { "auxiliary_loss_clip": 0.01128464, "auxiliary_loss_mlp": 0.00762199, "balance_loss_clip": 1.0465982, "balance_loss_mlp": 1.00061822, "epoch": 0.8817411170564541, "flos": 17779411701120.0, "grad_norm": 2.1193876958071423, "language_loss": 0.7871294, "learning_rate": 1.448033976377354e-07, "loss": 0.806036, "num_input_tokens_seen": 158692185, "step": 7333, "time_per_iteration": 2.546302318572998 }, { "auxiliary_loss_clip": 0.01155468, "auxiliary_loss_mlp": 0.0102446, "balance_loss_clip": 1.04479671, "balance_loss_mlp": 1.01801682, "epoch": 0.8818613599470931, "flos": 18551815439040.0, "grad_norm": 2.0437921612213263, "language_loss": 0.74216104, "learning_rate": 1.445125307843713e-07, "loss": 0.76396036, "num_input_tokens_seen": 158710410, "step": 7334, "time_per_iteration": 2.4721553325653076 }, { "auxiliary_loss_clip": 0.01150344, "auxiliary_loss_mlp": 0.01028104, "balance_loss_clip": 1.04643393, "balance_loss_mlp": 1.02155614, "epoch": 0.8819816028377322, "flos": 27599796543360.0, "grad_norm": 1.729089359918046, "language_loss": 0.75712132, "learning_rate": 1.442219454069813e-07, "loss": 0.77890587, "num_input_tokens_seen": 158731435, "step": 7335, "time_per_iteration": 3.3427324295043945 }, { "auxiliary_loss_clip": 0.01114632, "auxiliary_loss_mlp": 0.01026644, "balance_loss_clip": 1.0420599, "balance_loss_mlp": 1.019629, "epoch": 0.8821018457283714, "flos": 23404604182080.0, "grad_norm": 1.8522147908457167, "language_loss": 0.66489273, "learning_rate": 1.4393164154964676e-07, "loss": 0.68630552, "num_input_tokens_seen": 158750965, "step": 7336, "time_per_iteration": 2.5957558155059814 }, { "auxiliary_loss_clip": 0.0115291, "auxiliary_loss_mlp": 0.0102709, "balance_loss_clip": 1.04825544, "balance_loss_mlp": 1.02006602, "epoch": 0.8822220886190104, "flos": 29132464086720.0, "grad_norm": 1.819242031920203, "language_loss": 0.93802047, "learning_rate": 1.4364161925640649e-07, "loss": 0.95982051, "num_input_tokens_seen": 158772365, "step": 7337, "time_per_iteration": 2.579063892364502 }, { "auxiliary_loss_clip": 0.01167477, "auxiliary_loss_mlp": 0.01023729, "balance_loss_clip": 1.04752481, "balance_loss_mlp": 1.01753902, "epoch": 0.8823423315096495, "flos": 20485425569280.0, "grad_norm": 1.746698540679855, "language_loss": 0.85079741, "learning_rate": 1.4335187857125663e-07, "loss": 0.87270951, "num_input_tokens_seen": 158791065, "step": 7338, "time_per_iteration": 2.460958242416382 }, { "auxiliary_loss_clip": 0.01155427, "auxiliary_loss_mlp": 0.01026984, "balance_loss_clip": 1.04615772, "balance_loss_mlp": 1.01986432, "epoch": 0.8824625744002886, "flos": 24206381604960.0, "grad_norm": 1.6893726279596444, "language_loss": 0.75638008, "learning_rate": 1.4306241953815023e-07, "loss": 0.77820426, "num_input_tokens_seen": 158812125, "step": 7339, "time_per_iteration": 2.589061737060547 }, { "auxiliary_loss_clip": 0.01153058, "auxiliary_loss_mlp": 0.01023245, "balance_loss_clip": 1.04535413, "balance_loss_mlp": 1.01665568, "epoch": 0.8825828172909277, "flos": 24679504504800.0, "grad_norm": 1.692411602131599, "language_loss": 0.71015513, "learning_rate": 1.4277324220099862e-07, "loss": 0.73191822, "num_input_tokens_seen": 158834035, "step": 7340, "time_per_iteration": 2.5349555015563965 }, { "auxiliary_loss_clip": 0.01120066, "auxiliary_loss_mlp": 0.01028502, "balance_loss_clip": 1.04089081, "balance_loss_mlp": 1.02176118, "epoch": 0.8827030601815667, "flos": 22456167447840.0, "grad_norm": 1.8224661957014872, "language_loss": 0.74241078, "learning_rate": 1.4248434660366938e-07, "loss": 0.76389647, "num_input_tokens_seen": 158853510, "step": 7341, "time_per_iteration": 3.5043351650238037 }, { "auxiliary_loss_clip": 0.01137145, "auxiliary_loss_mlp": 0.01023584, "balance_loss_clip": 1.04443622, "balance_loss_mlp": 1.01692319, "epoch": 0.8828233030722058, "flos": 19865643358080.0, "grad_norm": 1.863943388728115, "language_loss": 0.70525473, "learning_rate": 1.4219573278998808e-07, "loss": 0.72686207, "num_input_tokens_seen": 158871970, "step": 7342, "time_per_iteration": 3.21268630027771 }, { "auxiliary_loss_clip": 0.01136786, "auxiliary_loss_mlp": 0.01028917, "balance_loss_clip": 1.042871, "balance_loss_mlp": 1.02155316, "epoch": 0.882943545962845, "flos": 39347218712160.0, "grad_norm": 1.9326431641057191, "language_loss": 0.64377797, "learning_rate": 1.4190740080373685e-07, "loss": 0.66543496, "num_input_tokens_seen": 158892250, "step": 7343, "time_per_iteration": 2.65563702583313 }, { "auxiliary_loss_clip": 0.01110331, "auxiliary_loss_mlp": 0.01028884, "balance_loss_clip": 1.04293227, "balance_loss_mlp": 1.02182102, "epoch": 0.883063788853484, "flos": 19054527525600.0, "grad_norm": 1.8451766342820688, "language_loss": 0.84148932, "learning_rate": 1.4161935068865538e-07, "loss": 0.86288142, "num_input_tokens_seen": 158907395, "step": 7344, "time_per_iteration": 2.552506923675537 }, { "auxiliary_loss_clip": 0.01165166, "auxiliary_loss_mlp": 0.01027226, "balance_loss_clip": 1.04573774, "balance_loss_mlp": 1.02011514, "epoch": 0.8831840317441231, "flos": 18733200062400.0, "grad_norm": 1.903058835949043, "language_loss": 0.75852299, "learning_rate": 1.4133158248844113e-07, "loss": 0.78044689, "num_input_tokens_seen": 158926300, "step": 7345, "time_per_iteration": 2.430797576904297 }, { "auxiliary_loss_clip": 0.01125371, "auxiliary_loss_mlp": 0.01022828, "balance_loss_clip": 1.0444994, "balance_loss_mlp": 1.01546049, "epoch": 0.8833042746347622, "flos": 26827716058080.0, "grad_norm": 1.9124446430055129, "language_loss": 0.73627639, "learning_rate": 1.4104409624674785e-07, "loss": 0.75775838, "num_input_tokens_seen": 158946085, "step": 7346, "time_per_iteration": 2.579702138900757 }, { "auxiliary_loss_clip": 0.01157347, "auxiliary_loss_mlp": 0.01029002, "balance_loss_clip": 1.0503695, "balance_loss_mlp": 1.02220988, "epoch": 0.8834245175254013, "flos": 26104081163520.0, "grad_norm": 4.522542878299821, "language_loss": 0.78414285, "learning_rate": 1.407568920071873e-07, "loss": 0.80600637, "num_input_tokens_seen": 158964950, "step": 7347, "time_per_iteration": 2.520136594772339 }, { "auxiliary_loss_clip": 0.01173313, "auxiliary_loss_mlp": 0.01027606, "balance_loss_clip": 1.04954171, "balance_loss_mlp": 1.02024472, "epoch": 0.8835447604160404, "flos": 30629041473600.0, "grad_norm": 2.210451584217563, "language_loss": 0.67981416, "learning_rate": 1.4046996981332782e-07, "loss": 0.70182335, "num_input_tokens_seen": 158984835, "step": 7348, "time_per_iteration": 2.5276596546173096 }, { "auxiliary_loss_clip": 0.01123038, "auxiliary_loss_mlp": 0.01030049, "balance_loss_clip": 1.04173684, "balance_loss_mlp": 1.02291441, "epoch": 0.8836650033066795, "flos": 24718360267200.0, "grad_norm": 1.974052571809991, "language_loss": 0.78135824, "learning_rate": 1.4018332970869516e-07, "loss": 0.80288911, "num_input_tokens_seen": 159002775, "step": 7349, "time_per_iteration": 2.558115243911743 }, { "auxiliary_loss_clip": 0.01134352, "auxiliary_loss_mlp": 0.01022285, "balance_loss_clip": 1.04586148, "balance_loss_mlp": 1.01551437, "epoch": 0.8837852461973186, "flos": 25413375732960.0, "grad_norm": 1.8271629322442624, "language_loss": 0.84878242, "learning_rate": 1.398969717367733e-07, "loss": 0.87034881, "num_input_tokens_seen": 159024100, "step": 7350, "time_per_iteration": 2.577359437942505 }, { "auxiliary_loss_clip": 0.01107767, "auxiliary_loss_mlp": 0.01024207, "balance_loss_clip": 1.04592514, "balance_loss_mlp": 1.01806164, "epoch": 0.8839054890879576, "flos": 17822577498720.0, "grad_norm": 2.0943898055131025, "language_loss": 0.76021957, "learning_rate": 1.396108959410014e-07, "loss": 0.78153926, "num_input_tokens_seen": 159043315, "step": 7351, "time_per_iteration": 2.578176736831665 }, { "auxiliary_loss_clip": 0.01154371, "auxiliary_loss_mlp": 0.00761923, "balance_loss_clip": 1.04808354, "balance_loss_mlp": 1.0005765, "epoch": 0.8840257319785968, "flos": 23769025608960.0, "grad_norm": 1.5333348839776821, "language_loss": 0.81556356, "learning_rate": 1.3932510236477745e-07, "loss": 0.83472657, "num_input_tokens_seen": 159063985, "step": 7352, "time_per_iteration": 2.5270509719848633 }, { "auxiliary_loss_clip": 0.01153945, "auxiliary_loss_mlp": 0.01025928, "balance_loss_clip": 1.04527593, "balance_loss_mlp": 1.01860213, "epoch": 0.8841459748692359, "flos": 29059780936320.0, "grad_norm": 1.8176397272139, "language_loss": 0.56159103, "learning_rate": 1.3903959105145636e-07, "loss": 0.58338976, "num_input_tokens_seen": 159084475, "step": 7353, "time_per_iteration": 2.573115348815918 }, { "auxiliary_loss_clip": 0.01167301, "auxiliary_loss_mlp": 0.01024604, "balance_loss_clip": 1.04792416, "balance_loss_mlp": 1.01741886, "epoch": 0.8842662177598749, "flos": 24311527298880.0, "grad_norm": 1.8617688195732467, "language_loss": 0.83002245, "learning_rate": 1.387543620443492e-07, "loss": 0.85194153, "num_input_tokens_seen": 159101320, "step": 7354, "time_per_iteration": 3.217106580734253 }, { "auxiliary_loss_clip": 0.0116812, "auxiliary_loss_mlp": 0.01026507, "balance_loss_clip": 1.04809737, "balance_loss_mlp": 1.01997185, "epoch": 0.8843864606505141, "flos": 25007871692160.0, "grad_norm": 1.629802786122829, "language_loss": 0.84291041, "learning_rate": 1.3846941538672606e-07, "loss": 0.86485672, "num_input_tokens_seen": 159120025, "step": 7355, "time_per_iteration": 2.541809558868408 }, { "auxiliary_loss_clip": 0.01116556, "auxiliary_loss_mlp": 0.0102518, "balance_loss_clip": 1.04405046, "balance_loss_mlp": 1.01827514, "epoch": 0.8845067035411531, "flos": 28183919601600.0, "grad_norm": 2.358693345081662, "language_loss": 0.8046661, "learning_rate": 1.3818475112181193e-07, "loss": 0.82608354, "num_input_tokens_seen": 159138820, "step": 7356, "time_per_iteration": 2.630537986755371 }, { "auxiliary_loss_clip": 0.01135962, "auxiliary_loss_mlp": 0.01023772, "balance_loss_clip": 1.04576135, "balance_loss_mlp": 1.01735842, "epoch": 0.8846269464317922, "flos": 12853221468480.0, "grad_norm": 2.0741531358087277, "language_loss": 0.79854178, "learning_rate": 1.3790036929279091e-07, "loss": 0.82013917, "num_input_tokens_seen": 159155975, "step": 7357, "time_per_iteration": 2.512056350708008 }, { "auxiliary_loss_clip": 0.01157107, "auxiliary_loss_mlp": 0.00761973, "balance_loss_clip": 1.0480175, "balance_loss_mlp": 1.0005244, "epoch": 0.8847471893224313, "flos": 18624354921600.0, "grad_norm": 2.524108315024888, "language_loss": 0.58711678, "learning_rate": 1.3761626994280363e-07, "loss": 0.60630751, "num_input_tokens_seen": 159173445, "step": 7358, "time_per_iteration": 2.45668888092041 }, { "auxiliary_loss_clip": 0.01131898, "auxiliary_loss_mlp": 0.01024729, "balance_loss_clip": 1.04350889, "balance_loss_mlp": 1.01826143, "epoch": 0.8848674322130704, "flos": 35769438042720.0, "grad_norm": 1.7630830967677686, "language_loss": 0.73445916, "learning_rate": 1.3733245311494735e-07, "loss": 0.75602543, "num_input_tokens_seen": 159196100, "step": 7359, "time_per_iteration": 2.701580286026001 }, { "auxiliary_loss_clip": 0.01156253, "auxiliary_loss_mlp": 0.01023497, "balance_loss_clip": 1.04836786, "balance_loss_mlp": 1.01638627, "epoch": 0.8849876751037095, "flos": 24243764772000.0, "grad_norm": 2.0071228772443104, "language_loss": 0.70539141, "learning_rate": 1.3704891885227676e-07, "loss": 0.72718894, "num_input_tokens_seen": 159216145, "step": 7360, "time_per_iteration": 2.5189402103424072 }, { "auxiliary_loss_clip": 0.01122821, "auxiliary_loss_mlp": 0.01025259, "balance_loss_clip": 1.03921115, "balance_loss_mlp": 1.01768279, "epoch": 0.8851079179943486, "flos": 21500583238560.0, "grad_norm": 1.9859178231466097, "language_loss": 0.77952504, "learning_rate": 1.367656671978037e-07, "loss": 0.80100584, "num_input_tokens_seen": 159233610, "step": 7361, "time_per_iteration": 2.5880837440490723 }, { "auxiliary_loss_clip": 0.01144802, "auxiliary_loss_mlp": 0.01026346, "balance_loss_clip": 1.0435189, "balance_loss_mlp": 1.01945877, "epoch": 0.8852281608849877, "flos": 15300713859840.0, "grad_norm": 1.8478376274299197, "language_loss": 0.73699331, "learning_rate": 1.36482698194498e-07, "loss": 0.75870478, "num_input_tokens_seen": 159250155, "step": 7362, "time_per_iteration": 3.2338359355926514 }, { "auxiliary_loss_clip": 0.01139871, "auxiliary_loss_mlp": 0.01026502, "balance_loss_clip": 1.04282165, "balance_loss_mlp": 1.01945639, "epoch": 0.8853484037756267, "flos": 23295723124320.0, "grad_norm": 1.9769896333366044, "language_loss": 0.7197181, "learning_rate": 1.3620001188528506e-07, "loss": 0.74138188, "num_input_tokens_seen": 159270875, "step": 7363, "time_per_iteration": 2.536457061767578 }, { "auxiliary_loss_clip": 0.01156667, "auxiliary_loss_mlp": 0.01031472, "balance_loss_clip": 1.04482925, "balance_loss_mlp": 1.02453148, "epoch": 0.8854686466662659, "flos": 25114777317120.0, "grad_norm": 2.7704994768298907, "language_loss": 0.73780072, "learning_rate": 1.3591760831304865e-07, "loss": 0.75968212, "num_input_tokens_seen": 159288565, "step": 7364, "time_per_iteration": 2.5181376934051514 }, { "auxiliary_loss_clip": 0.01167632, "auxiliary_loss_mlp": 0.01024854, "balance_loss_clip": 1.04734969, "balance_loss_mlp": 1.01783252, "epoch": 0.885588889556905, "flos": 21390876090720.0, "grad_norm": 1.6475796617089764, "language_loss": 0.79795337, "learning_rate": 1.356354875206287e-07, "loss": 0.81987822, "num_input_tokens_seen": 159306400, "step": 7365, "time_per_iteration": 2.448371171951294 }, { "auxiliary_loss_clip": 0.01127594, "auxiliary_loss_mlp": 0.01026647, "balance_loss_clip": 1.04729915, "balance_loss_mlp": 1.01993573, "epoch": 0.885709132447544, "flos": 26906756510400.0, "grad_norm": 2.143606443974315, "language_loss": 0.70085597, "learning_rate": 1.3535364955082296e-07, "loss": 0.7223984, "num_input_tokens_seen": 159326250, "step": 7366, "time_per_iteration": 2.613101005554199 }, { "auxiliary_loss_clip": 0.01166673, "auxiliary_loss_mlp": 0.01025663, "balance_loss_clip": 1.04864895, "balance_loss_mlp": 1.019104, "epoch": 0.8858293753381832, "flos": 26103398741280.0, "grad_norm": 1.8846249029079616, "language_loss": 0.64606255, "learning_rate": 1.3507209444638613e-07, "loss": 0.66798592, "num_input_tokens_seen": 159348250, "step": 7367, "time_per_iteration": 3.2659950256347656 }, { "auxiliary_loss_clip": 0.01154932, "auxiliary_loss_mlp": 0.01026404, "balance_loss_clip": 1.04751325, "balance_loss_mlp": 1.01929021, "epoch": 0.8859496182288222, "flos": 23292813850560.0, "grad_norm": 1.9590389043541114, "language_loss": 0.74301237, "learning_rate": 1.347908222500298e-07, "loss": 0.7648257, "num_input_tokens_seen": 159368325, "step": 7368, "time_per_iteration": 3.253079891204834 }, { "auxiliary_loss_clip": 0.01114084, "auxiliary_loss_mlp": 0.01021995, "balance_loss_clip": 1.04409838, "balance_loss_mlp": 1.01532531, "epoch": 0.8860698611194613, "flos": 16872919587840.0, "grad_norm": 1.9331129450208422, "language_loss": 0.69521433, "learning_rate": 1.3450983300442276e-07, "loss": 0.71657515, "num_input_tokens_seen": 159387555, "step": 7369, "time_per_iteration": 2.528491258621216 }, { "auxiliary_loss_clip": 0.0115566, "auxiliary_loss_mlp": 0.01022566, "balance_loss_clip": 1.04669118, "balance_loss_mlp": 1.01628923, "epoch": 0.8861901040101005, "flos": 24681408103680.0, "grad_norm": 1.838624387604199, "language_loss": 0.73613226, "learning_rate": 1.3422912675219068e-07, "loss": 0.75791454, "num_input_tokens_seen": 159407310, "step": 7370, "time_per_iteration": 2.536336898803711 }, { "auxiliary_loss_clip": 0.01164678, "auxiliary_loss_mlp": 0.01023301, "balance_loss_clip": 1.04834962, "balance_loss_mlp": 1.01681292, "epoch": 0.8863103469007395, "flos": 24423030294720.0, "grad_norm": 1.4987169316909603, "language_loss": 0.78896391, "learning_rate": 1.339487035359166e-07, "loss": 0.81084365, "num_input_tokens_seen": 159427680, "step": 7371, "time_per_iteration": 2.4845731258392334 }, { "auxiliary_loss_clip": 0.0114131, "auxiliary_loss_mlp": 0.00761019, "balance_loss_clip": 1.04690123, "balance_loss_mlp": 1.00056934, "epoch": 0.8864305897913786, "flos": 22053967767360.0, "grad_norm": 1.558803389236336, "language_loss": 0.84558547, "learning_rate": 1.336685633981409e-07, "loss": 0.86460871, "num_input_tokens_seen": 159448765, "step": 7372, "time_per_iteration": 2.5527756214141846 }, { "auxiliary_loss_clip": 0.01153342, "auxiliary_loss_mlp": 0.01026568, "balance_loss_clip": 1.04354858, "balance_loss_mlp": 1.01929331, "epoch": 0.8865508326820177, "flos": 19099453254240.0, "grad_norm": 1.808683984219512, "language_loss": 0.74511522, "learning_rate": 1.333887063813597e-07, "loss": 0.76691431, "num_input_tokens_seen": 159466870, "step": 7373, "time_per_iteration": 2.493108034133911 }, { "auxiliary_loss_clip": 0.01140497, "auxiliary_loss_mlp": 0.01023983, "balance_loss_clip": 1.04243684, "balance_loss_mlp": 1.01769459, "epoch": 0.8866710755726568, "flos": 15414192288480.0, "grad_norm": 1.800628698103216, "language_loss": 0.66116589, "learning_rate": 1.331091325280278e-07, "loss": 0.68281072, "num_input_tokens_seen": 159485840, "step": 7374, "time_per_iteration": 2.520509958267212 }, { "auxiliary_loss_clip": 0.01105322, "auxiliary_loss_mlp": 0.01021912, "balance_loss_clip": 1.04143596, "balance_loss_mlp": 1.01483107, "epoch": 0.8867913184632958, "flos": 20083692809280.0, "grad_norm": 2.6708951847549036, "language_loss": 0.78700757, "learning_rate": 1.3282984188055625e-07, "loss": 0.80827993, "num_input_tokens_seen": 159505630, "step": 7375, "time_per_iteration": 2.602083206176758 }, { "auxiliary_loss_clip": 0.0116643, "auxiliary_loss_mlp": 0.0102196, "balance_loss_clip": 1.04668438, "balance_loss_mlp": 1.01541519, "epoch": 0.8869115613539349, "flos": 23365856170560.0, "grad_norm": 1.755487687762549, "language_loss": 0.79304194, "learning_rate": 1.3255083448131288e-07, "loss": 0.81492579, "num_input_tokens_seen": 159524675, "step": 7376, "time_per_iteration": 2.4965269565582275 }, { "auxiliary_loss_clip": 0.01156675, "auxiliary_loss_mlp": 0.01021205, "balance_loss_clip": 1.04504216, "balance_loss_mlp": 1.01469648, "epoch": 0.8870318042445741, "flos": 21286843822560.0, "grad_norm": 2.289519332540613, "language_loss": 0.79087466, "learning_rate": 1.3227211037262365e-07, "loss": 0.81265348, "num_input_tokens_seen": 159541915, "step": 7377, "time_per_iteration": 2.465386152267456 }, { "auxiliary_loss_clip": 0.01110226, "auxiliary_loss_mlp": 0.01025545, "balance_loss_clip": 1.04075313, "balance_loss_mlp": 1.01807976, "epoch": 0.8871520471352131, "flos": 20010865991040.0, "grad_norm": 2.1118638712948496, "language_loss": 0.85198313, "learning_rate": 1.319936695967696e-07, "loss": 0.87334085, "num_input_tokens_seen": 159559740, "step": 7378, "time_per_iteration": 2.5729737281799316 }, { "auxiliary_loss_clip": 0.01173631, "auxiliary_loss_mlp": 0.01028914, "balance_loss_clip": 1.04817462, "balance_loss_mlp": 1.02109993, "epoch": 0.8872722900258522, "flos": 22601425997760.0, "grad_norm": 3.162403249330369, "language_loss": 0.82317483, "learning_rate": 1.3171551219599097e-07, "loss": 0.8452003, "num_input_tokens_seen": 159578265, "step": 7379, "time_per_iteration": 2.4520890712738037 }, { "auxiliary_loss_clip": 0.01170606, "auxiliary_loss_mlp": 0.01032708, "balance_loss_clip": 1.0507288, "balance_loss_mlp": 1.02543664, "epoch": 0.8873925329164913, "flos": 22163279828640.0, "grad_norm": 2.157808123624274, "language_loss": 0.77802706, "learning_rate": 1.3143763821248377e-07, "loss": 0.80006027, "num_input_tokens_seen": 159595350, "step": 7380, "time_per_iteration": 3.1456077098846436 }, { "auxiliary_loss_clip": 0.01167155, "auxiliary_loss_mlp": 0.01019949, "balance_loss_clip": 1.04723918, "balance_loss_mlp": 1.0137887, "epoch": 0.8875127758071304, "flos": 19208226561120.0, "grad_norm": 1.9537267873681081, "language_loss": 0.72112978, "learning_rate": 1.3116004768840118e-07, "loss": 0.74300075, "num_input_tokens_seen": 159613725, "step": 7381, "time_per_iteration": 2.4319162368774414 }, { "auxiliary_loss_clip": 0.01170806, "auxiliary_loss_mlp": 0.01024162, "balance_loss_clip": 1.04882383, "balance_loss_mlp": 1.01751065, "epoch": 0.8876330186977694, "flos": 18110903664000.0, "grad_norm": 1.7689019669604955, "language_loss": 0.74249989, "learning_rate": 1.3088274066585348e-07, "loss": 0.7644496, "num_input_tokens_seen": 159631335, "step": 7382, "time_per_iteration": 2.452903985977173 }, { "auxiliary_loss_clip": 0.01131326, "auxiliary_loss_mlp": 0.01022385, "balance_loss_clip": 1.04258311, "balance_loss_mlp": 1.01575387, "epoch": 0.8877532615884086, "flos": 22009437125280.0, "grad_norm": 2.0864314760176743, "language_loss": 0.90619034, "learning_rate": 1.3060571718690749e-07, "loss": 0.92772746, "num_input_tokens_seen": 159648830, "step": 7383, "time_per_iteration": 2.5538933277130127 }, { "auxiliary_loss_clip": 0.01033304, "auxiliary_loss_mlp": 0.00752764, "balance_loss_clip": 1.00779748, "balance_loss_mlp": 1.00019789, "epoch": 0.8878735044790477, "flos": 72136934924160.0, "grad_norm": 0.739915126902207, "language_loss": 0.56932402, "learning_rate": 1.3032897729358805e-07, "loss": 0.58718473, "num_input_tokens_seen": 159709785, "step": 7384, "time_per_iteration": 3.1617655754089355 }, { "auxiliary_loss_clip": 0.01086062, "auxiliary_loss_mlp": 0.00762423, "balance_loss_clip": 1.03745651, "balance_loss_mlp": 1.00056314, "epoch": 0.8879937473696867, "flos": 27526359136800.0, "grad_norm": 1.9715317609251772, "language_loss": 0.79694724, "learning_rate": 1.3005252102787645e-07, "loss": 0.81543207, "num_input_tokens_seen": 159728725, "step": 7385, "time_per_iteration": 2.694617748260498 }, { "auxiliary_loss_clip": 0.01157467, "auxiliary_loss_mlp": 0.01028818, "balance_loss_clip": 1.0461154, "balance_loss_mlp": 1.02187395, "epoch": 0.8881139902603259, "flos": 22234095297120.0, "grad_norm": 1.6798961666132421, "language_loss": 0.73438126, "learning_rate": 1.297763484317105e-07, "loss": 0.75624412, "num_input_tokens_seen": 159747020, "step": 7386, "time_per_iteration": 2.4897470474243164 }, { "auxiliary_loss_clip": 0.01102846, "auxiliary_loss_mlp": 0.00762196, "balance_loss_clip": 1.0387373, "balance_loss_mlp": 1.00056219, "epoch": 0.888234233150965, "flos": 20299551325920.0, "grad_norm": 1.9662794116939892, "language_loss": 0.70276082, "learning_rate": 1.2950045954698551e-07, "loss": 0.72141123, "num_input_tokens_seen": 159764855, "step": 7387, "time_per_iteration": 3.330000400543213 }, { "auxiliary_loss_clip": 0.01115754, "auxiliary_loss_mlp": 0.01020974, "balance_loss_clip": 1.04152358, "balance_loss_mlp": 1.01411319, "epoch": 0.888354476041604, "flos": 18147999495360.0, "grad_norm": 1.5090905868476172, "language_loss": 0.75196338, "learning_rate": 1.2922485441555343e-07, "loss": 0.77333075, "num_input_tokens_seen": 159783935, "step": 7388, "time_per_iteration": 2.560567855834961 }, { "auxiliary_loss_clip": 0.01165084, "auxiliary_loss_mlp": 0.0102591, "balance_loss_clip": 1.04483461, "balance_loss_mlp": 1.01902819, "epoch": 0.8884747189322432, "flos": 22014285914880.0, "grad_norm": 1.705808968738217, "language_loss": 0.81613779, "learning_rate": 1.2894953307922363e-07, "loss": 0.83804768, "num_input_tokens_seen": 159802895, "step": 7389, "time_per_iteration": 2.483058214187622 }, { "auxiliary_loss_clip": 0.01120448, "auxiliary_loss_mlp": 0.01028864, "balance_loss_clip": 1.04297018, "balance_loss_mlp": 1.02143407, "epoch": 0.8885949618228822, "flos": 19786782490560.0, "grad_norm": 1.9243704170545732, "language_loss": 0.8398701, "learning_rate": 1.2867449557976208e-07, "loss": 0.86136323, "num_input_tokens_seen": 159820995, "step": 7390, "time_per_iteration": 2.527362823486328 }, { "auxiliary_loss_clip": 0.0115251, "auxiliary_loss_mlp": 0.01025443, "balance_loss_clip": 1.04710484, "balance_loss_mlp": 1.019274, "epoch": 0.8887152047135213, "flos": 20047602652800.0, "grad_norm": 1.8663169519159761, "language_loss": 0.75864792, "learning_rate": 1.283997419588916e-07, "loss": 0.7804274, "num_input_tokens_seen": 159840465, "step": 7391, "time_per_iteration": 2.482179641723633 }, { "auxiliary_loss_clip": 0.01158736, "auxiliary_loss_mlp": 0.01023377, "balance_loss_clip": 1.04705286, "balance_loss_mlp": 1.0167613, "epoch": 0.8888354476041604, "flos": 18588121097280.0, "grad_norm": 1.9025731890517925, "language_loss": 0.62166405, "learning_rate": 1.2812527225829216e-07, "loss": 0.64348513, "num_input_tokens_seen": 159858690, "step": 7392, "time_per_iteration": 2.4664902687072754 }, { "auxiliary_loss_clip": 0.01160034, "auxiliary_loss_mlp": 0.01025708, "balance_loss_clip": 1.04729021, "balance_loss_mlp": 1.01829886, "epoch": 0.8889556904947995, "flos": 21689797759200.0, "grad_norm": 1.878055602232035, "language_loss": 0.76609862, "learning_rate": 1.2785108651960052e-07, "loss": 0.78795606, "num_input_tokens_seen": 159880325, "step": 7393, "time_per_iteration": 3.2903921604156494 }, { "auxiliary_loss_clip": 0.01156348, "auxiliary_loss_mlp": 0.01024359, "balance_loss_clip": 1.04456913, "balance_loss_mlp": 1.01746297, "epoch": 0.8890759333854386, "flos": 27381208337760.0, "grad_norm": 1.9351347877906677, "language_loss": 0.80371225, "learning_rate": 1.2757718478441094e-07, "loss": 0.82551932, "num_input_tokens_seen": 159901070, "step": 7394, "time_per_iteration": 3.30767822265625 }, { "auxiliary_loss_clip": 0.01136203, "auxiliary_loss_mlp": 0.01024333, "balance_loss_clip": 1.0425148, "balance_loss_mlp": 1.01744831, "epoch": 0.8891961762760777, "flos": 24498838220640.0, "grad_norm": 1.7446948908016793, "language_loss": 0.77569222, "learning_rate": 1.2730356709427302e-07, "loss": 0.7972976, "num_input_tokens_seen": 159919750, "step": 7395, "time_per_iteration": 2.5471866130828857 }, { "auxiliary_loss_clip": 0.01152212, "auxiliary_loss_mlp": 0.01024109, "balance_loss_clip": 1.04849505, "balance_loss_mlp": 1.01733804, "epoch": 0.8893164191667168, "flos": 41499776217600.0, "grad_norm": 1.5075760591519645, "language_loss": 0.59757817, "learning_rate": 1.2703023349069542e-07, "loss": 0.61934137, "num_input_tokens_seen": 159944600, "step": 7396, "time_per_iteration": 2.662708282470703 }, { "auxiliary_loss_clip": 0.01148646, "auxiliary_loss_mlp": 0.01023989, "balance_loss_clip": 1.04488587, "balance_loss_mlp": 1.01713181, "epoch": 0.8894366620573558, "flos": 33583627820640.0, "grad_norm": 2.646045358246408, "language_loss": 0.6168524, "learning_rate": 1.2675718401514223e-07, "loss": 0.63857871, "num_input_tokens_seen": 159968780, "step": 7397, "time_per_iteration": 2.5722992420196533 }, { "auxiliary_loss_clip": 0.01139123, "auxiliary_loss_mlp": 0.01029726, "balance_loss_clip": 1.04491997, "balance_loss_mlp": 1.02226937, "epoch": 0.889556904947995, "flos": 16909835834400.0, "grad_norm": 2.2032294643525754, "language_loss": 0.74545074, "learning_rate": 1.264844187090346e-07, "loss": 0.76713926, "num_input_tokens_seen": 159985905, "step": 7398, "time_per_iteration": 2.512648820877075 }, { "auxiliary_loss_clip": 0.01133806, "auxiliary_loss_mlp": 0.01023418, "balance_loss_clip": 1.04274118, "balance_loss_mlp": 1.01714778, "epoch": 0.889677147838634, "flos": 26030859258720.0, "grad_norm": 1.6496274046228614, "language_loss": 0.7492969, "learning_rate": 1.262119376137516e-07, "loss": 0.77086914, "num_input_tokens_seen": 160006965, "step": 7399, "time_per_iteration": 2.5460238456726074 }, { "auxiliary_loss_clip": 0.0114386, "auxiliary_loss_mlp": 0.01022443, "balance_loss_clip": 1.0438447, "balance_loss_mlp": 1.01586592, "epoch": 0.8897973907292731, "flos": 26468287088640.0, "grad_norm": 1.5662107914823284, "language_loss": 0.85010946, "learning_rate": 1.2593974077062707e-07, "loss": 0.87177253, "num_input_tokens_seen": 160028585, "step": 7400, "time_per_iteration": 2.5435547828674316 }, { "auxiliary_loss_clip": 0.01116171, "auxiliary_loss_mlp": 0.0102221, "balance_loss_clip": 1.04235792, "balance_loss_mlp": 1.01537609, "epoch": 0.8899176336199123, "flos": 26249698883040.0, "grad_norm": 2.2634111098274876, "language_loss": 0.63650239, "learning_rate": 1.2566782822095423e-07, "loss": 0.65788621, "num_input_tokens_seen": 160048840, "step": 7401, "time_per_iteration": 2.577956199645996 }, { "auxiliary_loss_clip": 0.01130664, "auxiliary_loss_mlp": 0.01023518, "balance_loss_clip": 1.04577017, "balance_loss_mlp": 1.01680374, "epoch": 0.8900378765105513, "flos": 20811745489920.0, "grad_norm": 1.7266862475620472, "language_loss": 0.71156138, "learning_rate": 1.2539620000598162e-07, "loss": 0.73310316, "num_input_tokens_seen": 160068175, "step": 7402, "time_per_iteration": 2.552755117416382 }, { "auxiliary_loss_clip": 0.01167971, "auxiliary_loss_mlp": 0.01024529, "balance_loss_clip": 1.04715466, "balance_loss_mlp": 1.01762414, "epoch": 0.8901581194011904, "flos": 16472336170560.0, "grad_norm": 2.1615037808241513, "language_loss": 0.7942816, "learning_rate": 1.2512485616691492e-07, "loss": 0.81620657, "num_input_tokens_seen": 160085230, "step": 7403, "time_per_iteration": 2.4207208156585693 }, { "auxiliary_loss_clip": 0.01129533, "auxiliary_loss_mlp": 0.01026595, "balance_loss_clip": 1.04219484, "balance_loss_mlp": 1.01908481, "epoch": 0.8902783622918296, "flos": 35155258877280.0, "grad_norm": 1.5450693510574003, "language_loss": 0.80943346, "learning_rate": 1.2485379674491681e-07, "loss": 0.83099478, "num_input_tokens_seen": 160111425, "step": 7404, "time_per_iteration": 2.6825811862945557 }, { "auxiliary_loss_clip": 0.01141205, "auxiliary_loss_mlp": 0.01031437, "balance_loss_clip": 1.04732132, "balance_loss_mlp": 1.02417064, "epoch": 0.8903986051824686, "flos": 17201071273440.0, "grad_norm": 2.1621843827393845, "language_loss": 0.79245061, "learning_rate": 1.2458302178110657e-07, "loss": 0.81417704, "num_input_tokens_seen": 160129790, "step": 7405, "time_per_iteration": 2.472982883453369 }, { "auxiliary_loss_clip": 0.01116469, "auxiliary_loss_mlp": 0.01020656, "balance_loss_clip": 1.0423193, "balance_loss_mlp": 1.01435566, "epoch": 0.8905188480731077, "flos": 25483867948800.0, "grad_norm": 5.936251655309059, "language_loss": 0.82442987, "learning_rate": 1.2431253131656118e-07, "loss": 0.84580112, "num_input_tokens_seen": 160149265, "step": 7406, "time_per_iteration": 3.3268039226531982 }, { "auxiliary_loss_clip": 0.01132639, "auxiliary_loss_mlp": 0.0102464, "balance_loss_clip": 1.04400253, "balance_loss_mlp": 1.01802969, "epoch": 0.8906390909637467, "flos": 23365892087520.0, "grad_norm": 1.6503804364095447, "language_loss": 0.76714343, "learning_rate": 1.240423253923133e-07, "loss": 0.7887162, "num_input_tokens_seen": 160168870, "step": 7407, "time_per_iteration": 2.5093178749084473 }, { "auxiliary_loss_clip": 0.01154844, "auxiliary_loss_mlp": 0.01023091, "balance_loss_clip": 1.04565227, "balance_loss_mlp": 1.01563454, "epoch": 0.8907593338543859, "flos": 21068794371360.0, "grad_norm": 2.248107179714532, "language_loss": 0.69092315, "learning_rate": 1.237724040493533e-07, "loss": 0.71270251, "num_input_tokens_seen": 160187495, "step": 7408, "time_per_iteration": 2.486229419708252 }, { "auxiliary_loss_clip": 0.01176066, "auxiliary_loss_mlp": 0.01030847, "balance_loss_clip": 1.05180645, "balance_loss_mlp": 1.02297306, "epoch": 0.8908795767450249, "flos": 21869566119360.0, "grad_norm": 3.370533144375442, "language_loss": 0.72951066, "learning_rate": 1.2350276732862773e-07, "loss": 0.75157982, "num_input_tokens_seen": 160208520, "step": 7409, "time_per_iteration": 2.4750282764434814 }, { "auxiliary_loss_clip": 0.0105118, "auxiliary_loss_mlp": 0.01001115, "balance_loss_clip": 1.00588012, "balance_loss_mlp": 0.99981576, "epoch": 0.890999819635664, "flos": 66307880191200.0, "grad_norm": 0.8451126051690835, "language_loss": 0.56724548, "learning_rate": 1.2323341527103993e-07, "loss": 0.58776844, "num_input_tokens_seen": 160263720, "step": 7410, "time_per_iteration": 3.0172274112701416 }, { "auxiliary_loss_clip": 0.01167405, "auxiliary_loss_mlp": 0.01027024, "balance_loss_clip": 1.04750741, "balance_loss_mlp": 1.02041936, "epoch": 0.8911200625263032, "flos": 26869912097760.0, "grad_norm": 2.0273978602449008, "language_loss": 0.85099292, "learning_rate": 1.2296434791745135e-07, "loss": 0.8729372, "num_input_tokens_seen": 160282170, "step": 7411, "time_per_iteration": 2.4978997707366943 }, { "auxiliary_loss_clip": 0.01157554, "auxiliary_loss_mlp": 0.01020979, "balance_loss_clip": 1.04896522, "balance_loss_mlp": 1.01398444, "epoch": 0.8912403054169422, "flos": 20885829401760.0, "grad_norm": 1.7224832260949299, "language_loss": 0.7682755, "learning_rate": 1.2269556530867875e-07, "loss": 0.79006088, "num_input_tokens_seen": 160300725, "step": 7412, "time_per_iteration": 2.5358774662017822 }, { "auxiliary_loss_clip": 0.01173474, "auxiliary_loss_mlp": 0.01031362, "balance_loss_clip": 1.0486939, "balance_loss_mlp": 1.02368462, "epoch": 0.8913605483075813, "flos": 27016571409120.0, "grad_norm": 2.818008594608146, "language_loss": 0.82072675, "learning_rate": 1.2242706748549614e-07, "loss": 0.84277511, "num_input_tokens_seen": 160318720, "step": 7413, "time_per_iteration": 3.268984317779541 }, { "auxiliary_loss_clip": 0.01138753, "auxiliary_loss_mlp": 0.0102014, "balance_loss_clip": 1.04068828, "balance_loss_mlp": 1.01372039, "epoch": 0.8914807911982204, "flos": 23621504290560.0, "grad_norm": 1.7512816501818131, "language_loss": 0.81962585, "learning_rate": 1.2215885448863473e-07, "loss": 0.84121478, "num_input_tokens_seen": 160339595, "step": 7414, "time_per_iteration": 2.5632522106170654 }, { "auxiliary_loss_clip": 0.01139946, "auxiliary_loss_mlp": 0.01024072, "balance_loss_clip": 1.0462234, "balance_loss_mlp": 1.01750672, "epoch": 0.8916010340888595, "flos": 24462281143680.0, "grad_norm": 1.7167696647771187, "language_loss": 0.80410624, "learning_rate": 1.2189092635878152e-07, "loss": 0.82574642, "num_input_tokens_seen": 160361045, "step": 7415, "time_per_iteration": 2.5569686889648438 }, { "auxiliary_loss_clip": 0.01116584, "auxiliary_loss_mlp": 0.0102009, "balance_loss_clip": 1.0421586, "balance_loss_mlp": 1.01290715, "epoch": 0.8917212769794985, "flos": 21215776935360.0, "grad_norm": 1.62721441971652, "language_loss": 0.76890862, "learning_rate": 1.216232831365822e-07, "loss": 0.79027534, "num_input_tokens_seen": 160379990, "step": 7416, "time_per_iteration": 2.532235860824585 }, { "auxiliary_loss_clip": 0.0114789, "auxiliary_loss_mlp": 0.01027102, "balance_loss_clip": 1.04703975, "balance_loss_mlp": 1.0202682, "epoch": 0.8918415198701377, "flos": 25513995889920.0, "grad_norm": 1.9101411669652737, "language_loss": 0.80857056, "learning_rate": 1.2135592486263678e-07, "loss": 0.83032048, "num_input_tokens_seen": 160399240, "step": 7417, "time_per_iteration": 2.560516595840454 }, { "auxiliary_loss_clip": 0.0113963, "auxiliary_loss_mlp": 0.0102342, "balance_loss_clip": 1.04392481, "balance_loss_mlp": 1.01671708, "epoch": 0.8919617627607768, "flos": 37853011844640.0, "grad_norm": 1.523774156150636, "language_loss": 0.61078715, "learning_rate": 1.2108885157750415e-07, "loss": 0.63241762, "num_input_tokens_seen": 160421600, "step": 7418, "time_per_iteration": 2.6485888957977295 }, { "auxiliary_loss_clip": 0.01120641, "auxiliary_loss_mlp": 0.00762165, "balance_loss_clip": 1.04377913, "balance_loss_mlp": 1.00055707, "epoch": 0.8920820056514158, "flos": 26213680560480.0, "grad_norm": 1.6591773135118788, "language_loss": 0.80196977, "learning_rate": 1.2082206332169897e-07, "loss": 0.8207978, "num_input_tokens_seen": 160441695, "step": 7419, "time_per_iteration": 2.6122896671295166 }, { "auxiliary_loss_clip": 0.0113568, "auxiliary_loss_mlp": 0.01028127, "balance_loss_clip": 1.04558134, "balance_loss_mlp": 1.02179098, "epoch": 0.892202248542055, "flos": 17383138319040.0, "grad_norm": 2.792825172344517, "language_loss": 0.73132378, "learning_rate": 1.2055556013569225e-07, "loss": 0.75296181, "num_input_tokens_seen": 160457205, "step": 7420, "time_per_iteration": 3.9996094703674316 }, { "auxiliary_loss_clip": 0.01139574, "auxiliary_loss_mlp": 0.01022457, "balance_loss_clip": 1.04549158, "balance_loss_mlp": 1.01544785, "epoch": 0.892322491432694, "flos": 21324226989600.0, "grad_norm": 1.7238521912747429, "language_loss": 0.82192743, "learning_rate": 1.2028934205991315e-07, "loss": 0.84354776, "num_input_tokens_seen": 160476525, "step": 7421, "time_per_iteration": 2.5159313678741455 }, { "auxiliary_loss_clip": 0.01153677, "auxiliary_loss_mlp": 0.01022811, "balance_loss_clip": 1.04443645, "balance_loss_mlp": 1.0157268, "epoch": 0.8924427343233331, "flos": 24029378850720.0, "grad_norm": 1.5934247158416948, "language_loss": 0.76826137, "learning_rate": 1.2002340913474607e-07, "loss": 0.79002619, "num_input_tokens_seen": 160500160, "step": 7422, "time_per_iteration": 2.584671974182129 }, { "auxiliary_loss_clip": 0.01169333, "auxiliary_loss_mlp": 0.01026286, "balance_loss_clip": 1.04763675, "balance_loss_mlp": 1.01865649, "epoch": 0.8925629772139723, "flos": 30008074002720.0, "grad_norm": 2.6924995120982294, "language_loss": 0.73825735, "learning_rate": 1.1975776140053317e-07, "loss": 0.76021349, "num_input_tokens_seen": 160520130, "step": 7423, "time_per_iteration": 2.525522232055664 }, { "auxiliary_loss_clip": 0.01113904, "auxiliary_loss_mlp": 0.01025879, "balance_loss_clip": 1.04275513, "balance_loss_mlp": 1.01831532, "epoch": 0.8926832201046113, "flos": 22601713333440.0, "grad_norm": 1.8555233644933242, "language_loss": 0.7344569, "learning_rate": 1.194923988975729e-07, "loss": 0.75585473, "num_input_tokens_seen": 160539730, "step": 7424, "time_per_iteration": 2.5996901988983154 }, { "auxiliary_loss_clip": 0.01122293, "auxiliary_loss_mlp": 0.010243, "balance_loss_clip": 1.0428766, "balance_loss_mlp": 1.01731145, "epoch": 0.8928034629952504, "flos": 13297724939520.0, "grad_norm": 2.1881728724407026, "language_loss": 0.73769188, "learning_rate": 1.192273216661206e-07, "loss": 0.75915784, "num_input_tokens_seen": 160557820, "step": 7425, "time_per_iteration": 2.49861216545105 }, { "auxiliary_loss_clip": 0.01010133, "auxiliary_loss_mlp": 0.01004173, "balance_loss_clip": 1.00696874, "balance_loss_mlp": 1.00274837, "epoch": 0.8929237058858895, "flos": 54854566818240.0, "grad_norm": 0.7699498100550787, "language_loss": 0.57544005, "learning_rate": 1.189625297463881e-07, "loss": 0.59558308, "num_input_tokens_seen": 160619510, "step": 7426, "time_per_iteration": 3.2659196853637695 }, { "auxiliary_loss_clip": 0.01091368, "auxiliary_loss_mlp": 0.01023815, "balance_loss_clip": 1.03682697, "balance_loss_mlp": 1.01751208, "epoch": 0.8930439487765286, "flos": 28883855690880.0, "grad_norm": 1.592139870195339, "language_loss": 0.79691321, "learning_rate": 1.1869802317854394e-07, "loss": 0.81806505, "num_input_tokens_seen": 160643295, "step": 7427, "time_per_iteration": 2.8537235260009766 }, { "auxiliary_loss_clip": 0.01114964, "auxiliary_loss_mlp": 0.01026779, "balance_loss_clip": 1.04258418, "balance_loss_mlp": 1.01942968, "epoch": 0.8931641916671677, "flos": 22419287118240.0, "grad_norm": 1.967196241912493, "language_loss": 0.72155607, "learning_rate": 1.1843380200271425e-07, "loss": 0.74297345, "num_input_tokens_seen": 160662495, "step": 7428, "time_per_iteration": 2.587372303009033 }, { "auxiliary_loss_clip": 0.01117398, "auxiliary_loss_mlp": 0.01025106, "balance_loss_clip": 1.04381657, "balance_loss_mlp": 1.01813865, "epoch": 0.8932844345578068, "flos": 25843153250400.0, "grad_norm": 2.0412097519591117, "language_loss": 0.80158347, "learning_rate": 1.181698662589805e-07, "loss": 0.82300854, "num_input_tokens_seen": 160682080, "step": 7429, "time_per_iteration": 2.5836446285247803 }, { "auxiliary_loss_clip": 0.01152751, "auxiliary_loss_mlp": 0.01022078, "balance_loss_clip": 1.04544199, "balance_loss_mlp": 1.01516354, "epoch": 0.8934046774484459, "flos": 22925806402560.0, "grad_norm": 1.7022713345043043, "language_loss": 0.75719702, "learning_rate": 1.1790621598738249e-07, "loss": 0.77894533, "num_input_tokens_seen": 160700395, "step": 7430, "time_per_iteration": 2.521206855773926 }, { "auxiliary_loss_clip": 0.0116594, "auxiliary_loss_mlp": 0.0102537, "balance_loss_clip": 1.04878569, "balance_loss_mlp": 1.01915622, "epoch": 0.8935249203390849, "flos": 24462101558880.0, "grad_norm": 3.4498500523634488, "language_loss": 0.74654484, "learning_rate": 1.1764285122791461e-07, "loss": 0.76845789, "num_input_tokens_seen": 160721115, "step": 7431, "time_per_iteration": 2.4988834857940674 }, { "auxiliary_loss_clip": 0.01152919, "auxiliary_loss_mlp": 0.01024875, "balance_loss_clip": 1.04464018, "balance_loss_mlp": 1.01811028, "epoch": 0.8936451632297241, "flos": 15742739060640.0, "grad_norm": 1.7960065364175875, "language_loss": 0.77212912, "learning_rate": 1.173797720205294e-07, "loss": 0.79390705, "num_input_tokens_seen": 160739150, "step": 7432, "time_per_iteration": 3.508333206176758 }, { "auxiliary_loss_clip": 0.01156002, "auxiliary_loss_mlp": 0.01030166, "balance_loss_clip": 1.04685903, "balance_loss_mlp": 1.02241492, "epoch": 0.8937654061203631, "flos": 35115505190880.0, "grad_norm": 2.2031011161758065, "language_loss": 0.71413445, "learning_rate": 1.1711697840513602e-07, "loss": 0.73599619, "num_input_tokens_seen": 160758585, "step": 7433, "time_per_iteration": 2.6208791732788086 }, { "auxiliary_loss_clip": 0.01145038, "auxiliary_loss_mlp": 0.01027148, "balance_loss_clip": 1.04347277, "balance_loss_mlp": 1.02035582, "epoch": 0.8938856490110022, "flos": 16107447823200.0, "grad_norm": 1.9065909406334676, "language_loss": 0.70610809, "learning_rate": 1.1685447042160012e-07, "loss": 0.72782993, "num_input_tokens_seen": 160776620, "step": 7434, "time_per_iteration": 2.450546979904175 }, { "auxiliary_loss_clip": 0.01169768, "auxiliary_loss_mlp": 0.01028853, "balance_loss_clip": 1.04804361, "balance_loss_mlp": 1.02122951, "epoch": 0.8940058919016414, "flos": 20704193359680.0, "grad_norm": 1.6426759535280475, "language_loss": 0.71427238, "learning_rate": 1.1659224810974367e-07, "loss": 0.73625863, "num_input_tokens_seen": 160796580, "step": 7435, "time_per_iteration": 2.4638662338256836 }, { "auxiliary_loss_clip": 0.01139981, "auxiliary_loss_mlp": 0.01022669, "balance_loss_clip": 1.04672301, "balance_loss_mlp": 1.01589751, "epoch": 0.8941261347922804, "flos": 25229045918880.0, "grad_norm": 1.5733933724195568, "language_loss": 0.68488443, "learning_rate": 1.1633031150934591e-07, "loss": 0.70651102, "num_input_tokens_seen": 160819610, "step": 7436, "time_per_iteration": 2.5997259616851807 }, { "auxiliary_loss_clip": 0.01156093, "auxiliary_loss_mlp": 0.01030059, "balance_loss_clip": 1.04792881, "balance_loss_mlp": 1.02287626, "epoch": 0.8942463776829195, "flos": 19537240253760.0, "grad_norm": 1.9309145276733228, "language_loss": 0.80067444, "learning_rate": 1.1606866066014176e-07, "loss": 0.82253599, "num_input_tokens_seen": 160838660, "step": 7437, "time_per_iteration": 2.4747233390808105 }, { "auxiliary_loss_clip": 0.01124604, "auxiliary_loss_mlp": 0.01023591, "balance_loss_clip": 1.04349709, "balance_loss_mlp": 1.01685536, "epoch": 0.8943666205735585, "flos": 22301570488320.0, "grad_norm": 3.071852063208398, "language_loss": 0.7565608, "learning_rate": 1.1580729560182434e-07, "loss": 0.77804279, "num_input_tokens_seen": 160854515, "step": 7438, "time_per_iteration": 2.513991594314575 }, { "auxiliary_loss_clip": 0.01167148, "auxiliary_loss_mlp": 0.00761623, "balance_loss_clip": 1.04690707, "balance_loss_mlp": 1.00053859, "epoch": 0.8944868634641977, "flos": 18912896588640.0, "grad_norm": 2.0592996149139435, "language_loss": 0.70895231, "learning_rate": 1.1554621637404171e-07, "loss": 0.72824001, "num_input_tokens_seen": 160872605, "step": 7439, "time_per_iteration": 3.3327929973602295 }, { "auxiliary_loss_clip": 0.01153466, "auxiliary_loss_mlp": 0.01024848, "balance_loss_clip": 1.04454136, "balance_loss_mlp": 1.01809525, "epoch": 0.8946071063548368, "flos": 14460906764640.0, "grad_norm": 2.4828920928580884, "language_loss": 0.6113925, "learning_rate": 1.1528542301639999e-07, "loss": 0.63317561, "num_input_tokens_seen": 160889395, "step": 7440, "time_per_iteration": 2.451439142227173 }, { "auxiliary_loss_clip": 0.01127159, "auxiliary_loss_mlp": 0.0102559, "balance_loss_clip": 1.04217219, "balance_loss_mlp": 1.01877069, "epoch": 0.8947273492454758, "flos": 20084087895840.0, "grad_norm": 2.2596198787102217, "language_loss": 0.82077765, "learning_rate": 1.1502491556846105e-07, "loss": 0.84230518, "num_input_tokens_seen": 160907890, "step": 7441, "time_per_iteration": 2.560849189758301 }, { "auxiliary_loss_clip": 0.01141213, "auxiliary_loss_mlp": 0.01024443, "balance_loss_clip": 1.04668415, "balance_loss_mlp": 1.01729941, "epoch": 0.894847592136115, "flos": 18550558345440.0, "grad_norm": 2.756584431650395, "language_loss": 0.81392503, "learning_rate": 1.1476469406974331e-07, "loss": 0.83558154, "num_input_tokens_seen": 160923490, "step": 7442, "time_per_iteration": 2.475550889968872 }, { "auxiliary_loss_clip": 0.01165966, "auxiliary_loss_mlp": 0.01018967, "balance_loss_clip": 1.04873371, "balance_loss_mlp": 1.01185918, "epoch": 0.894967835026754, "flos": 23478472592160.0, "grad_norm": 1.6879471541243825, "language_loss": 0.77244163, "learning_rate": 1.1450475855972341e-07, "loss": 0.79429102, "num_input_tokens_seen": 160944280, "step": 7443, "time_per_iteration": 2.497819185256958 }, { "auxiliary_loss_clip": 0.01138743, "auxiliary_loss_mlp": 0.00762206, "balance_loss_clip": 1.04478979, "balance_loss_mlp": 1.00053859, "epoch": 0.8950880779173931, "flos": 15188312940000.0, "grad_norm": 1.9002001466756366, "language_loss": 0.70461649, "learning_rate": 1.1424510907783158e-07, "loss": 0.72362602, "num_input_tokens_seen": 160961560, "step": 7444, "time_per_iteration": 2.486125946044922 }, { "auxiliary_loss_clip": 0.01144816, "auxiliary_loss_mlp": 0.01019809, "balance_loss_clip": 1.04390955, "balance_loss_mlp": 1.01353848, "epoch": 0.8952083208080323, "flos": 22091961522720.0, "grad_norm": 1.6264273050265938, "language_loss": 0.82734287, "learning_rate": 1.1398574566345787e-07, "loss": 0.84898913, "num_input_tokens_seen": 160982195, "step": 7445, "time_per_iteration": 3.3238704204559326 }, { "auxiliary_loss_clip": 0.01145222, "auxiliary_loss_mlp": 0.01029394, "balance_loss_clip": 1.04485106, "balance_loss_mlp": 1.0224086, "epoch": 0.8953285636986713, "flos": 23254029922080.0, "grad_norm": 2.588696964783765, "language_loss": 0.82517558, "learning_rate": 1.1372666835594702e-07, "loss": 0.84692168, "num_input_tokens_seen": 161000520, "step": 7446, "time_per_iteration": 3.2931740283966064 }, { "auxiliary_loss_clip": 0.01139207, "auxiliary_loss_mlp": 0.01022136, "balance_loss_clip": 1.04534853, "balance_loss_mlp": 1.01513219, "epoch": 0.8954488065893104, "flos": 16362664939680.0, "grad_norm": 1.9503478131777323, "language_loss": 0.71989357, "learning_rate": 1.1346787719460071e-07, "loss": 0.74150705, "num_input_tokens_seen": 161019405, "step": 7447, "time_per_iteration": 2.513423442840576 }, { "auxiliary_loss_clip": 0.01139364, "auxiliary_loss_mlp": 0.01019923, "balance_loss_clip": 1.04555774, "balance_loss_mlp": 1.01357174, "epoch": 0.8955690494799495, "flos": 18257886228000.0, "grad_norm": 1.81406304129455, "language_loss": 0.72568536, "learning_rate": 1.1320937221867732e-07, "loss": 0.74727827, "num_input_tokens_seen": 161036985, "step": 7448, "time_per_iteration": 2.516411781311035 }, { "auxiliary_loss_clip": 0.01137119, "auxiliary_loss_mlp": 0.01027533, "balance_loss_clip": 1.04299307, "balance_loss_mlp": 1.02164972, "epoch": 0.8956892923705886, "flos": 25447490456640.0, "grad_norm": 1.7213206792245193, "language_loss": 0.79737115, "learning_rate": 1.1295115346739192e-07, "loss": 0.81901777, "num_input_tokens_seen": 161056985, "step": 7449, "time_per_iteration": 2.5640265941619873 }, { "auxiliary_loss_clip": 0.0114454, "auxiliary_loss_mlp": 0.01024051, "balance_loss_clip": 1.04573584, "balance_loss_mlp": 1.01724672, "epoch": 0.8958095352612276, "flos": 52661890068480.0, "grad_norm": 3.48288128635, "language_loss": 0.72956824, "learning_rate": 1.1269322097991629e-07, "loss": 0.75125408, "num_input_tokens_seen": 161080270, "step": 7450, "time_per_iteration": 2.8075668811798096 }, { "auxiliary_loss_clip": 0.01160295, "auxiliary_loss_mlp": 0.01025192, "balance_loss_clip": 1.04871655, "balance_loss_mlp": 1.01756263, "epoch": 0.8959297781518668, "flos": 23186339229120.0, "grad_norm": 2.2771580540501093, "language_loss": 0.67876101, "learning_rate": 1.1243557479537846e-07, "loss": 0.70061588, "num_input_tokens_seen": 161100160, "step": 7451, "time_per_iteration": 2.488302230834961 }, { "auxiliary_loss_clip": 0.01168846, "auxiliary_loss_mlp": 0.01030467, "balance_loss_clip": 1.04698133, "balance_loss_mlp": 1.0228231, "epoch": 0.8960500210425059, "flos": 20334312554880.0, "grad_norm": 2.2115705210459464, "language_loss": 0.68245161, "learning_rate": 1.121782149528634e-07, "loss": 0.70444471, "num_input_tokens_seen": 161117260, "step": 7452, "time_per_iteration": 2.455167055130005 }, { "auxiliary_loss_clip": 0.0114388, "auxiliary_loss_mlp": 0.01023438, "balance_loss_clip": 1.04953456, "balance_loss_mlp": 1.01682806, "epoch": 0.8961702639331449, "flos": 19901697597600.0, "grad_norm": 3.394728262756261, "language_loss": 0.7864632, "learning_rate": 1.1192114149141208e-07, "loss": 0.80813634, "num_input_tokens_seen": 161136895, "step": 7453, "time_per_iteration": 2.525634765625 }, { "auxiliary_loss_clip": 0.01142523, "auxiliary_loss_mlp": 0.01025692, "balance_loss_clip": 1.04382908, "balance_loss_mlp": 1.01833677, "epoch": 0.8962905068237841, "flos": 12896351349120.0, "grad_norm": 2.6863549430679288, "language_loss": 0.65217912, "learning_rate": 1.1166435445002197e-07, "loss": 0.67386127, "num_input_tokens_seen": 161154565, "step": 7454, "time_per_iteration": 2.5037009716033936 }, { "auxiliary_loss_clip": 0.01156917, "auxiliary_loss_mlp": 0.01028757, "balance_loss_clip": 1.04817986, "balance_loss_mlp": 1.02182484, "epoch": 0.8964107497144231, "flos": 23440335168960.0, "grad_norm": 2.224356470859216, "language_loss": 0.68486881, "learning_rate": 1.1140785386764818e-07, "loss": 0.70672554, "num_input_tokens_seen": 161173265, "step": 7455, "time_per_iteration": 2.490295648574829 }, { "auxiliary_loss_clip": 0.01146787, "auxiliary_loss_mlp": 0.01026163, "balance_loss_clip": 1.0438993, "balance_loss_mlp": 1.01905847, "epoch": 0.8965309926050622, "flos": 19500180339360.0, "grad_norm": 2.123355763386612, "language_loss": 0.696172, "learning_rate": 1.1115163978320153e-07, "loss": 0.71790153, "num_input_tokens_seen": 161191995, "step": 7456, "time_per_iteration": 2.470759868621826 }, { "auxiliary_loss_clip": 0.01158879, "auxiliary_loss_mlp": 0.00762267, "balance_loss_clip": 1.04800606, "balance_loss_mlp": 1.0005486, "epoch": 0.8966512354957014, "flos": 28658012259360.0, "grad_norm": 3.5664108826864225, "language_loss": 0.82365483, "learning_rate": 1.1089571223554917e-07, "loss": 0.8428663, "num_input_tokens_seen": 161212880, "step": 7457, "time_per_iteration": 2.5322794914245605 }, { "auxiliary_loss_clip": 0.01153505, "auxiliary_loss_mlp": 0.01031217, "balance_loss_clip": 1.04347312, "balance_loss_mlp": 1.02398133, "epoch": 0.8967714783863404, "flos": 23370920461920.0, "grad_norm": 1.851383878544963, "language_loss": 0.8556397, "learning_rate": 1.1064007126351537e-07, "loss": 0.87748694, "num_input_tokens_seen": 161233595, "step": 7458, "time_per_iteration": 3.2762742042541504 }, { "auxiliary_loss_clip": 0.01135291, "auxiliary_loss_mlp": 0.010288, "balance_loss_clip": 1.04548001, "balance_loss_mlp": 1.02173722, "epoch": 0.8968917212769795, "flos": 24535179795840.0, "grad_norm": 2.480213917708433, "language_loss": 0.7634244, "learning_rate": 1.1038471690588003e-07, "loss": 0.78506529, "num_input_tokens_seen": 161252740, "step": 7459, "time_per_iteration": 2.5232205390930176 }, { "auxiliary_loss_clip": 0.01113817, "auxiliary_loss_mlp": 0.0103227, "balance_loss_clip": 1.04603982, "balance_loss_mlp": 1.02538562, "epoch": 0.8970119641676186, "flos": 23475419650560.0, "grad_norm": 1.8671558755489233, "language_loss": 0.80319262, "learning_rate": 1.1012964920138145e-07, "loss": 0.82465351, "num_input_tokens_seen": 161272325, "step": 7460, "time_per_iteration": 2.621098279953003 }, { "auxiliary_loss_clip": 0.01133946, "auxiliary_loss_mlp": 0.01021977, "balance_loss_clip": 1.04218102, "balance_loss_mlp": 1.01454449, "epoch": 0.8971322070582577, "flos": 24538196820480.0, "grad_norm": 1.776377377190912, "language_loss": 0.75694448, "learning_rate": 1.0987486818871205e-07, "loss": 0.77850372, "num_input_tokens_seen": 161295915, "step": 7461, "time_per_iteration": 2.5772805213928223 }, { "auxiliary_loss_clip": 0.01153639, "auxiliary_loss_mlp": 0.00762222, "balance_loss_clip": 1.04737115, "balance_loss_mlp": 1.00055981, "epoch": 0.8972524499488967, "flos": 21797457640320.0, "grad_norm": 2.3460126980093237, "language_loss": 0.73144764, "learning_rate": 1.0962037390652245e-07, "loss": 0.75060624, "num_input_tokens_seen": 161314935, "step": 7462, "time_per_iteration": 2.517014265060425 }, { "auxiliary_loss_clip": 0.01142698, "auxiliary_loss_mlp": 0.01021875, "balance_loss_clip": 1.04634142, "balance_loss_mlp": 1.01483846, "epoch": 0.8973726928395359, "flos": 21726247085280.0, "grad_norm": 2.4075443831693506, "language_loss": 0.72242993, "learning_rate": 1.0936616639341911e-07, "loss": 0.74407566, "num_input_tokens_seen": 161335225, "step": 7463, "time_per_iteration": 2.532789707183838 }, { "auxiliary_loss_clip": 0.0104833, "auxiliary_loss_mlp": 0.00999466, "balance_loss_clip": 1.00916719, "balance_loss_mlp": 0.99841684, "epoch": 0.897492935730175, "flos": 53837110463520.0, "grad_norm": 0.7858830340691062, "language_loss": 0.54824102, "learning_rate": 1.0911224568796473e-07, "loss": 0.56871903, "num_input_tokens_seen": 161393420, "step": 7464, "time_per_iteration": 3.140841007232666 }, { "auxiliary_loss_clip": 0.01153217, "auxiliary_loss_mlp": 0.01026926, "balance_loss_clip": 1.0481143, "balance_loss_mlp": 1.0204885, "epoch": 0.897613178620814, "flos": 18290348771520.0, "grad_norm": 1.8349364810897562, "language_loss": 0.70693445, "learning_rate": 1.0885861182867984e-07, "loss": 0.72873586, "num_input_tokens_seen": 161411525, "step": 7465, "time_per_iteration": 3.2130823135375977 }, { "auxiliary_loss_clip": 0.01142655, "auxiliary_loss_mlp": 0.01022479, "balance_loss_clip": 1.04490423, "balance_loss_mlp": 1.01560032, "epoch": 0.8977334215114532, "flos": 32993722131840.0, "grad_norm": 3.543810173289813, "language_loss": 0.70619857, "learning_rate": 1.0860526485403942e-07, "loss": 0.72784996, "num_input_tokens_seen": 161432800, "step": 7466, "time_per_iteration": 2.6358065605163574 }, { "auxiliary_loss_clip": 0.01167339, "auxiliary_loss_mlp": 0.01022776, "balance_loss_clip": 1.04763162, "balance_loss_mlp": 1.01571, "epoch": 0.8978536644020922, "flos": 15195639999840.0, "grad_norm": 1.570409660437083, "language_loss": 0.77040458, "learning_rate": 1.0835220480247675e-07, "loss": 0.79230583, "num_input_tokens_seen": 161451295, "step": 7467, "time_per_iteration": 2.443830728530884 }, { "auxiliary_loss_clip": 0.01138307, "auxiliary_loss_mlp": 0.01025065, "balance_loss_clip": 1.04568255, "balance_loss_mlp": 1.01818681, "epoch": 0.8979739072927313, "flos": 18004393125600.0, "grad_norm": 1.9740230500285816, "language_loss": 0.83845508, "learning_rate": 1.0809943171238067e-07, "loss": 0.86008883, "num_input_tokens_seen": 161469220, "step": 7468, "time_per_iteration": 2.5020267963409424 }, { "auxiliary_loss_clip": 0.01148681, "auxiliary_loss_mlp": 0.01027515, "balance_loss_clip": 1.04790711, "balance_loss_mlp": 1.01987958, "epoch": 0.8980941501833704, "flos": 22271550298080.0, "grad_norm": 1.937655498426031, "language_loss": 0.62629092, "learning_rate": 1.078469456220965e-07, "loss": 0.64805287, "num_input_tokens_seen": 161489375, "step": 7469, "time_per_iteration": 2.5375208854675293 }, { "auxiliary_loss_clip": 0.01150964, "auxiliary_loss_mlp": 0.01029037, "balance_loss_clip": 1.04444444, "balance_loss_mlp": 1.02223945, "epoch": 0.8982143930740095, "flos": 37560734813760.0, "grad_norm": 1.8039512860316183, "language_loss": 0.69679523, "learning_rate": 1.0759474656992606e-07, "loss": 0.71859527, "num_input_tokens_seen": 161512145, "step": 7470, "time_per_iteration": 2.636214017868042 }, { "auxiliary_loss_clip": 0.01144872, "auxiliary_loss_mlp": 0.01025173, "balance_loss_clip": 1.04361987, "balance_loss_mlp": 1.01782954, "epoch": 0.8983346359646486, "flos": 18076896691200.0, "grad_norm": 2.3034384546579334, "language_loss": 0.77670956, "learning_rate": 1.0734283459412785e-07, "loss": 0.79841, "num_input_tokens_seen": 161528995, "step": 7471, "time_per_iteration": 3.2641141414642334 }, { "auxiliary_loss_clip": 0.01117297, "auxiliary_loss_mlp": 0.01022694, "balance_loss_clip": 1.04249787, "balance_loss_mlp": 1.01551163, "epoch": 0.8984548788552876, "flos": 20558898892800.0, "grad_norm": 1.6630053842512893, "language_loss": 0.80405229, "learning_rate": 1.0709120973291707e-07, "loss": 0.82545221, "num_input_tokens_seen": 161548775, "step": 7472, "time_per_iteration": 3.2989888191223145 }, { "auxiliary_loss_clip": 0.01172474, "auxiliary_loss_mlp": 0.01029792, "balance_loss_clip": 1.0512979, "balance_loss_mlp": 1.0223918, "epoch": 0.8985751217459268, "flos": 17785445750400.0, "grad_norm": 2.133590015328565, "language_loss": 0.77660036, "learning_rate": 1.0683987202446475e-07, "loss": 0.79862303, "num_input_tokens_seen": 161566960, "step": 7473, "time_per_iteration": 2.4464328289031982 }, { "auxiliary_loss_clip": 0.01157357, "auxiliary_loss_mlp": 0.01023416, "balance_loss_clip": 1.04593253, "balance_loss_mlp": 1.01648724, "epoch": 0.8986953646365659, "flos": 21617006857920.0, "grad_norm": 1.7553513432619419, "language_loss": 0.69650996, "learning_rate": 1.0658882150689862e-07, "loss": 0.71831763, "num_input_tokens_seen": 161585820, "step": 7474, "time_per_iteration": 2.49820613861084 }, { "auxiliary_loss_clip": 0.01128731, "auxiliary_loss_mlp": 0.01023403, "balance_loss_clip": 1.0434382, "balance_loss_mlp": 1.01656985, "epoch": 0.8988156075272049, "flos": 14027357966400.0, "grad_norm": 2.3342561399576662, "language_loss": 0.78692919, "learning_rate": 1.0633805821830288e-07, "loss": 0.80845052, "num_input_tokens_seen": 161602505, "step": 7475, "time_per_iteration": 2.521698236465454 }, { "auxiliary_loss_clip": 0.01140037, "auxiliary_loss_mlp": 0.01028824, "balance_loss_clip": 1.0440917, "balance_loss_mlp": 1.02148974, "epoch": 0.8989358504178441, "flos": 29059206264960.0, "grad_norm": 8.14216617843104, "language_loss": 0.83036369, "learning_rate": 1.0608758219671753e-07, "loss": 0.85205233, "num_input_tokens_seen": 161621545, "step": 7476, "time_per_iteration": 2.5739846229553223 }, { "auxiliary_loss_clip": 0.01145783, "auxiliary_loss_mlp": 0.01025654, "balance_loss_clip": 1.04617035, "balance_loss_mlp": 1.01884437, "epoch": 0.8990560933084831, "flos": 20230423954560.0, "grad_norm": 1.711230386733151, "language_loss": 0.70564187, "learning_rate": 1.0583739348014065e-07, "loss": 0.72735626, "num_input_tokens_seen": 161642630, "step": 7477, "time_per_iteration": 2.5390195846557617 }, { "auxiliary_loss_clip": 0.0116849, "auxiliary_loss_mlp": 0.01023125, "balance_loss_clip": 1.04830551, "balance_loss_mlp": 1.0159483, "epoch": 0.8991763361991222, "flos": 25520676444480.0, "grad_norm": 1.927159041936174, "language_loss": 0.84736043, "learning_rate": 1.0558749210652518e-07, "loss": 0.86927658, "num_input_tokens_seen": 161662560, "step": 7478, "time_per_iteration": 2.4930355548858643 }, { "auxiliary_loss_clip": 0.01132248, "auxiliary_loss_mlp": 0.01026313, "balance_loss_clip": 1.04449391, "balance_loss_mlp": 1.01980686, "epoch": 0.8992965790897613, "flos": 25119195103200.0, "grad_norm": 1.6816754504252343, "language_loss": 0.85429448, "learning_rate": 1.053378781137808e-07, "loss": 0.87588006, "num_input_tokens_seen": 161683480, "step": 7479, "time_per_iteration": 2.5870256423950195 }, { "auxiliary_loss_clip": 0.0114102, "auxiliary_loss_mlp": 0.0102733, "balance_loss_clip": 1.04625821, "balance_loss_mlp": 1.02074337, "epoch": 0.8994168219804004, "flos": 16070818912320.0, "grad_norm": 1.9333705986721623, "language_loss": 0.77802706, "learning_rate": 1.0508855153977392e-07, "loss": 0.79971051, "num_input_tokens_seen": 161699945, "step": 7480, "time_per_iteration": 2.491135358810425 }, { "auxiliary_loss_clip": 0.0115247, "auxiliary_loss_mlp": 0.0102513, "balance_loss_clip": 1.0435735, "balance_loss_mlp": 1.01813245, "epoch": 0.8995370648710395, "flos": 24825768729600.0, "grad_norm": 2.320647936917176, "language_loss": 0.67198765, "learning_rate": 1.0483951242232669e-07, "loss": 0.69376373, "num_input_tokens_seen": 161720420, "step": 7481, "time_per_iteration": 2.5200140476226807 }, { "auxiliary_loss_clip": 0.0105933, "auxiliary_loss_mlp": 0.01000602, "balance_loss_clip": 1.00540924, "balance_loss_mlp": 0.99936265, "epoch": 0.8996573077616786, "flos": 63116250709440.0, "grad_norm": 0.9731228747462083, "language_loss": 0.5775584, "learning_rate": 1.0459076079921936e-07, "loss": 0.59815776, "num_input_tokens_seen": 161773080, "step": 7482, "time_per_iteration": 3.1214864253997803 }, { "auxiliary_loss_clip": 0.01132656, "auxiliary_loss_mlp": 0.01034647, "balance_loss_clip": 1.04398441, "balance_loss_mlp": 1.02751839, "epoch": 0.8997775506523177, "flos": 18219676970880.0, "grad_norm": 2.1919411953494836, "language_loss": 0.84756088, "learning_rate": 1.0434229670818618e-07, "loss": 0.86923397, "num_input_tokens_seen": 161789755, "step": 7483, "time_per_iteration": 2.4641616344451904 }, { "auxiliary_loss_clip": 0.01133814, "auxiliary_loss_mlp": 0.01028604, "balance_loss_clip": 1.04365718, "balance_loss_mlp": 1.02154636, "epoch": 0.8998977935429567, "flos": 24166771586400.0, "grad_norm": 1.6704108332333916, "language_loss": 0.79790831, "learning_rate": 1.0409412018691944e-07, "loss": 0.81953239, "num_input_tokens_seen": 161810220, "step": 7484, "time_per_iteration": 3.2893078327178955 }, { "auxiliary_loss_clip": 0.01136313, "auxiliary_loss_mlp": 0.01024208, "balance_loss_clip": 1.04447532, "balance_loss_mlp": 1.01632881, "epoch": 0.9000180364335959, "flos": 20773033395360.0, "grad_norm": 1.8217543069550945, "language_loss": 0.7483018, "learning_rate": 1.0384623127306724e-07, "loss": 0.76990706, "num_input_tokens_seen": 161827565, "step": 7485, "time_per_iteration": 2.5235378742218018 }, { "auxiliary_loss_clip": 0.0112059, "auxiliary_loss_mlp": 0.01022507, "balance_loss_clip": 1.03959394, "balance_loss_mlp": 1.01594758, "epoch": 0.900138279324235, "flos": 19205748290880.0, "grad_norm": 1.767120321030453, "language_loss": 0.79419404, "learning_rate": 1.0359863000423397e-07, "loss": 0.81562501, "num_input_tokens_seen": 161845700, "step": 7486, "time_per_iteration": 2.5832974910736084 }, { "auxiliary_loss_clip": 0.01168035, "auxiliary_loss_mlp": 0.01027995, "balance_loss_clip": 1.04694343, "balance_loss_mlp": 1.02109241, "epoch": 0.900258522214874, "flos": 28731162330240.0, "grad_norm": 1.5663059876797156, "language_loss": 0.7175836, "learning_rate": 1.0335131641798112e-07, "loss": 0.73954391, "num_input_tokens_seen": 161867660, "step": 7487, "time_per_iteration": 2.508596897125244 }, { "auxiliary_loss_clip": 0.01040263, "auxiliary_loss_mlp": 0.01002237, "balance_loss_clip": 1.00589383, "balance_loss_mlp": 1.00098515, "epoch": 0.9003787651055132, "flos": 58280695718880.0, "grad_norm": 0.8116515154422682, "language_loss": 0.55645651, "learning_rate": 1.0310429055182512e-07, "loss": 0.57688153, "num_input_tokens_seen": 161921980, "step": 7488, "time_per_iteration": 2.9756832122802734 }, { "auxiliary_loss_clip": 0.01130146, "auxiliary_loss_mlp": 0.01029981, "balance_loss_clip": 1.04560518, "balance_loss_mlp": 1.02293003, "epoch": 0.9004990079961522, "flos": 25556479265280.0, "grad_norm": 1.8716802779597173, "language_loss": 0.73978126, "learning_rate": 1.0285755244324024e-07, "loss": 0.76138258, "num_input_tokens_seen": 161942725, "step": 7489, "time_per_iteration": 2.598557472229004 }, { "auxiliary_loss_clip": 0.01142662, "auxiliary_loss_mlp": 0.0076163, "balance_loss_clip": 1.04407978, "balance_loss_mlp": 1.00051403, "epoch": 0.9006192508867913, "flos": 23335189475040.0, "grad_norm": 1.548256991922202, "language_loss": 0.68596071, "learning_rate": 1.0261110212965629e-07, "loss": 0.70500362, "num_input_tokens_seen": 161964520, "step": 7490, "time_per_iteration": 2.5859344005584717 }, { "auxiliary_loss_clip": 0.01140377, "auxiliary_loss_mlp": 0.01021789, "balance_loss_clip": 1.04548657, "balance_loss_mlp": 1.01536953, "epoch": 0.9007394937774305, "flos": 18040303697280.0, "grad_norm": 1.9027948242859556, "language_loss": 0.79056782, "learning_rate": 1.023649396484596e-07, "loss": 0.81218946, "num_input_tokens_seen": 161983575, "step": 7491, "time_per_iteration": 3.222874879837036 }, { "auxiliary_loss_clip": 0.01169655, "auxiliary_loss_mlp": 0.01026059, "balance_loss_clip": 1.0487752, "balance_loss_mlp": 1.01892757, "epoch": 0.9008597366680695, "flos": 43068462083520.0, "grad_norm": 2.305242726906074, "language_loss": 0.67633176, "learning_rate": 1.0211906503699275e-07, "loss": 0.69828892, "num_input_tokens_seen": 162006550, "step": 7492, "time_per_iteration": 2.666166305541992 }, { "auxiliary_loss_clip": 0.01157952, "auxiliary_loss_mlp": 0.01027431, "balance_loss_clip": 1.04825318, "balance_loss_mlp": 1.02096725, "epoch": 0.9009799795587086, "flos": 14939058038880.0, "grad_norm": 2.393308448456894, "language_loss": 0.81781042, "learning_rate": 1.0187347833255455e-07, "loss": 0.83966422, "num_input_tokens_seen": 162022455, "step": 7493, "time_per_iteration": 2.4542365074157715 }, { "auxiliary_loss_clip": 0.01165131, "auxiliary_loss_mlp": 0.01023915, "balance_loss_clip": 1.04776239, "balance_loss_mlp": 1.01741183, "epoch": 0.9011002224493477, "flos": 21579587773920.0, "grad_norm": 2.278339828265342, "language_loss": 0.79688621, "learning_rate": 1.0162817957240056e-07, "loss": 0.81877673, "num_input_tokens_seen": 162042350, "step": 7494, "time_per_iteration": 2.478501081466675 }, { "auxiliary_loss_clip": 0.01050367, "auxiliary_loss_mlp": 0.01000566, "balance_loss_clip": 1.0052917, "balance_loss_mlp": 0.99938011, "epoch": 0.9012204653399868, "flos": 71166379730880.0, "grad_norm": 0.8954874271837989, "language_loss": 0.63048494, "learning_rate": 1.0138316879374253e-07, "loss": 0.65099424, "num_input_tokens_seen": 162111640, "step": 7495, "time_per_iteration": 3.2376465797424316 }, { "auxiliary_loss_clip": 0.01144548, "auxiliary_loss_mlp": 0.01024572, "balance_loss_clip": 1.04826999, "balance_loss_mlp": 1.01802135, "epoch": 0.9013407082306258, "flos": 15594966323520.0, "grad_norm": 2.020183850154325, "language_loss": 0.74217665, "learning_rate": 1.0113844603374833e-07, "loss": 0.76386786, "num_input_tokens_seen": 162128165, "step": 7496, "time_per_iteration": 2.504608631134033 }, { "auxiliary_loss_clip": 0.01136317, "auxiliary_loss_mlp": 0.010262, "balance_loss_clip": 1.04099822, "balance_loss_mlp": 1.01865387, "epoch": 0.901460951121265, "flos": 15049160273280.0, "grad_norm": 2.2587039394975377, "language_loss": 0.71703368, "learning_rate": 1.0089401132954178e-07, "loss": 0.73865891, "num_input_tokens_seen": 162146145, "step": 7497, "time_per_iteration": 2.5064783096313477 }, { "auxiliary_loss_clip": 0.01138909, "auxiliary_loss_mlp": 0.01025898, "balance_loss_clip": 1.04452753, "balance_loss_mlp": 1.01916862, "epoch": 0.9015811940119041, "flos": 22236860903040.0, "grad_norm": 2.3219303462475063, "language_loss": 0.72296816, "learning_rate": 1.006498647182037e-07, "loss": 0.74461627, "num_input_tokens_seen": 162164800, "step": 7498, "time_per_iteration": 4.25041127204895 }, { "auxiliary_loss_clip": 0.0109352, "auxiliary_loss_mlp": 0.01029551, "balance_loss_clip": 1.03870916, "balance_loss_mlp": 1.02252078, "epoch": 0.9017014369025431, "flos": 24973828802400.0, "grad_norm": 2.1476180837935286, "language_loss": 0.7123872, "learning_rate": 1.004060062367713e-07, "loss": 0.7336179, "num_input_tokens_seen": 162185895, "step": 7499, "time_per_iteration": 2.6514337062835693 }, { "auxiliary_loss_clip": 0.01153781, "auxiliary_loss_mlp": 0.01020425, "balance_loss_clip": 1.04462671, "balance_loss_mlp": 1.01347828, "epoch": 0.9018216797931822, "flos": 18114172107360.0, "grad_norm": 1.9185837813793445, "language_loss": 0.69586271, "learning_rate": 1.0016243592223728e-07, "loss": 0.71760476, "num_input_tokens_seen": 162206295, "step": 7500, "time_per_iteration": 2.503235340118408 }, { "auxiliary_loss_clip": 0.0109508, "auxiliary_loss_mlp": 0.01027812, "balance_loss_clip": 1.04044223, "balance_loss_mlp": 1.02089238, "epoch": 0.9019419226838213, "flos": 37268457782880.0, "grad_norm": 2.033260627216769, "language_loss": 0.65315342, "learning_rate": 9.991915381155114e-08, "loss": 0.67438233, "num_input_tokens_seen": 162229275, "step": 7501, "time_per_iteration": 2.7510087490081787 }, { "auxiliary_loss_clip": 0.01159227, "auxiliary_loss_mlp": 0.01027833, "balance_loss_clip": 1.04716396, "balance_loss_mlp": 1.02096331, "epoch": 0.9020621655744604, "flos": 23441125342080.0, "grad_norm": 2.0388314361038855, "language_loss": 0.74581361, "learning_rate": 9.967615994161871e-08, "loss": 0.76768422, "num_input_tokens_seen": 162248935, "step": 7502, "time_per_iteration": 2.516348123550415 }, { "auxiliary_loss_clip": 0.011666, "auxiliary_loss_mlp": 0.01025625, "balance_loss_clip": 1.04724514, "balance_loss_mlp": 1.01890182, "epoch": 0.9021824084650995, "flos": 22857469204320.0, "grad_norm": 1.6523497281653798, "language_loss": 0.78103364, "learning_rate": 9.943345434930161e-08, "loss": 0.80295593, "num_input_tokens_seen": 162269185, "step": 7503, "time_per_iteration": 2.4600701332092285 }, { "auxiliary_loss_clip": 0.01126595, "auxiliary_loss_mlp": 0.01026778, "balance_loss_clip": 1.04606915, "balance_loss_mlp": 1.02010489, "epoch": 0.9023026513557386, "flos": 22127584758720.0, "grad_norm": 2.078873159465074, "language_loss": 0.68852997, "learning_rate": 9.919103707141885e-08, "loss": 0.7100637, "num_input_tokens_seen": 162288065, "step": 7504, "time_per_iteration": 2.5726935863494873 }, { "auxiliary_loss_clip": 0.01154463, "auxiliary_loss_mlp": 0.01028525, "balance_loss_clip": 1.0475204, "balance_loss_mlp": 1.0211668, "epoch": 0.9024228942463777, "flos": 24199090462080.0, "grad_norm": 1.7585682254342767, "language_loss": 0.7634263, "learning_rate": 9.89489081447441e-08, "loss": 0.78525615, "num_input_tokens_seen": 162305265, "step": 7505, "time_per_iteration": 2.4787545204162598 }, { "auxiliary_loss_clip": 0.01138165, "auxiliary_loss_mlp": 0.01025238, "balance_loss_clip": 1.04310179, "balance_loss_mlp": 1.01796281, "epoch": 0.9025431371370167, "flos": 25008266778720.0, "grad_norm": 1.834763449590264, "language_loss": 0.82906997, "learning_rate": 9.870706760600844e-08, "loss": 0.85070395, "num_input_tokens_seen": 162325215, "step": 7506, "time_per_iteration": 2.5557212829589844 }, { "auxiliary_loss_clip": 0.01116438, "auxiliary_loss_mlp": 0.01026028, "balance_loss_clip": 1.04673553, "balance_loss_mlp": 1.01917326, "epoch": 0.9026633800276559, "flos": 18952865776800.0, "grad_norm": 1.9377469630684034, "language_loss": 0.72825938, "learning_rate": 9.846551549189918e-08, "loss": 0.74968398, "num_input_tokens_seen": 162344820, "step": 7507, "time_per_iteration": 2.553088903427124 }, { "auxiliary_loss_clip": 0.01139926, "auxiliary_loss_mlp": 0.01025189, "balance_loss_clip": 1.04633665, "balance_loss_mlp": 1.01824486, "epoch": 0.902783622918295, "flos": 32416064126400.0, "grad_norm": 2.5692219156595133, "language_loss": 0.68908381, "learning_rate": 9.822425183905902e-08, "loss": 0.71073496, "num_input_tokens_seen": 162365345, "step": 7508, "time_per_iteration": 2.6396119594573975 }, { "auxiliary_loss_clip": 0.01030628, "auxiliary_loss_mlp": 0.01002017, "balance_loss_clip": 1.00535083, "balance_loss_mlp": 1.00076509, "epoch": 0.902903865808934, "flos": 63717463852320.0, "grad_norm": 0.9168825094892656, "language_loss": 0.75272107, "learning_rate": 9.798327668408823e-08, "loss": 0.77304757, "num_input_tokens_seen": 162426980, "step": 7509, "time_per_iteration": 3.239729404449463 }, { "auxiliary_loss_clip": 0.01171539, "auxiliary_loss_mlp": 0.01027987, "balance_loss_clip": 1.0475204, "balance_loss_mlp": 1.0205065, "epoch": 0.9030241086995732, "flos": 23804038256640.0, "grad_norm": 2.1589762886552624, "language_loss": 0.68793535, "learning_rate": 9.774259006354158e-08, "loss": 0.7099306, "num_input_tokens_seen": 162447050, "step": 7510, "time_per_iteration": 3.2052369117736816 }, { "auxiliary_loss_clip": 0.01143914, "auxiliary_loss_mlp": 0.01025758, "balance_loss_clip": 1.04442549, "balance_loss_mlp": 1.0189693, "epoch": 0.9031443515902122, "flos": 26395891273920.0, "grad_norm": 2.1218050090125504, "language_loss": 0.75904959, "learning_rate": 9.750219201393184e-08, "loss": 0.78074634, "num_input_tokens_seen": 162467015, "step": 7511, "time_per_iteration": 2.5764737129211426 }, { "auxiliary_loss_clip": 0.01152084, "auxiliary_loss_mlp": 0.01020042, "balance_loss_clip": 1.04533589, "balance_loss_mlp": 1.01318192, "epoch": 0.9032645944808513, "flos": 24939354909120.0, "grad_norm": 1.7828137861617397, "language_loss": 0.77873385, "learning_rate": 9.726208257172697e-08, "loss": 0.80045509, "num_input_tokens_seen": 162488710, "step": 7512, "time_per_iteration": 2.5394015312194824 }, { "auxiliary_loss_clip": 0.01166685, "auxiliary_loss_mlp": 0.01023681, "balance_loss_clip": 1.0472728, "balance_loss_mlp": 1.01614344, "epoch": 0.9033848373714904, "flos": 21178824771840.0, "grad_norm": 2.459593517566949, "language_loss": 0.74755067, "learning_rate": 9.702226177335115e-08, "loss": 0.76945436, "num_input_tokens_seen": 162507205, "step": 7513, "time_per_iteration": 2.4480483531951904 }, { "auxiliary_loss_clip": 0.01143683, "auxiliary_loss_mlp": 0.01023653, "balance_loss_clip": 1.04980421, "balance_loss_mlp": 1.01643491, "epoch": 0.9035050802621295, "flos": 26286363710880.0, "grad_norm": 1.5348660472019073, "language_loss": 0.72543091, "learning_rate": 9.67827296551853e-08, "loss": 0.74710429, "num_input_tokens_seen": 162528490, "step": 7514, "time_per_iteration": 2.544106960296631 }, { "auxiliary_loss_clip": 0.01131732, "auxiliary_loss_mlp": 0.00762093, "balance_loss_clip": 1.04180765, "balance_loss_mlp": 1.00055563, "epoch": 0.9036253231527686, "flos": 24204549840000.0, "grad_norm": 2.1550323854401943, "language_loss": 0.68183964, "learning_rate": 9.65434862535659e-08, "loss": 0.70077795, "num_input_tokens_seen": 162547860, "step": 7515, "time_per_iteration": 2.550680637359619 }, { "auxiliary_loss_clip": 0.01144812, "auxiliary_loss_mlp": 0.01021127, "balance_loss_clip": 1.04582191, "balance_loss_mlp": 1.01387906, "epoch": 0.9037455660434077, "flos": 18072658489920.0, "grad_norm": 2.7411016045918783, "language_loss": 0.65122819, "learning_rate": 9.630453160478635e-08, "loss": 0.67288768, "num_input_tokens_seen": 162563215, "step": 7516, "time_per_iteration": 2.462005853652954 }, { "auxiliary_loss_clip": 0.0111276, "auxiliary_loss_mlp": 0.01024641, "balance_loss_clip": 1.04236972, "balance_loss_mlp": 1.01807213, "epoch": 0.9038658089340468, "flos": 24060799802400.0, "grad_norm": 1.5311086505125948, "language_loss": 0.8236919, "learning_rate": 9.60658657450959e-08, "loss": 0.84506595, "num_input_tokens_seen": 162583515, "step": 7517, "time_per_iteration": 3.5354299545288086 }, { "auxiliary_loss_clip": 0.01126351, "auxiliary_loss_mlp": 0.01021536, "balance_loss_clip": 1.04171491, "balance_loss_mlp": 1.01489639, "epoch": 0.9039860518246858, "flos": 21834302052960.0, "grad_norm": 2.1148712022697107, "language_loss": 0.79724491, "learning_rate": 9.582748871069979e-08, "loss": 0.81872374, "num_input_tokens_seen": 162602955, "step": 7518, "time_per_iteration": 2.5001626014709473 }, { "auxiliary_loss_clip": 0.01144382, "auxiliary_loss_mlp": 0.00761903, "balance_loss_clip": 1.04526794, "balance_loss_mlp": 1.00055981, "epoch": 0.904106294715325, "flos": 26614874566080.0, "grad_norm": 2.092936859095697, "language_loss": 0.83308923, "learning_rate": 9.558940053775954e-08, "loss": 0.85215205, "num_input_tokens_seen": 162621595, "step": 7519, "time_per_iteration": 2.5713577270507812 }, { "auxiliary_loss_clip": 0.01154, "auxiliary_loss_mlp": 0.0102778, "balance_loss_clip": 1.04695702, "balance_loss_mlp": 1.02053237, "epoch": 0.904226537605964, "flos": 17785697169120.0, "grad_norm": 1.9641979722367031, "language_loss": 0.67825449, "learning_rate": 9.535160126239294e-08, "loss": 0.70007223, "num_input_tokens_seen": 162638220, "step": 7520, "time_per_iteration": 2.4469552040100098 }, { "auxiliary_loss_clip": 0.01155009, "auxiliary_loss_mlp": 0.01023231, "balance_loss_clip": 1.04778028, "balance_loss_mlp": 1.01652598, "epoch": 0.9043467804966031, "flos": 24790432829280.0, "grad_norm": 1.6483932064033735, "language_loss": 0.70815086, "learning_rate": 9.511409092067424e-08, "loss": 0.72993326, "num_input_tokens_seen": 162658575, "step": 7521, "time_per_iteration": 2.5291128158569336 }, { "auxiliary_loss_clip": 0.01141278, "auxiliary_loss_mlp": 0.01023613, "balance_loss_clip": 1.04753709, "balance_loss_mlp": 1.01623356, "epoch": 0.9044670233872423, "flos": 22632128610240.0, "grad_norm": 2.0481003560764046, "language_loss": 0.67505717, "learning_rate": 9.487686954863327e-08, "loss": 0.69670606, "num_input_tokens_seen": 162678295, "step": 7522, "time_per_iteration": 2.508415460586548 }, { "auxiliary_loss_clip": 0.01156657, "auxiliary_loss_mlp": 0.01026484, "balance_loss_clip": 1.04898417, "balance_loss_mlp": 1.01949883, "epoch": 0.9045872662778813, "flos": 23771324294400.0, "grad_norm": 2.292956890868452, "language_loss": 0.77128708, "learning_rate": 9.46399371822566e-08, "loss": 0.79311854, "num_input_tokens_seen": 162698070, "step": 7523, "time_per_iteration": 3.3207149505615234 }, { "auxiliary_loss_clip": 0.01167467, "auxiliary_loss_mlp": 0.0102709, "balance_loss_clip": 1.04667568, "balance_loss_mlp": 1.01998484, "epoch": 0.9047075091685204, "flos": 15191042628960.0, "grad_norm": 2.414078375368101, "language_loss": 0.72140443, "learning_rate": 9.440329385748657e-08, "loss": 0.74334997, "num_input_tokens_seen": 162715140, "step": 7524, "time_per_iteration": 3.1989431381225586 }, { "auxiliary_loss_clip": 0.0112429, "auxiliary_loss_mlp": 0.0102798, "balance_loss_clip": 1.04458642, "balance_loss_mlp": 1.02136087, "epoch": 0.9048277520591596, "flos": 18003710703360.0, "grad_norm": 2.0511820128379, "language_loss": 0.70503092, "learning_rate": 9.416693961022137e-08, "loss": 0.72655368, "num_input_tokens_seen": 162733390, "step": 7525, "time_per_iteration": 2.520369052886963 }, { "auxiliary_loss_clip": 0.01083953, "auxiliary_loss_mlp": 0.01024144, "balance_loss_clip": 1.03710079, "balance_loss_mlp": 1.01765919, "epoch": 0.9049479949497986, "flos": 21872475393120.0, "grad_norm": 1.9220269239246672, "language_loss": 0.7706039, "learning_rate": 9.393087447631654e-08, "loss": 0.79168493, "num_input_tokens_seen": 162751670, "step": 7526, "time_per_iteration": 2.619542121887207 }, { "auxiliary_loss_clip": 0.01135992, "auxiliary_loss_mlp": 0.01021313, "balance_loss_clip": 1.04399657, "balance_loss_mlp": 1.01479506, "epoch": 0.9050682378404377, "flos": 20773931319360.0, "grad_norm": 1.8726765547068505, "language_loss": 0.72727662, "learning_rate": 9.36950984915823e-08, "loss": 0.74884963, "num_input_tokens_seen": 162770025, "step": 7527, "time_per_iteration": 2.50856351852417 }, { "auxiliary_loss_clip": 0.01170442, "auxiliary_loss_mlp": 0.01023793, "balance_loss_clip": 1.0494318, "balance_loss_mlp": 1.01698875, "epoch": 0.9051884807310768, "flos": 21580018777440.0, "grad_norm": 1.6635865610239928, "language_loss": 0.68941963, "learning_rate": 9.345961169178607e-08, "loss": 0.711362, "num_input_tokens_seen": 162789710, "step": 7528, "time_per_iteration": 2.4866623878479004 }, { "auxiliary_loss_clip": 0.01116894, "auxiliary_loss_mlp": 0.0102721, "balance_loss_clip": 1.04735219, "balance_loss_mlp": 1.02052212, "epoch": 0.9053087236217159, "flos": 21908062712160.0, "grad_norm": 1.4439468923620327, "language_loss": 0.72597265, "learning_rate": 9.322441411265081e-08, "loss": 0.74741364, "num_input_tokens_seen": 162810695, "step": 7529, "time_per_iteration": 2.552551746368408 }, { "auxiliary_loss_clip": 0.01135688, "auxiliary_loss_mlp": 0.01023767, "balance_loss_clip": 1.04556251, "balance_loss_mlp": 1.01689792, "epoch": 0.9054289665123549, "flos": 17055812723520.0, "grad_norm": 2.1059371095643398, "language_loss": 0.73103714, "learning_rate": 9.298950578985554e-08, "loss": 0.75263166, "num_input_tokens_seen": 162827770, "step": 7530, "time_per_iteration": 2.5108139514923096 }, { "auxiliary_loss_clip": 0.01149358, "auxiliary_loss_mlp": 0.00762206, "balance_loss_clip": 1.04684103, "balance_loss_mlp": 1.00049078, "epoch": 0.905549209402994, "flos": 20777271596640.0, "grad_norm": 1.9217371388973299, "language_loss": 0.70915389, "learning_rate": 9.275488675903665e-08, "loss": 0.72826958, "num_input_tokens_seen": 162846715, "step": 7531, "time_per_iteration": 2.4808642864227295 }, { "auxiliary_loss_clip": 0.01112355, "auxiliary_loss_mlp": 0.01023885, "balance_loss_clip": 1.04384136, "balance_loss_mlp": 1.01693809, "epoch": 0.9056694522936332, "flos": 21686816651520.0, "grad_norm": 2.3944507743042984, "language_loss": 0.73715532, "learning_rate": 9.252055705578454e-08, "loss": 0.75851774, "num_input_tokens_seen": 162866215, "step": 7532, "time_per_iteration": 2.609435558319092 }, { "auxiliary_loss_clip": 0.0115297, "auxiliary_loss_mlp": 0.01023601, "balance_loss_clip": 1.04473937, "balance_loss_mlp": 1.01744986, "epoch": 0.9057896951842722, "flos": 29569137660480.0, "grad_norm": 1.6131923654597993, "language_loss": 0.72157836, "learning_rate": 9.228651671564747e-08, "loss": 0.74334407, "num_input_tokens_seen": 162888245, "step": 7533, "time_per_iteration": 2.541398525238037 }, { "auxiliary_loss_clip": 0.0110823, "auxiliary_loss_mlp": 0.01023535, "balance_loss_clip": 1.04512525, "balance_loss_mlp": 1.01708007, "epoch": 0.9059099380749113, "flos": 27892253159040.0, "grad_norm": 1.4956025724853848, "language_loss": 0.77677131, "learning_rate": 9.205276577412901e-08, "loss": 0.79808897, "num_input_tokens_seen": 162911025, "step": 7534, "time_per_iteration": 2.645186424255371 }, { "auxiliary_loss_clip": 0.0114539, "auxiliary_loss_mlp": 0.00761764, "balance_loss_clip": 1.04307723, "balance_loss_mlp": 1.00054073, "epoch": 0.9060301809655504, "flos": 17748996424320.0, "grad_norm": 4.913737943105547, "language_loss": 0.76974851, "learning_rate": 9.181930426668905e-08, "loss": 0.78882003, "num_input_tokens_seen": 162927820, "step": 7535, "time_per_iteration": 2.4916067123413086 }, { "auxiliary_loss_clip": 0.01104787, "auxiliary_loss_mlp": 0.01026725, "balance_loss_clip": 1.04290116, "balance_loss_mlp": 1.02011776, "epoch": 0.9061504238561895, "flos": 31759437502560.0, "grad_norm": 1.5485925637343028, "language_loss": 0.67774665, "learning_rate": 9.158613222874346e-08, "loss": 0.69906175, "num_input_tokens_seen": 162949445, "step": 7536, "time_per_iteration": 3.4416568279266357 }, { "auxiliary_loss_clip": 0.01138926, "auxiliary_loss_mlp": 0.01027732, "balance_loss_clip": 1.04412866, "balance_loss_mlp": 1.02131605, "epoch": 0.9062706667468285, "flos": 20048069573280.0, "grad_norm": 1.5990911591660188, "language_loss": 0.81764328, "learning_rate": 9.135324969566394e-08, "loss": 0.83930993, "num_input_tokens_seen": 162968945, "step": 7537, "time_per_iteration": 2.5205252170562744 }, { "auxiliary_loss_clip": 0.01161059, "auxiliary_loss_mlp": 0.01024245, "balance_loss_clip": 1.04826784, "balance_loss_mlp": 1.01691377, "epoch": 0.9063909096374677, "flos": 18437295418560.0, "grad_norm": 2.655037250517359, "language_loss": 0.75745809, "learning_rate": 9.112065670277913e-08, "loss": 0.77931118, "num_input_tokens_seen": 162985310, "step": 7538, "time_per_iteration": 2.4616124629974365 }, { "auxiliary_loss_clip": 0.01138033, "auxiliary_loss_mlp": 0.0102979, "balance_loss_clip": 1.04395366, "balance_loss_mlp": 1.02301884, "epoch": 0.9065111525281068, "flos": 33547358079360.0, "grad_norm": 2.1434949019855276, "language_loss": 0.72676402, "learning_rate": 9.088835328537303e-08, "loss": 0.74844217, "num_input_tokens_seen": 163006900, "step": 7539, "time_per_iteration": 2.6195120811462402 }, { "auxiliary_loss_clip": 0.01144257, "auxiliary_loss_mlp": 0.01033739, "balance_loss_clip": 1.04642701, "balance_loss_mlp": 1.02684236, "epoch": 0.9066313954187458, "flos": 23367867520320.0, "grad_norm": 1.9537726800057749, "language_loss": 0.71615499, "learning_rate": 9.065633947868568e-08, "loss": 0.73793495, "num_input_tokens_seen": 163026505, "step": 7540, "time_per_iteration": 2.5392096042633057 }, { "auxiliary_loss_clip": 0.01127174, "auxiliary_loss_mlp": 0.00761474, "balance_loss_clip": 1.04619002, "balance_loss_mlp": 1.00052953, "epoch": 0.906751638309385, "flos": 26249627049120.0, "grad_norm": 2.7164665771417287, "language_loss": 0.80116224, "learning_rate": 9.042461531791379e-08, "loss": 0.82004869, "num_input_tokens_seen": 163044925, "step": 7541, "time_per_iteration": 2.5764083862304688 }, { "auxiliary_loss_clip": 0.01164064, "auxiliary_loss_mlp": 0.01021084, "balance_loss_clip": 1.04608595, "balance_loss_mlp": 1.01431262, "epoch": 0.906871881200024, "flos": 16544480566560.0, "grad_norm": 1.808893221916467, "language_loss": 0.77830124, "learning_rate": 9.019318083820903e-08, "loss": 0.80015278, "num_input_tokens_seen": 163063505, "step": 7542, "time_per_iteration": 2.4403085708618164 }, { "auxiliary_loss_clip": 0.01153885, "auxiliary_loss_mlp": 0.01024874, "balance_loss_clip": 1.04785228, "balance_loss_mlp": 1.01771879, "epoch": 0.9069921240906631, "flos": 24605133257280.0, "grad_norm": 1.5768509025517419, "language_loss": 0.85298347, "learning_rate": 8.996203607468045e-08, "loss": 0.874771, "num_input_tokens_seen": 163082505, "step": 7543, "time_per_iteration": 3.2743701934814453 }, { "auxiliary_loss_clip": 0.01148997, "auxiliary_loss_mlp": 0.01024948, "balance_loss_clip": 1.04390407, "balance_loss_mlp": 1.01814663, "epoch": 0.9071123669813023, "flos": 25374735472320.0, "grad_norm": 1.4013151996125899, "language_loss": 0.75469929, "learning_rate": 8.973118106239241e-08, "loss": 0.77643871, "num_input_tokens_seen": 163105110, "step": 7544, "time_per_iteration": 2.5335168838500977 }, { "auxiliary_loss_clip": 0.01096214, "auxiliary_loss_mlp": 0.01025843, "balance_loss_clip": 1.03716481, "balance_loss_mlp": 1.01935506, "epoch": 0.9072326098719413, "flos": 26725802890560.0, "grad_norm": 1.8689879280369315, "language_loss": 0.94692719, "learning_rate": 8.95006158363656e-08, "loss": 0.96814775, "num_input_tokens_seen": 163125295, "step": 7545, "time_per_iteration": 2.665591239929199 }, { "auxiliary_loss_clip": 0.01158332, "auxiliary_loss_mlp": 0.01024728, "balance_loss_clip": 1.05068302, "balance_loss_mlp": 1.01744127, "epoch": 0.9073528527625804, "flos": 23878804590720.0, "grad_norm": 2.569717693796077, "language_loss": 0.772892, "learning_rate": 8.9270340431576e-08, "loss": 0.79472268, "num_input_tokens_seen": 163144385, "step": 7546, "time_per_iteration": 2.4916253089904785 }, { "auxiliary_loss_clip": 0.01153956, "auxiliary_loss_mlp": 0.01025609, "balance_loss_clip": 1.04477584, "balance_loss_mlp": 1.01891255, "epoch": 0.9074730956532195, "flos": 37852149837600.0, "grad_norm": 2.054503945717244, "language_loss": 0.7342726, "learning_rate": 8.904035488295658e-08, "loss": 0.75606823, "num_input_tokens_seen": 163163885, "step": 7547, "time_per_iteration": 2.6495275497436523 }, { "auxiliary_loss_clip": 0.01049839, "auxiliary_loss_mlp": 0.00752899, "balance_loss_clip": 1.00579345, "balance_loss_mlp": 1.0000695, "epoch": 0.9075933385438586, "flos": 65173317794880.0, "grad_norm": 0.6618798970534877, "language_loss": 0.53251022, "learning_rate": 8.881065922539632e-08, "loss": 0.55053759, "num_input_tokens_seen": 163224325, "step": 7548, "time_per_iteration": 3.022874593734741 }, { "auxiliary_loss_clip": 0.01119782, "auxiliary_loss_mlp": 0.01022079, "balance_loss_clip": 1.04440761, "balance_loss_mlp": 1.01532936, "epoch": 0.9077135814344977, "flos": 19931574120000.0, "grad_norm": 2.7235899870892113, "language_loss": 0.73426777, "learning_rate": 8.85812534937389e-08, "loss": 0.7556864, "num_input_tokens_seen": 163242425, "step": 7549, "time_per_iteration": 3.3150415420532227 }, { "auxiliary_loss_clip": 0.01157301, "auxiliary_loss_mlp": 0.01027058, "balance_loss_clip": 1.04781008, "balance_loss_mlp": 1.01993489, "epoch": 0.9078338243251368, "flos": 17529653962560.0, "grad_norm": 2.517225266774885, "language_loss": 0.66895235, "learning_rate": 8.835213772278583e-08, "loss": 0.69079596, "num_input_tokens_seen": 163259280, "step": 7550, "time_per_iteration": 3.4737534523010254 }, { "auxiliary_loss_clip": 0.01118885, "auxiliary_loss_mlp": 0.01028511, "balance_loss_clip": 1.04443097, "balance_loss_mlp": 1.02170682, "epoch": 0.9079540672157759, "flos": 28803414477120.0, "grad_norm": 10.334215670325907, "language_loss": 0.79169655, "learning_rate": 8.812331194729373e-08, "loss": 0.81317055, "num_input_tokens_seen": 163278925, "step": 7551, "time_per_iteration": 2.6034443378448486 }, { "auxiliary_loss_clip": 0.01172895, "auxiliary_loss_mlp": 0.01029815, "balance_loss_clip": 1.05085349, "balance_loss_mlp": 1.02253985, "epoch": 0.9080743101064149, "flos": 23513844409440.0, "grad_norm": 1.6819717603546909, "language_loss": 0.72197163, "learning_rate": 8.789477620197461e-08, "loss": 0.74399877, "num_input_tokens_seen": 163298450, "step": 7552, "time_per_iteration": 2.476863145828247 }, { "auxiliary_loss_clip": 0.01141349, "auxiliary_loss_mlp": 0.01029223, "balance_loss_clip": 1.04614973, "balance_loss_mlp": 1.02171624, "epoch": 0.9081945529970541, "flos": 22778105499360.0, "grad_norm": 2.2634472296188077, "language_loss": 0.79061085, "learning_rate": 8.766653052149831e-08, "loss": 0.81231654, "num_input_tokens_seen": 163313635, "step": 7553, "time_per_iteration": 2.4909842014312744 }, { "auxiliary_loss_clip": 0.01139211, "auxiliary_loss_mlp": 0.01023741, "balance_loss_clip": 1.04579186, "balance_loss_mlp": 1.01664495, "epoch": 0.9083147958876931, "flos": 18873717573600.0, "grad_norm": 2.445392686223333, "language_loss": 0.74891186, "learning_rate": 8.743857494048823e-08, "loss": 0.77054137, "num_input_tokens_seen": 163330450, "step": 7554, "time_per_iteration": 2.4974992275238037 }, { "auxiliary_loss_clip": 0.01126858, "auxiliary_loss_mlp": 0.01022274, "balance_loss_clip": 1.04365134, "balance_loss_mlp": 1.01570559, "epoch": 0.9084350387783322, "flos": 18909376726560.0, "grad_norm": 2.4334530955487628, "language_loss": 0.62663871, "learning_rate": 8.721090949352605e-08, "loss": 0.64813, "num_input_tokens_seen": 163346690, "step": 7555, "time_per_iteration": 2.5261337757110596 }, { "auxiliary_loss_clip": 0.01162121, "auxiliary_loss_mlp": 0.01027748, "balance_loss_clip": 1.05024397, "balance_loss_mlp": 1.02074194, "epoch": 0.9085552816689714, "flos": 20595491886720.0, "grad_norm": 1.8521718586322613, "language_loss": 0.72878182, "learning_rate": 8.698353421514793e-08, "loss": 0.75068051, "num_input_tokens_seen": 163365065, "step": 7556, "time_per_iteration": 2.492607593536377 }, { "auxiliary_loss_clip": 0.01153832, "auxiliary_loss_mlp": 0.01023662, "balance_loss_clip": 1.04739833, "balance_loss_mlp": 1.01686656, "epoch": 0.9086755245596104, "flos": 18113166432480.0, "grad_norm": 2.269145456851975, "language_loss": 0.80262887, "learning_rate": 8.67564491398467e-08, "loss": 0.82440388, "num_input_tokens_seen": 163382070, "step": 7557, "time_per_iteration": 2.441349983215332 }, { "auxiliary_loss_clip": 0.01159196, "auxiliary_loss_mlp": 0.01027618, "balance_loss_clip": 1.0473659, "balance_loss_mlp": 1.0201019, "epoch": 0.9087957674502495, "flos": 19129796697120.0, "grad_norm": 1.702800098791065, "language_loss": 0.73760772, "learning_rate": 8.652965430207104e-08, "loss": 0.75947583, "num_input_tokens_seen": 163399975, "step": 7558, "time_per_iteration": 2.476034164428711 }, { "auxiliary_loss_clip": 0.01156093, "auxiliary_loss_mlp": 0.01028483, "balance_loss_clip": 1.04559374, "balance_loss_mlp": 1.02097011, "epoch": 0.9089160103408886, "flos": 18109933906080.0, "grad_norm": 2.038094673018678, "language_loss": 0.65698773, "learning_rate": 8.630314973622521e-08, "loss": 0.67883348, "num_input_tokens_seen": 163417520, "step": 7559, "time_per_iteration": 2.4555466175079346 }, { "auxiliary_loss_clip": 0.01153104, "auxiliary_loss_mlp": 0.01022057, "balance_loss_clip": 1.04900098, "balance_loss_mlp": 1.01525009, "epoch": 0.9090362532315277, "flos": 33364860030240.0, "grad_norm": 1.8037440839286942, "language_loss": 0.7115885, "learning_rate": 8.607693547666995e-08, "loss": 0.73334014, "num_input_tokens_seen": 163440060, "step": 7560, "time_per_iteration": 2.5923705101013184 }, { "auxiliary_loss_clip": 0.01031123, "auxiliary_loss_mlp": 0.00999846, "balance_loss_clip": 1.00633836, "balance_loss_mlp": 0.99852878, "epoch": 0.9091564961221668, "flos": 71480595636000.0, "grad_norm": 0.8807711333605461, "language_loss": 0.57936573, "learning_rate": 8.585101155772201e-08, "loss": 0.59967542, "num_input_tokens_seen": 163502180, "step": 7561, "time_per_iteration": 3.2255661487579346 }, { "auxiliary_loss_clip": 0.011332, "auxiliary_loss_mlp": 0.0102701, "balance_loss_clip": 1.04226947, "balance_loss_mlp": 1.01966083, "epoch": 0.9092767390128058, "flos": 24712577636640.0, "grad_norm": 1.7554238376595535, "language_loss": 0.68613869, "learning_rate": 8.562537801365377e-08, "loss": 0.70774078, "num_input_tokens_seen": 163521915, "step": 7562, "time_per_iteration": 3.3047542572021484 }, { "auxiliary_loss_clip": 0.01170228, "auxiliary_loss_mlp": 0.0102952, "balance_loss_clip": 1.04824209, "balance_loss_mlp": 1.02143204, "epoch": 0.909396981903445, "flos": 23586491642880.0, "grad_norm": 1.7235481139033393, "language_loss": 0.70113629, "learning_rate": 8.540003487869362e-08, "loss": 0.7231338, "num_input_tokens_seen": 163543585, "step": 7563, "time_per_iteration": 2.466827630996704 }, { "auxiliary_loss_clip": 0.01112325, "auxiliary_loss_mlp": 0.01023123, "balance_loss_clip": 1.04085541, "balance_loss_mlp": 1.01637578, "epoch": 0.909517224794084, "flos": 23404173178560.0, "grad_norm": 4.323122556367772, "language_loss": 0.79728043, "learning_rate": 8.517498218702557e-08, "loss": 0.81863487, "num_input_tokens_seen": 163561515, "step": 7564, "time_per_iteration": 2.560333728790283 }, { "auxiliary_loss_clip": 0.01119705, "auxiliary_loss_mlp": 0.01022916, "balance_loss_clip": 1.04052615, "balance_loss_mlp": 1.01607645, "epoch": 0.9096374676847231, "flos": 19208621647680.0, "grad_norm": 1.8017751922702268, "language_loss": 0.69523865, "learning_rate": 8.49502199727905e-08, "loss": 0.71666485, "num_input_tokens_seen": 163579540, "step": 7565, "time_per_iteration": 2.5221335887908936 }, { "auxiliary_loss_clip": 0.01146207, "auxiliary_loss_mlp": 0.01032321, "balance_loss_clip": 1.04084206, "balance_loss_mlp": 1.02484393, "epoch": 0.9097577105753623, "flos": 33292500132480.0, "grad_norm": 2.44129017846117, "language_loss": 0.6620729, "learning_rate": 8.472574827008428e-08, "loss": 0.68385816, "num_input_tokens_seen": 163600425, "step": 7566, "time_per_iteration": 2.578636884689331 }, { "auxiliary_loss_clip": 0.0115565, "auxiliary_loss_mlp": 0.01020818, "balance_loss_clip": 1.04718578, "balance_loss_mlp": 1.01380849, "epoch": 0.9098779534660013, "flos": 21906446448960.0, "grad_norm": 1.7557584708165628, "language_loss": 0.83609819, "learning_rate": 8.450156711295942e-08, "loss": 0.85786283, "num_input_tokens_seen": 163620595, "step": 7567, "time_per_iteration": 2.490661144256592 }, { "auxiliary_loss_clip": 0.01135265, "auxiliary_loss_mlp": 0.01029773, "balance_loss_clip": 1.04491425, "balance_loss_mlp": 1.02297831, "epoch": 0.9099981963566404, "flos": 25730357244000.0, "grad_norm": 2.510006871436997, "language_loss": 0.86450589, "learning_rate": 8.427767653542383e-08, "loss": 0.88615632, "num_input_tokens_seen": 163635765, "step": 7568, "time_per_iteration": 2.508160352706909 }, { "auxiliary_loss_clip": 0.01109061, "auxiliary_loss_mlp": 0.01023807, "balance_loss_clip": 1.0407238, "balance_loss_mlp": 1.01725328, "epoch": 0.9101184392472795, "flos": 21069440876640.0, "grad_norm": 1.9594382389474714, "language_loss": 0.70125353, "learning_rate": 8.405407657144125e-08, "loss": 0.72258222, "num_input_tokens_seen": 163654925, "step": 7569, "time_per_iteration": 3.3472957611083984 }, { "auxiliary_loss_clip": 0.01132449, "auxiliary_loss_mlp": 0.01021259, "balance_loss_clip": 1.04325485, "balance_loss_mlp": 1.01420772, "epoch": 0.9102386821379186, "flos": 24752618658720.0, "grad_norm": 1.8420631729203871, "language_loss": 0.72774363, "learning_rate": 8.383076725493232e-08, "loss": 0.74928069, "num_input_tokens_seen": 163672245, "step": 7570, "time_per_iteration": 2.5260982513427734 }, { "auxiliary_loss_clip": 0.01154796, "auxiliary_loss_mlp": 0.01024179, "balance_loss_clip": 1.0464294, "balance_loss_mlp": 1.01729488, "epoch": 0.9103589250285576, "flos": 22562821654080.0, "grad_norm": 1.722346376390719, "language_loss": 0.6805774, "learning_rate": 8.360774861977216e-08, "loss": 0.70236707, "num_input_tokens_seen": 163691365, "step": 7571, "time_per_iteration": 2.499291181564331 }, { "auxiliary_loss_clip": 0.01139189, "auxiliary_loss_mlp": 0.01025105, "balance_loss_clip": 1.04204679, "balance_loss_mlp": 1.01855481, "epoch": 0.9104791679191968, "flos": 25373478378720.0, "grad_norm": 2.312358639189398, "language_loss": 0.74602431, "learning_rate": 8.338502069979281e-08, "loss": 0.76766729, "num_input_tokens_seen": 163711675, "step": 7572, "time_per_iteration": 2.5449178218841553 }, { "auxiliary_loss_clip": 0.01154338, "auxiliary_loss_mlp": 0.01027568, "balance_loss_clip": 1.04269171, "balance_loss_mlp": 1.02074003, "epoch": 0.9105994108098359, "flos": 14426684290080.0, "grad_norm": 3.158852843792431, "language_loss": 0.79841185, "learning_rate": 8.316258352878214e-08, "loss": 0.8202309, "num_input_tokens_seen": 163728095, "step": 7573, "time_per_iteration": 2.4628615379333496 }, { "auxiliary_loss_clip": 0.01157128, "auxiliary_loss_mlp": 0.01023776, "balance_loss_clip": 1.0445931, "balance_loss_mlp": 1.01635528, "epoch": 0.9107196537004749, "flos": 26718296245920.0, "grad_norm": 1.7406727712636447, "language_loss": 0.70976996, "learning_rate": 8.294043714048338e-08, "loss": 0.73157901, "num_input_tokens_seen": 163747175, "step": 7574, "time_per_iteration": 2.5303211212158203 }, { "auxiliary_loss_clip": 0.01041289, "auxiliary_loss_mlp": 0.01001996, "balance_loss_clip": 1.00736165, "balance_loss_mlp": 1.00070906, "epoch": 0.9108398965911141, "flos": 66532646113920.0, "grad_norm": 0.7525549996389429, "language_loss": 0.60532624, "learning_rate": 8.271858156859624e-08, "loss": 0.62575907, "num_input_tokens_seen": 163812545, "step": 7575, "time_per_iteration": 3.917018413543701 }, { "auxiliary_loss_clip": 0.01167681, "auxiliary_loss_mlp": 0.01025982, "balance_loss_clip": 1.04884887, "balance_loss_mlp": 1.01905596, "epoch": 0.9109601394817531, "flos": 25411077047520.0, "grad_norm": 1.6544634369313727, "language_loss": 0.73894131, "learning_rate": 8.249701684677557e-08, "loss": 0.76087797, "num_input_tokens_seen": 163833870, "step": 7576, "time_per_iteration": 3.31015682220459 }, { "auxiliary_loss_clip": 0.01157256, "auxiliary_loss_mlp": 0.01027826, "balance_loss_clip": 1.05057609, "balance_loss_mlp": 1.02118576, "epoch": 0.9110803823723922, "flos": 22747797973440.0, "grad_norm": 1.612657712241696, "language_loss": 0.80801433, "learning_rate": 8.227574300863294e-08, "loss": 0.8298651, "num_input_tokens_seen": 163854040, "step": 7577, "time_per_iteration": 2.4880080223083496 }, { "auxiliary_loss_clip": 0.01145803, "auxiliary_loss_mlp": 0.01027475, "balance_loss_clip": 1.04672766, "balance_loss_mlp": 1.0205462, "epoch": 0.9112006252630314, "flos": 48469930233600.0, "grad_norm": 2.7615955476209892, "language_loss": 0.69550443, "learning_rate": 8.205476008773548e-08, "loss": 0.71723723, "num_input_tokens_seen": 163878040, "step": 7578, "time_per_iteration": 2.765814781188965 }, { "auxiliary_loss_clip": 0.01117785, "auxiliary_loss_mlp": 0.01024661, "balance_loss_clip": 1.04319513, "balance_loss_mlp": 1.01830089, "epoch": 0.9113208681536704, "flos": 30009654348960.0, "grad_norm": 2.2417575832578445, "language_loss": 0.82633674, "learning_rate": 8.183406811760596e-08, "loss": 0.84776115, "num_input_tokens_seen": 163897770, "step": 7579, "time_per_iteration": 2.6045539379119873 }, { "auxiliary_loss_clip": 0.01110309, "auxiliary_loss_mlp": 0.01020443, "balance_loss_clip": 1.03919244, "balance_loss_mlp": 1.01342797, "epoch": 0.9114411110443095, "flos": 25594975858080.0, "grad_norm": 1.5522166791414878, "language_loss": 0.74062014, "learning_rate": 8.161366713172313e-08, "loss": 0.76192766, "num_input_tokens_seen": 163920160, "step": 7580, "time_per_iteration": 2.621108293533325 }, { "auxiliary_loss_clip": 0.01130694, "auxiliary_loss_mlp": 0.01023726, "balance_loss_clip": 1.04299569, "balance_loss_mlp": 1.01684487, "epoch": 0.9115613539349486, "flos": 18399732666720.0, "grad_norm": 4.530801430167912, "language_loss": 0.84117198, "learning_rate": 8.139355716352137e-08, "loss": 0.8627162, "num_input_tokens_seen": 163935000, "step": 7581, "time_per_iteration": 2.5043463706970215 }, { "auxiliary_loss_clip": 0.01136993, "auxiliary_loss_mlp": 0.01021222, "balance_loss_clip": 1.04269814, "balance_loss_mlp": 1.01420093, "epoch": 0.9116815968255877, "flos": 21726175251360.0, "grad_norm": 1.5126909728919744, "language_loss": 0.70009267, "learning_rate": 8.117373824639196e-08, "loss": 0.7216748, "num_input_tokens_seen": 163955265, "step": 7582, "time_per_iteration": 2.5305888652801514 }, { "auxiliary_loss_clip": 0.01059575, "auxiliary_loss_mlp": 0.01001186, "balance_loss_clip": 1.00560427, "balance_loss_mlp": 0.99992794, "epoch": 0.9118018397162267, "flos": 65363537990400.0, "grad_norm": 0.7267001929077292, "language_loss": 0.59330845, "learning_rate": 8.095421041368067e-08, "loss": 0.61391604, "num_input_tokens_seen": 164014680, "step": 7583, "time_per_iteration": 2.9828155040740967 }, { "auxiliary_loss_clip": 0.01138948, "auxiliary_loss_mlp": 0.00762089, "balance_loss_clip": 1.04684484, "balance_loss_mlp": 1.00060678, "epoch": 0.9119220826068659, "flos": 20922889316160.0, "grad_norm": 1.7652659946675269, "language_loss": 0.70447719, "learning_rate": 8.073497369868999e-08, "loss": 0.72348756, "num_input_tokens_seen": 164033140, "step": 7584, "time_per_iteration": 2.534954309463501 }, { "auxiliary_loss_clip": 0.01148172, "auxiliary_loss_mlp": 0.01025466, "balance_loss_clip": 1.04603028, "balance_loss_mlp": 1.01805401, "epoch": 0.912042325497505, "flos": 28366453567680.0, "grad_norm": 1.6566659914977442, "language_loss": 0.75379145, "learning_rate": 8.051602813467772e-08, "loss": 0.77552783, "num_input_tokens_seen": 164054995, "step": 7585, "time_per_iteration": 2.5825681686401367 }, { "auxiliary_loss_clip": 0.0115707, "auxiliary_loss_mlp": 0.01020616, "balance_loss_clip": 1.04704547, "balance_loss_mlp": 1.01388991, "epoch": 0.912162568388144, "flos": 17566785710880.0, "grad_norm": 1.7143963070782084, "language_loss": 0.71167886, "learning_rate": 8.029737375485756e-08, "loss": 0.73345572, "num_input_tokens_seen": 164074225, "step": 7586, "time_per_iteration": 2.492694854736328 }, { "auxiliary_loss_clip": 0.01167675, "auxiliary_loss_mlp": 0.01028107, "balance_loss_clip": 1.04671299, "balance_loss_mlp": 1.02086544, "epoch": 0.9122828112787832, "flos": 19827901021440.0, "grad_norm": 1.7413860568896928, "language_loss": 0.72424489, "learning_rate": 8.007901059239986e-08, "loss": 0.74620271, "num_input_tokens_seen": 164093505, "step": 7587, "time_per_iteration": 2.4442765712738037 }, { "auxiliary_loss_clip": 0.01139508, "auxiliary_loss_mlp": 0.01023156, "balance_loss_clip": 1.04406357, "balance_loss_mlp": 1.01641488, "epoch": 0.9124030541694222, "flos": 20813792756640.0, "grad_norm": 1.5831583817109425, "language_loss": 0.80066687, "learning_rate": 7.986093868042964e-08, "loss": 0.82229352, "num_input_tokens_seen": 164113750, "step": 7588, "time_per_iteration": 3.254568099975586 }, { "auxiliary_loss_clip": 0.01151425, "auxiliary_loss_mlp": 0.01019448, "balance_loss_clip": 1.0454607, "balance_loss_mlp": 1.01357365, "epoch": 0.9125232970600613, "flos": 25192309257120.0, "grad_norm": 1.5786246617261466, "language_loss": 0.67834663, "learning_rate": 7.964315805202826e-08, "loss": 0.7000553, "num_input_tokens_seen": 164134330, "step": 7589, "time_per_iteration": 2.522508144378662 }, { "auxiliary_loss_clip": 0.01142554, "auxiliary_loss_mlp": 0.01024648, "balance_loss_clip": 1.04613471, "balance_loss_mlp": 1.01767755, "epoch": 0.9126435399507005, "flos": 19719594635040.0, "grad_norm": 1.8006915114904358, "language_loss": 0.73215866, "learning_rate": 7.942566874023304e-08, "loss": 0.75383067, "num_input_tokens_seen": 164153515, "step": 7590, "time_per_iteration": 2.506655216217041 }, { "auxiliary_loss_clip": 0.01133344, "auxiliary_loss_mlp": 0.01025617, "balance_loss_clip": 1.04287291, "balance_loss_mlp": 1.01833379, "epoch": 0.9127637828413395, "flos": 19573617745920.0, "grad_norm": 2.253249223471491, "language_loss": 0.69551969, "learning_rate": 7.920847077803649e-08, "loss": 0.71710932, "num_input_tokens_seen": 164171305, "step": 7591, "time_per_iteration": 2.48409366607666 }, { "auxiliary_loss_clip": 0.01097555, "auxiliary_loss_mlp": 0.01027455, "balance_loss_clip": 1.03572869, "balance_loss_mlp": 1.02046371, "epoch": 0.9128840257319786, "flos": 20230639456320.0, "grad_norm": 3.284859940785787, "language_loss": 0.82068598, "learning_rate": 7.899156419838826e-08, "loss": 0.84193611, "num_input_tokens_seen": 164190275, "step": 7592, "time_per_iteration": 2.578444719314575 }, { "auxiliary_loss_clip": 0.01124484, "auxiliary_loss_mlp": 0.01021893, "balance_loss_clip": 1.04368162, "balance_loss_mlp": 1.01499069, "epoch": 0.9130042686226177, "flos": 24858662276640.0, "grad_norm": 1.6698747814469337, "language_loss": 0.65438443, "learning_rate": 7.87749490341918e-08, "loss": 0.67584819, "num_input_tokens_seen": 164210550, "step": 7593, "time_per_iteration": 2.572030544281006 }, { "auxiliary_loss_clip": 0.01171055, "auxiliary_loss_mlp": 0.01025582, "balance_loss_clip": 1.04879558, "balance_loss_mlp": 1.01829839, "epoch": 0.9131245115132568, "flos": 23581750604160.0, "grad_norm": 2.3219691000547193, "language_loss": 0.83843529, "learning_rate": 7.855862531830836e-08, "loss": 0.86040163, "num_input_tokens_seen": 164226660, "step": 7594, "time_per_iteration": 2.463949203491211 }, { "auxiliary_loss_clip": 0.01152105, "auxiliary_loss_mlp": 0.0102269, "balance_loss_clip": 1.04494011, "balance_loss_mlp": 1.01518583, "epoch": 0.9132447544038959, "flos": 19931610036960.0, "grad_norm": 1.6136974823116406, "language_loss": 0.72548908, "learning_rate": 7.834259308355373e-08, "loss": 0.74723697, "num_input_tokens_seen": 164245425, "step": 7595, "time_per_iteration": 3.2963478565216064 }, { "auxiliary_loss_clip": 0.0108189, "auxiliary_loss_mlp": 0.01028249, "balance_loss_clip": 1.03750014, "balance_loss_mlp": 1.02144492, "epoch": 0.9133649972945349, "flos": 21981751537440.0, "grad_norm": 1.8990111411618167, "language_loss": 0.75264871, "learning_rate": 7.812685236269989e-08, "loss": 0.77375007, "num_input_tokens_seen": 164264085, "step": 7596, "time_per_iteration": 2.6452009677886963 }, { "auxiliary_loss_clip": 0.01030098, "auxiliary_loss_mlp": 0.01000421, "balance_loss_clip": 1.01067853, "balance_loss_mlp": 0.99937755, "epoch": 0.9134852401851741, "flos": 71240535476640.0, "grad_norm": 0.7935692256166538, "language_loss": 0.58647758, "learning_rate": 7.791140318847445e-08, "loss": 0.60678279, "num_input_tokens_seen": 164322220, "step": 7597, "time_per_iteration": 3.150259256362915 }, { "auxiliary_loss_clip": 0.01137288, "auxiliary_loss_mlp": 0.0102406, "balance_loss_clip": 1.04717755, "balance_loss_mlp": 1.01711559, "epoch": 0.9136054830758131, "flos": 23626927751520.0, "grad_norm": 1.4572943463446475, "language_loss": 0.80220711, "learning_rate": 7.769624559356081e-08, "loss": 0.82382059, "num_input_tokens_seen": 164345615, "step": 7598, "time_per_iteration": 2.5835354328155518 }, { "auxiliary_loss_clip": 0.011529, "auxiliary_loss_mlp": 0.01025831, "balance_loss_clip": 1.04612875, "balance_loss_mlp": 1.01861, "epoch": 0.9137257259664522, "flos": 23438862573600.0, "grad_norm": 2.8175286241719633, "language_loss": 0.75206375, "learning_rate": 7.748137961059842e-08, "loss": 0.77385116, "num_input_tokens_seen": 164359595, "step": 7599, "time_per_iteration": 2.5168519020080566 }, { "auxiliary_loss_clip": 0.01163988, "auxiliary_loss_mlp": 0.01021026, "balance_loss_clip": 1.0467087, "balance_loss_mlp": 1.01440704, "epoch": 0.9138459688570914, "flos": 19127857181280.0, "grad_norm": 3.0382548903606366, "language_loss": 0.65075123, "learning_rate": 7.726680527218211e-08, "loss": 0.67260146, "num_input_tokens_seen": 164376635, "step": 7600, "time_per_iteration": 2.4288978576660156 }, { "auxiliary_loss_clip": 0.01166206, "auxiliary_loss_mlp": 0.0102667, "balance_loss_clip": 1.04480946, "balance_loss_mlp": 1.02006841, "epoch": 0.9139662117477304, "flos": 46281246654720.0, "grad_norm": 1.8922405542320606, "language_loss": 0.75575793, "learning_rate": 7.70525226108627e-08, "loss": 0.77768672, "num_input_tokens_seen": 164400305, "step": 7601, "time_per_iteration": 3.4487977027893066 }, { "auxiliary_loss_clip": 0.011557, "auxiliary_loss_mlp": 0.01026085, "balance_loss_clip": 1.04759824, "balance_loss_mlp": 1.01959968, "epoch": 0.9140864546383695, "flos": 22273202478240.0, "grad_norm": 1.9063396982243646, "language_loss": 0.80105776, "learning_rate": 7.683853165914666e-08, "loss": 0.82287562, "num_input_tokens_seen": 164418075, "step": 7602, "time_per_iteration": 3.2446272373199463 }, { "auxiliary_loss_clip": 0.01111828, "auxiliary_loss_mlp": 0.01029488, "balance_loss_clip": 1.0412637, "balance_loss_mlp": 1.02295494, "epoch": 0.9142066975290086, "flos": 17530013132160.0, "grad_norm": 1.8029855714030731, "language_loss": 0.77020597, "learning_rate": 7.662483244949602e-08, "loss": 0.79161918, "num_input_tokens_seen": 164435335, "step": 7603, "time_per_iteration": 2.556779146194458 }, { "auxiliary_loss_clip": 0.01120232, "auxiliary_loss_mlp": 0.01027175, "balance_loss_clip": 1.04641247, "balance_loss_mlp": 1.02076125, "epoch": 0.9143269404196477, "flos": 17712152011680.0, "grad_norm": 2.247453307043681, "language_loss": 0.80474651, "learning_rate": 7.641142501432951e-08, "loss": 0.82622057, "num_input_tokens_seen": 164451530, "step": 7604, "time_per_iteration": 2.5081403255462646 }, { "auxiliary_loss_clip": 0.01135316, "auxiliary_loss_mlp": 0.01027692, "balance_loss_clip": 1.04375386, "balance_loss_mlp": 1.02111411, "epoch": 0.9144471833102867, "flos": 33323418246720.0, "grad_norm": 1.600959821568801, "language_loss": 0.73639792, "learning_rate": 7.619830938602013e-08, "loss": 0.75802803, "num_input_tokens_seen": 164472755, "step": 7605, "time_per_iteration": 2.6113879680633545 }, { "auxiliary_loss_clip": 0.01150777, "auxiliary_loss_mlp": 0.01025362, "balance_loss_clip": 1.0461812, "balance_loss_mlp": 1.01805782, "epoch": 0.9145674262009259, "flos": 21068973956160.0, "grad_norm": 3.6259274805799935, "language_loss": 0.82608271, "learning_rate": 7.598548559689777e-08, "loss": 0.84784418, "num_input_tokens_seen": 164491155, "step": 7606, "time_per_iteration": 2.4942514896392822 }, { "auxiliary_loss_clip": 0.01119157, "auxiliary_loss_mlp": 0.01022792, "balance_loss_clip": 1.04093289, "balance_loss_mlp": 1.01592541, "epoch": 0.914687669091565, "flos": 16800272354400.0, "grad_norm": 2.2042693338393637, "language_loss": 0.81409061, "learning_rate": 7.577295367924751e-08, "loss": 0.83551013, "num_input_tokens_seen": 164507555, "step": 7607, "time_per_iteration": 2.5123343467712402 }, { "auxiliary_loss_clip": 0.01146498, "auxiliary_loss_mlp": 0.01026858, "balance_loss_clip": 1.0473994, "balance_loss_mlp": 1.01962805, "epoch": 0.914807911982204, "flos": 25773630792480.0, "grad_norm": 1.5716828740563376, "language_loss": 0.82084692, "learning_rate": 7.556071366531002e-08, "loss": 0.84258056, "num_input_tokens_seen": 164528525, "step": 7608, "time_per_iteration": 2.554624557495117 }, { "auxiliary_loss_clip": 0.01153034, "auxiliary_loss_mlp": 0.01030573, "balance_loss_clip": 1.04654741, "balance_loss_mlp": 1.02317882, "epoch": 0.9149281548728432, "flos": 19208046976320.0, "grad_norm": 2.517768335278035, "language_loss": 0.79263496, "learning_rate": 7.53487655872822e-08, "loss": 0.81447101, "num_input_tokens_seen": 164547695, "step": 7609, "time_per_iteration": 2.470973014831543 }, { "auxiliary_loss_clip": 0.01114716, "auxiliary_loss_mlp": 0.01024564, "balance_loss_clip": 1.0405097, "balance_loss_mlp": 1.01747394, "epoch": 0.9150483977634822, "flos": 26870558603040.0, "grad_norm": 2.3341827592286943, "language_loss": 0.74100977, "learning_rate": 7.513710947731656e-08, "loss": 0.76240259, "num_input_tokens_seen": 164568905, "step": 7610, "time_per_iteration": 2.628030776977539 }, { "auxiliary_loss_clip": 0.01131312, "auxiliary_loss_mlp": 0.01027, "balance_loss_clip": 1.0437336, "balance_loss_mlp": 1.02009177, "epoch": 0.9151686406541213, "flos": 21908960636160.0, "grad_norm": 1.6797798167040863, "language_loss": 0.85230196, "learning_rate": 7.492574536752095e-08, "loss": 0.87388504, "num_input_tokens_seen": 164588895, "step": 7611, "time_per_iteration": 2.511397123336792 }, { "auxiliary_loss_clip": 0.01151223, "auxiliary_loss_mlp": 0.0102326, "balance_loss_clip": 1.04721773, "balance_loss_mlp": 1.01695943, "epoch": 0.9152888835447605, "flos": 27308561104320.0, "grad_norm": 1.7404873747330352, "language_loss": 0.78064847, "learning_rate": 7.471467328995907e-08, "loss": 0.80239326, "num_input_tokens_seen": 164607705, "step": 7612, "time_per_iteration": 2.5462589263916016 }, { "auxiliary_loss_clip": 0.01076794, "auxiliary_loss_mlp": 0.01023269, "balance_loss_clip": 1.03900409, "balance_loss_mlp": 1.01704347, "epoch": 0.9154091264353995, "flos": 13370731342560.0, "grad_norm": 2.519198685475057, "language_loss": 0.60730815, "learning_rate": 7.450389327665018e-08, "loss": 0.62830877, "num_input_tokens_seen": 164625540, "step": 7613, "time_per_iteration": 3.4038338661193848 }, { "auxiliary_loss_clip": 0.01129803, "auxiliary_loss_mlp": 0.01036848, "balance_loss_clip": 1.04878807, "balance_loss_mlp": 1.02930498, "epoch": 0.9155293693260386, "flos": 20193040787520.0, "grad_norm": 2.8821018323074967, "language_loss": 0.67184782, "learning_rate": 7.429340535957029e-08, "loss": 0.69351435, "num_input_tokens_seen": 164640735, "step": 7614, "time_per_iteration": 3.105135917663574 }, { "auxiliary_loss_clip": 0.01141804, "auxiliary_loss_mlp": 0.01020599, "balance_loss_clip": 1.04511642, "balance_loss_mlp": 1.01400721, "epoch": 0.9156496122166777, "flos": 19354993623360.0, "grad_norm": 2.157607292373027, "language_loss": 0.70878601, "learning_rate": 7.40832095706494e-08, "loss": 0.73040998, "num_input_tokens_seen": 164657430, "step": 7615, "time_per_iteration": 2.50250244140625 }, { "auxiliary_loss_clip": 0.01132374, "auxiliary_loss_mlp": 0.01021978, "balance_loss_clip": 1.04469252, "balance_loss_mlp": 1.01577032, "epoch": 0.9157698551073168, "flos": 21107290964160.0, "grad_norm": 1.6128632431970695, "language_loss": 0.80260319, "learning_rate": 7.387330594177443e-08, "loss": 0.82414675, "num_input_tokens_seen": 164679505, "step": 7616, "time_per_iteration": 2.6099932193756104 }, { "auxiliary_loss_clip": 0.01118532, "auxiliary_loss_mlp": 0.01020789, "balance_loss_clip": 1.04239249, "balance_loss_mlp": 1.01365685, "epoch": 0.9158900979979558, "flos": 25193171264160.0, "grad_norm": 1.7409906322578141, "language_loss": 0.79251319, "learning_rate": 7.366369450478749e-08, "loss": 0.81390643, "num_input_tokens_seen": 164700615, "step": 7617, "time_per_iteration": 2.586257219314575 }, { "auxiliary_loss_clip": 0.01122856, "auxiliary_loss_mlp": 0.01025418, "balance_loss_clip": 1.04470384, "balance_loss_mlp": 1.01852465, "epoch": 0.916010340888595, "flos": 30146651998080.0, "grad_norm": 1.7463562659074152, "language_loss": 0.66532719, "learning_rate": 7.345437529148646e-08, "loss": 0.6868099, "num_input_tokens_seen": 164719625, "step": 7618, "time_per_iteration": 2.6107773780822754 }, { "auxiliary_loss_clip": 0.01127763, "auxiliary_loss_mlp": 0.01022816, "balance_loss_clip": 1.04329824, "balance_loss_mlp": 1.01614976, "epoch": 0.9161305837792341, "flos": 17091831046080.0, "grad_norm": 2.7241996955740957, "language_loss": 0.72505939, "learning_rate": 7.324534833362483e-08, "loss": 0.74656516, "num_input_tokens_seen": 164737200, "step": 7619, "time_per_iteration": 2.5418996810913086 }, { "auxiliary_loss_clip": 0.01139924, "auxiliary_loss_mlp": 0.01032372, "balance_loss_clip": 1.04551995, "balance_loss_mlp": 1.02529955, "epoch": 0.9162508266698731, "flos": 22893703028640.0, "grad_norm": 1.8307471445321524, "language_loss": 0.6861909, "learning_rate": 7.303661366291192e-08, "loss": 0.70791388, "num_input_tokens_seen": 164757870, "step": 7620, "time_per_iteration": 3.4379687309265137 }, { "auxiliary_loss_clip": 0.01107891, "auxiliary_loss_mlp": 0.0102551, "balance_loss_clip": 1.03978872, "balance_loss_mlp": 1.01896548, "epoch": 0.9163710695605123, "flos": 19974811751520.0, "grad_norm": 2.2776982348855253, "language_loss": 0.81629735, "learning_rate": 7.28281713110126e-08, "loss": 0.83763134, "num_input_tokens_seen": 164775945, "step": 7621, "time_per_iteration": 2.651475429534912 }, { "auxiliary_loss_clip": 0.01135154, "auxiliary_loss_mlp": 0.01025608, "balance_loss_clip": 1.04509878, "balance_loss_mlp": 1.01877737, "epoch": 0.9164913124511513, "flos": 22783816296000.0, "grad_norm": 2.3044399146928636, "language_loss": 0.77077997, "learning_rate": 7.262002130954759e-08, "loss": 0.7923876, "num_input_tokens_seen": 164794400, "step": 7622, "time_per_iteration": 2.526127338409424 }, { "auxiliary_loss_clip": 0.01113784, "auxiliary_loss_mlp": 0.01028611, "balance_loss_clip": 1.04199362, "balance_loss_mlp": 1.02175665, "epoch": 0.9166115553417904, "flos": 24900858316320.0, "grad_norm": 1.7511198464790698, "language_loss": 0.79093081, "learning_rate": 7.241216369009296e-08, "loss": 0.81235474, "num_input_tokens_seen": 164814585, "step": 7623, "time_per_iteration": 2.633225440979004 }, { "auxiliary_loss_clip": 0.01166198, "auxiliary_loss_mlp": 0.01029201, "balance_loss_clip": 1.04584503, "balance_loss_mlp": 1.02177989, "epoch": 0.9167317982324296, "flos": 25702923074880.0, "grad_norm": 1.8035738785359032, "language_loss": 0.66433072, "learning_rate": 7.220459848418037e-08, "loss": 0.68628478, "num_input_tokens_seen": 164834660, "step": 7624, "time_per_iteration": 2.4766573905944824 }, { "auxiliary_loss_clip": 0.01168482, "auxiliary_loss_mlp": 0.01026203, "balance_loss_clip": 1.04964304, "balance_loss_mlp": 1.01961708, "epoch": 0.9168520411230686, "flos": 15632816411040.0, "grad_norm": 1.720822725633037, "language_loss": 0.79493761, "learning_rate": 7.199732572329708e-08, "loss": 0.81688452, "num_input_tokens_seen": 164852560, "step": 7625, "time_per_iteration": 2.468226194381714 }, { "auxiliary_loss_clip": 0.01124102, "auxiliary_loss_mlp": 0.01028186, "balance_loss_clip": 1.0426265, "balance_loss_mlp": 1.02124238, "epoch": 0.9169722840137077, "flos": 30258154993920.0, "grad_norm": 2.063277191123224, "language_loss": 0.75722498, "learning_rate": 7.179034543888684e-08, "loss": 0.77874786, "num_input_tokens_seen": 164872065, "step": 7626, "time_per_iteration": 2.5938730239868164 }, { "auxiliary_loss_clip": 0.01153009, "auxiliary_loss_mlp": 0.01022687, "balance_loss_clip": 1.04402888, "balance_loss_mlp": 1.01586246, "epoch": 0.9170925269043467, "flos": 22491647016000.0, "grad_norm": 1.949411561603863, "language_loss": 0.77294999, "learning_rate": 7.158365766234808e-08, "loss": 0.794707, "num_input_tokens_seen": 164890915, "step": 7627, "time_per_iteration": 2.4917759895324707 }, { "auxiliary_loss_clip": 0.0111579, "auxiliary_loss_mlp": 0.01025856, "balance_loss_clip": 1.03885293, "balance_loss_mlp": 1.01836395, "epoch": 0.9172127697949859, "flos": 22893918530400.0, "grad_norm": 1.8390090410453575, "language_loss": 0.72349125, "learning_rate": 7.137726242503527e-08, "loss": 0.74490768, "num_input_tokens_seen": 164909835, "step": 7628, "time_per_iteration": 4.0361926555633545 }, { "auxiliary_loss_clip": 0.01152117, "auxiliary_loss_mlp": 0.00761981, "balance_loss_clip": 1.04643154, "balance_loss_mlp": 1.00051188, "epoch": 0.917333012685625, "flos": 17451870603840.0, "grad_norm": 2.369775004395654, "language_loss": 0.78669691, "learning_rate": 7.11711597582585e-08, "loss": 0.80583799, "num_input_tokens_seen": 164927195, "step": 7629, "time_per_iteration": 2.44877290725708 }, { "auxiliary_loss_clip": 0.01124815, "auxiliary_loss_mlp": 0.01019156, "balance_loss_clip": 1.03965032, "balance_loss_mlp": 1.0125699, "epoch": 0.917453255576264, "flos": 14318952575040.0, "grad_norm": 1.6689737318397695, "language_loss": 0.79683053, "learning_rate": 7.096534969328271e-08, "loss": 0.81827021, "num_input_tokens_seen": 164944640, "step": 7630, "time_per_iteration": 2.5192599296569824 }, { "auxiliary_loss_clip": 0.01142479, "auxiliary_loss_mlp": 0.01028085, "balance_loss_clip": 1.04281294, "balance_loss_mlp": 1.0214361, "epoch": 0.9175734984669032, "flos": 20741181440160.0, "grad_norm": 1.8672151605169853, "language_loss": 0.84036565, "learning_rate": 7.075983226132987e-08, "loss": 0.86207128, "num_input_tokens_seen": 164963570, "step": 7631, "time_per_iteration": 2.5151259899139404 }, { "auxiliary_loss_clip": 0.01144328, "auxiliary_loss_mlp": 0.00762514, "balance_loss_clip": 1.04448056, "balance_loss_mlp": 1.00055623, "epoch": 0.9176937413575422, "flos": 14830500233760.0, "grad_norm": 2.5226127881242575, "language_loss": 0.79331094, "learning_rate": 7.055460749357656e-08, "loss": 0.81237936, "num_input_tokens_seen": 164979850, "step": 7632, "time_per_iteration": 2.5008180141448975 }, { "auxiliary_loss_clip": 0.01139599, "auxiliary_loss_mlp": 0.01026866, "balance_loss_clip": 1.04561222, "balance_loss_mlp": 1.01896524, "epoch": 0.9178139842481813, "flos": 18474606751680.0, "grad_norm": 2.1961618892298067, "language_loss": 0.70291185, "learning_rate": 7.034967542115521e-08, "loss": 0.72457659, "num_input_tokens_seen": 164998115, "step": 7633, "time_per_iteration": 2.4849863052368164 }, { "auxiliary_loss_clip": 0.01144601, "auxiliary_loss_mlp": 0.00761425, "balance_loss_clip": 1.04491949, "balance_loss_mlp": 1.00048459, "epoch": 0.9179342271388204, "flos": 20047458984960.0, "grad_norm": 1.9189585645920897, "language_loss": 0.75202256, "learning_rate": 7.014503607515388e-08, "loss": 0.77108282, "num_input_tokens_seen": 165017420, "step": 7634, "time_per_iteration": 2.5046658515930176 }, { "auxiliary_loss_clip": 0.01142614, "auxiliary_loss_mlp": 0.01023547, "balance_loss_clip": 1.04902995, "balance_loss_mlp": 1.01701164, "epoch": 0.9180544700294595, "flos": 24676236061440.0, "grad_norm": 2.344538504432883, "language_loss": 0.68594694, "learning_rate": 6.994068948661592e-08, "loss": 0.70760858, "num_input_tokens_seen": 165035575, "step": 7635, "time_per_iteration": 2.5463123321533203 }, { "auxiliary_loss_clip": 0.01157655, "auxiliary_loss_mlp": 0.0102507, "balance_loss_clip": 1.04789495, "balance_loss_mlp": 1.01742291, "epoch": 0.9181747129200986, "flos": 16727481453120.0, "grad_norm": 1.9286415827317231, "language_loss": 0.76172107, "learning_rate": 6.973663568654142e-08, "loss": 0.7835483, "num_input_tokens_seen": 165053280, "step": 7636, "time_per_iteration": 2.4686200618743896 }, { "auxiliary_loss_clip": 0.01165388, "auxiliary_loss_mlp": 0.01025401, "balance_loss_clip": 1.04724646, "balance_loss_mlp": 1.0184691, "epoch": 0.9182949558107377, "flos": 24271629944640.0, "grad_norm": 2.29227450289815, "language_loss": 0.65241534, "learning_rate": 6.953287470588386e-08, "loss": 0.6743232, "num_input_tokens_seen": 165071235, "step": 7637, "time_per_iteration": 2.4650919437408447 }, { "auxiliary_loss_clip": 0.0115575, "auxiliary_loss_mlp": 0.01023573, "balance_loss_clip": 1.04394341, "balance_loss_mlp": 1.01711226, "epoch": 0.9184151987013768, "flos": 22082120275680.0, "grad_norm": 2.524002381101356, "language_loss": 0.85709596, "learning_rate": 6.932940657555452e-08, "loss": 0.87888914, "num_input_tokens_seen": 165087365, "step": 7638, "time_per_iteration": 2.483574867248535 }, { "auxiliary_loss_clip": 0.01161595, "auxiliary_loss_mlp": 0.01027007, "balance_loss_clip": 1.0459516, "balance_loss_mlp": 1.02064967, "epoch": 0.9185354415920158, "flos": 32166737391360.0, "grad_norm": 1.3683675179220167, "language_loss": 0.7649163, "learning_rate": 6.912623132641938e-08, "loss": 0.78680229, "num_input_tokens_seen": 165112455, "step": 7639, "time_per_iteration": 2.575813055038452 }, { "auxiliary_loss_clip": 0.01142334, "auxiliary_loss_mlp": 0.0102836, "balance_loss_clip": 1.04639494, "balance_loss_mlp": 1.02103508, "epoch": 0.918655684482655, "flos": 20997835235040.0, "grad_norm": 1.7391609400285897, "language_loss": 0.765082, "learning_rate": 6.892334898929952e-08, "loss": 0.78678894, "num_input_tokens_seen": 165132700, "step": 7640, "time_per_iteration": 3.492267370223999 }, { "auxiliary_loss_clip": 0.01147082, "auxiliary_loss_mlp": 0.01022395, "balance_loss_clip": 1.04466534, "balance_loss_mlp": 1.01557338, "epoch": 0.918775927373294, "flos": 15560707932000.0, "grad_norm": 1.9605030584254988, "language_loss": 0.84932095, "learning_rate": 6.872075959497236e-08, "loss": 0.87101567, "num_input_tokens_seen": 165151475, "step": 7641, "time_per_iteration": 2.4668121337890625 }, { "auxiliary_loss_clip": 0.01150918, "auxiliary_loss_mlp": 0.01023966, "balance_loss_clip": 1.04477882, "balance_loss_mlp": 1.01711488, "epoch": 0.9188961702639331, "flos": 29934061924800.0, "grad_norm": 2.331502705542635, "language_loss": 0.82681423, "learning_rate": 6.85184631741702e-08, "loss": 0.84856308, "num_input_tokens_seen": 165172040, "step": 7642, "time_per_iteration": 2.5423717498779297 }, { "auxiliary_loss_clip": 0.01152313, "auxiliary_loss_mlp": 0.01021287, "balance_loss_clip": 1.0463022, "balance_loss_mlp": 1.01443839, "epoch": 0.9190164131545723, "flos": 20701248168960.0, "grad_norm": 1.930165494195383, "language_loss": 0.77501243, "learning_rate": 6.831645975758161e-08, "loss": 0.79674846, "num_input_tokens_seen": 165189980, "step": 7643, "time_per_iteration": 2.464858055114746 }, { "auxiliary_loss_clip": 0.01132275, "auxiliary_loss_mlp": 0.01026023, "balance_loss_clip": 1.04317355, "balance_loss_mlp": 1.01746929, "epoch": 0.9191366560452113, "flos": 25629916671840.0, "grad_norm": 2.095917332546757, "language_loss": 0.67096949, "learning_rate": 6.811474937585026e-08, "loss": 0.69255245, "num_input_tokens_seen": 165209770, "step": 7644, "time_per_iteration": 2.5693840980529785 }, { "auxiliary_loss_clip": 0.01122289, "auxiliary_loss_mlp": 0.01021808, "balance_loss_clip": 1.0424993, "balance_loss_mlp": 1.01490545, "epoch": 0.9192568989358504, "flos": 21434329224000.0, "grad_norm": 1.630058656073105, "language_loss": 0.79185033, "learning_rate": 6.79133320595755e-08, "loss": 0.81329131, "num_input_tokens_seen": 165229690, "step": 7645, "time_per_iteration": 2.5379672050476074 }, { "auxiliary_loss_clip": 0.01143998, "auxiliary_loss_mlp": 0.0102197, "balance_loss_clip": 1.04798698, "balance_loss_mlp": 1.01502907, "epoch": 0.9193771418264896, "flos": 23185082135520.0, "grad_norm": 2.0723978833193324, "language_loss": 0.75568765, "learning_rate": 6.771220783931198e-08, "loss": 0.77734733, "num_input_tokens_seen": 165249850, "step": 7646, "time_per_iteration": 2.522090435028076 }, { "auxiliary_loss_clip": 0.01013487, "auxiliary_loss_mlp": 0.00753093, "balance_loss_clip": 1.03540003, "balance_loss_mlp": 1.00020111, "epoch": 0.9194973847171286, "flos": 70582975011840.0, "grad_norm": 0.8566168844598004, "language_loss": 0.64605045, "learning_rate": 6.751137674556994e-08, "loss": 0.6637162, "num_input_tokens_seen": 165310235, "step": 7647, "time_per_iteration": 4.319162368774414 }, { "auxiliary_loss_clip": 0.01157439, "auxiliary_loss_mlp": 0.01023245, "balance_loss_clip": 1.04539263, "balance_loss_mlp": 1.01677823, "epoch": 0.9196176276077677, "flos": 14720685335040.0, "grad_norm": 2.033157578355754, "language_loss": 0.77747315, "learning_rate": 6.731083880881572e-08, "loss": 0.79927999, "num_input_tokens_seen": 165326455, "step": 7648, "time_per_iteration": 2.6360599994659424 }, { "auxiliary_loss_clip": 0.01139081, "auxiliary_loss_mlp": 0.01021983, "balance_loss_clip": 1.04425335, "balance_loss_mlp": 1.01540828, "epoch": 0.9197378704984068, "flos": 23294897034240.0, "grad_norm": 2.403032572780323, "language_loss": 0.81002402, "learning_rate": 6.711059405947072e-08, "loss": 0.83163464, "num_input_tokens_seen": 165344645, "step": 7649, "time_per_iteration": 2.535015344619751 }, { "auxiliary_loss_clip": 0.0112468, "auxiliary_loss_mlp": 0.01022214, "balance_loss_clip": 1.04545283, "balance_loss_mlp": 1.01569009, "epoch": 0.9198581133890459, "flos": 20302568350560.0, "grad_norm": 1.7100408055184975, "language_loss": 0.7666474, "learning_rate": 6.691064252791156e-08, "loss": 0.78811634, "num_input_tokens_seen": 165364120, "step": 7650, "time_per_iteration": 2.540363311767578 }, { "auxiliary_loss_clip": 0.01104835, "auxiliary_loss_mlp": 0.0102242, "balance_loss_clip": 1.04109764, "balance_loss_mlp": 1.01516557, "epoch": 0.9199783562796849, "flos": 17675666768640.0, "grad_norm": 1.625126856225058, "language_loss": 0.77743137, "learning_rate": 6.67109842444713e-08, "loss": 0.79870391, "num_input_tokens_seen": 165383050, "step": 7651, "time_per_iteration": 2.5569076538085938 }, { "auxiliary_loss_clip": 0.0115506, "auxiliary_loss_mlp": 0.0076185, "balance_loss_clip": 1.05002499, "balance_loss_mlp": 1.00053513, "epoch": 0.9200985991703241, "flos": 17676025938240.0, "grad_norm": 1.881363238643454, "language_loss": 0.76446438, "learning_rate": 6.651161923943704e-08, "loss": 0.78363347, "num_input_tokens_seen": 165400955, "step": 7652, "time_per_iteration": 2.4635019302368164 }, { "auxiliary_loss_clip": 0.01148412, "auxiliary_loss_mlp": 0.0102845, "balance_loss_clip": 1.04360652, "balance_loss_mlp": 1.02150273, "epoch": 0.9202188420609632, "flos": 20996578141440.0, "grad_norm": 1.7555543916833947, "language_loss": 0.76746845, "learning_rate": 6.631254754305326e-08, "loss": 0.78923714, "num_input_tokens_seen": 165420415, "step": 7653, "time_per_iteration": 2.49660062789917 }, { "auxiliary_loss_clip": 0.01169733, "auxiliary_loss_mlp": 0.01028543, "balance_loss_clip": 1.04712558, "balance_loss_mlp": 1.0216856, "epoch": 0.9203390849516022, "flos": 13918225489920.0, "grad_norm": 2.242726564006129, "language_loss": 0.78052104, "learning_rate": 6.611376918551848e-08, "loss": 0.80250382, "num_input_tokens_seen": 165439200, "step": 7654, "time_per_iteration": 4.2878358364105225 }, { "auxiliary_loss_clip": 0.01116177, "auxiliary_loss_mlp": 0.00761786, "balance_loss_clip": 1.0400331, "balance_loss_mlp": 1.00054491, "epoch": 0.9204593278422414, "flos": 21175915498080.0, "grad_norm": 2.0246209360987057, "language_loss": 0.79300338, "learning_rate": 6.591528419698744e-08, "loss": 0.81178296, "num_input_tokens_seen": 165458985, "step": 7655, "time_per_iteration": 2.554116725921631 }, { "auxiliary_loss_clip": 0.01140754, "auxiliary_loss_mlp": 0.01021998, "balance_loss_clip": 1.04313302, "balance_loss_mlp": 1.01545358, "epoch": 0.9205795707328804, "flos": 14501378790240.0, "grad_norm": 3.340098860511029, "language_loss": 0.83014828, "learning_rate": 6.571709260756986e-08, "loss": 0.85177577, "num_input_tokens_seen": 165475630, "step": 7656, "time_per_iteration": 2.5247962474823 }, { "auxiliary_loss_clip": 0.01155748, "auxiliary_loss_mlp": 0.01027523, "balance_loss_clip": 1.04861104, "balance_loss_mlp": 1.02017379, "epoch": 0.9206998136235195, "flos": 22417563104160.0, "grad_norm": 2.409659127741345, "language_loss": 0.76315659, "learning_rate": 6.551919444733122e-08, "loss": 0.7849893, "num_input_tokens_seen": 165493445, "step": 7657, "time_per_iteration": 2.4752111434936523 }, { "auxiliary_loss_clip": 0.01138534, "auxiliary_loss_mlp": 0.0102508, "balance_loss_clip": 1.04549229, "balance_loss_mlp": 1.01814818, "epoch": 0.9208200565141585, "flos": 53358414046560.0, "grad_norm": 1.733138281056469, "language_loss": 0.65892798, "learning_rate": 6.53215897462931e-08, "loss": 0.68056417, "num_input_tokens_seen": 165517200, "step": 7658, "time_per_iteration": 2.814298629760742 }, { "auxiliary_loss_clip": 0.01150808, "auxiliary_loss_mlp": 0.01020505, "balance_loss_clip": 1.04426014, "balance_loss_mlp": 1.01387739, "epoch": 0.9209402994047977, "flos": 30589144119360.0, "grad_norm": 2.019614535672254, "language_loss": 0.75314611, "learning_rate": 6.512427853443103e-08, "loss": 0.77485925, "num_input_tokens_seen": 165539280, "step": 7659, "time_per_iteration": 2.5434014797210693 }, { "auxiliary_loss_clip": 0.01156267, "auxiliary_loss_mlp": 0.01026311, "balance_loss_clip": 1.04624391, "balance_loss_mlp": 1.01975775, "epoch": 0.9210605422954368, "flos": 29132715505440.0, "grad_norm": 1.5414055581943613, "language_loss": 0.75633568, "learning_rate": 6.492726084167799e-08, "loss": 0.77816153, "num_input_tokens_seen": 165561395, "step": 7660, "time_per_iteration": 2.566406011581421 }, { "auxiliary_loss_clip": 0.01059091, "auxiliary_loss_mlp": 0.01000764, "balance_loss_clip": 1.0051465, "balance_loss_mlp": 0.99953616, "epoch": 0.9211807851860758, "flos": 54853848479040.0, "grad_norm": 0.7762248051866697, "language_loss": 0.57529283, "learning_rate": 6.473053669792072e-08, "loss": 0.59589136, "num_input_tokens_seen": 165616085, "step": 7661, "time_per_iteration": 2.9340391159057617 }, { "auxiliary_loss_clip": 0.01150866, "auxiliary_loss_mlp": 0.01028896, "balance_loss_clip": 1.04423916, "balance_loss_mlp": 1.02177, "epoch": 0.921301028076715, "flos": 19201977010080.0, "grad_norm": 3.025067356833053, "language_loss": 0.72460961, "learning_rate": 6.453410613300248e-08, "loss": 0.74640727, "num_input_tokens_seen": 165634015, "step": 7662, "time_per_iteration": 2.4886319637298584 }, { "auxiliary_loss_clip": 0.01097541, "auxiliary_loss_mlp": 0.01030191, "balance_loss_clip": 1.03983808, "balance_loss_mlp": 1.02283013, "epoch": 0.921421270967354, "flos": 27526897891200.0, "grad_norm": 1.5748841282617319, "language_loss": 0.58065939, "learning_rate": 6.43379691767214e-08, "loss": 0.60193664, "num_input_tokens_seen": 165653220, "step": 7663, "time_per_iteration": 2.671966791152954 }, { "auxiliary_loss_clip": 0.0102042, "auxiliary_loss_mlp": 0.01000348, "balance_loss_clip": 1.00566959, "balance_loss_mlp": 0.99911398, "epoch": 0.9215415138579931, "flos": 70209322926240.0, "grad_norm": 0.7225306096489736, "language_loss": 0.55182099, "learning_rate": 6.414212585883105e-08, "loss": 0.5720287, "num_input_tokens_seen": 165715850, "step": 7664, "time_per_iteration": 3.218163251876831 }, { "auxiliary_loss_clip": 0.01145007, "auxiliary_loss_mlp": 0.0102285, "balance_loss_clip": 1.04661179, "balance_loss_mlp": 1.01610255, "epoch": 0.9216617567486323, "flos": 35553112605600.0, "grad_norm": 1.6126753096630093, "language_loss": 0.69884419, "learning_rate": 6.394657620904143e-08, "loss": 0.72052276, "num_input_tokens_seen": 165738960, "step": 7665, "time_per_iteration": 2.654012441635132 }, { "auxiliary_loss_clip": 0.01174498, "auxiliary_loss_mlp": 0.01028066, "balance_loss_clip": 1.05048609, "balance_loss_mlp": 1.02084494, "epoch": 0.9217819996392713, "flos": 29533334839680.0, "grad_norm": 1.9749033251347476, "language_loss": 0.71580064, "learning_rate": 6.375132025701657e-08, "loss": 0.73782635, "num_input_tokens_seen": 165761260, "step": 7666, "time_per_iteration": 3.268242120742798 }, { "auxiliary_loss_clip": 0.01173991, "auxiliary_loss_mlp": 0.01032401, "balance_loss_clip": 1.05131412, "balance_loss_mlp": 1.02554035, "epoch": 0.9219022425299104, "flos": 14574672528960.0, "grad_norm": 2.329506261964218, "language_loss": 0.69528615, "learning_rate": 6.355635803237724e-08, "loss": 0.71735007, "num_input_tokens_seen": 165776960, "step": 7667, "time_per_iteration": 2.4385318756103516 }, { "auxiliary_loss_clip": 0.0115279, "auxiliary_loss_mlp": 0.01030645, "balance_loss_clip": 1.04490232, "balance_loss_mlp": 1.02349222, "epoch": 0.9220224854205495, "flos": 18077507279520.0, "grad_norm": 1.944892266556806, "language_loss": 0.79433227, "learning_rate": 6.336168956469867e-08, "loss": 0.81616664, "num_input_tokens_seen": 165795435, "step": 7668, "time_per_iteration": 2.4517078399658203 }, { "auxiliary_loss_clip": 0.01133439, "auxiliary_loss_mlp": 0.01025581, "balance_loss_clip": 1.04521894, "balance_loss_mlp": 1.0186398, "epoch": 0.9221427283111886, "flos": 24790468746240.0, "grad_norm": 1.5661506287102365, "language_loss": 0.71879888, "learning_rate": 6.316731488351168e-08, "loss": 0.74038911, "num_input_tokens_seen": 165816625, "step": 7669, "time_per_iteration": 2.5567781925201416 }, { "auxiliary_loss_clip": 0.01152826, "auxiliary_loss_mlp": 0.01029819, "balance_loss_clip": 1.04648471, "balance_loss_mlp": 1.02259493, "epoch": 0.9222629712018277, "flos": 13845039502080.0, "grad_norm": 1.6877922006529864, "language_loss": 0.63447118, "learning_rate": 6.297323401830334e-08, "loss": 0.65629762, "num_input_tokens_seen": 165835410, "step": 7670, "time_per_iteration": 2.4495179653167725 }, { "auxiliary_loss_clip": 0.01157458, "auxiliary_loss_mlp": 0.01027767, "balance_loss_clip": 1.04610276, "balance_loss_mlp": 1.02036417, "epoch": 0.9223832140924668, "flos": 21616180767840.0, "grad_norm": 1.9698741897933854, "language_loss": 0.68973196, "learning_rate": 6.277944699851523e-08, "loss": 0.71158421, "num_input_tokens_seen": 165854930, "step": 7671, "time_per_iteration": 2.485123634338379 }, { "auxiliary_loss_clip": 0.01167776, "auxiliary_loss_mlp": 0.01030088, "balance_loss_clip": 1.04655993, "balance_loss_mlp": 1.02303672, "epoch": 0.9225034569831059, "flos": 21142088110080.0, "grad_norm": 1.7769898971807032, "language_loss": 0.73122168, "learning_rate": 6.25859538535447e-08, "loss": 0.75320035, "num_input_tokens_seen": 165875725, "step": 7672, "time_per_iteration": 3.198199510574341 }, { "auxiliary_loss_clip": 0.01137647, "auxiliary_loss_mlp": 0.0103193, "balance_loss_clip": 1.04502439, "balance_loss_mlp": 1.02462292, "epoch": 0.9226236998737449, "flos": 12495049592640.0, "grad_norm": 2.5174125352852683, "language_loss": 0.78006566, "learning_rate": 6.239275461274474e-08, "loss": 0.80176139, "num_input_tokens_seen": 165892100, "step": 7673, "time_per_iteration": 2.4939568042755127 }, { "auxiliary_loss_clip": 0.01153246, "auxiliary_loss_mlp": 0.01024427, "balance_loss_clip": 1.04611921, "balance_loss_mlp": 1.01707125, "epoch": 0.9227439427643841, "flos": 26214075647040.0, "grad_norm": 1.7389701115087246, "language_loss": 0.85926163, "learning_rate": 6.219984930542299e-08, "loss": 0.88103843, "num_input_tokens_seen": 165912840, "step": 7674, "time_per_iteration": 2.5299742221832275 }, { "auxiliary_loss_clip": 0.01157035, "auxiliary_loss_mlp": 0.0102622, "balance_loss_clip": 1.04596615, "balance_loss_mlp": 1.01871872, "epoch": 0.9228641856550232, "flos": 17967584629920.0, "grad_norm": 2.320900188673536, "language_loss": 0.760324, "learning_rate": 6.200723796084383e-08, "loss": 0.78215653, "num_input_tokens_seen": 165930935, "step": 7675, "time_per_iteration": 2.4851040840148926 }, { "auxiliary_loss_clip": 0.01032796, "auxiliary_loss_mlp": 0.01002431, "balance_loss_clip": 1.00562453, "balance_loss_mlp": 1.00122666, "epoch": 0.9229844285456622, "flos": 70420619988960.0, "grad_norm": 0.7618645684804656, "language_loss": 0.63037902, "learning_rate": 6.181492060822546e-08, "loss": 0.65073127, "num_input_tokens_seen": 165991110, "step": 7676, "time_per_iteration": 3.062164783477783 }, { "auxiliary_loss_clip": 0.01108872, "auxiliary_loss_mlp": 0.01030772, "balance_loss_clip": 1.04058385, "balance_loss_mlp": 1.02394485, "epoch": 0.9231046714363014, "flos": 17967836048640.0, "grad_norm": 2.0900292177984645, "language_loss": 0.81651056, "learning_rate": 6.162289727674274e-08, "loss": 0.83790702, "num_input_tokens_seen": 166008790, "step": 7677, "time_per_iteration": 2.5641205310821533 }, { "auxiliary_loss_clip": 0.01124941, "auxiliary_loss_mlp": 0.01022464, "balance_loss_clip": 1.04299712, "balance_loss_mlp": 1.01617551, "epoch": 0.9232249143269404, "flos": 17858236651680.0, "grad_norm": 2.440104116144501, "language_loss": 0.88042963, "learning_rate": 6.143116799552527e-08, "loss": 0.90190363, "num_input_tokens_seen": 166025035, "step": 7678, "time_per_iteration": 2.521756172180176 }, { "auxiliary_loss_clip": 0.01155227, "auxiliary_loss_mlp": 0.01032001, "balance_loss_clip": 1.04600346, "balance_loss_mlp": 1.02477407, "epoch": 0.9233451572175795, "flos": 23404388680320.0, "grad_norm": 2.347766539206291, "language_loss": 0.55925322, "learning_rate": 6.123973279365802e-08, "loss": 0.5811255, "num_input_tokens_seen": 166044010, "step": 7679, "time_per_iteration": 2.510371685028076 }, { "auxiliary_loss_clip": 0.01158876, "auxiliary_loss_mlp": 0.01027499, "balance_loss_clip": 1.04959631, "balance_loss_mlp": 1.02072501, "epoch": 0.9234654001082186, "flos": 17999328834240.0, "grad_norm": 1.8817461579940367, "language_loss": 0.77468061, "learning_rate": 6.10485917001824e-08, "loss": 0.79654443, "num_input_tokens_seen": 166061865, "step": 7680, "time_per_iteration": 3.261979818344116 }, { "auxiliary_loss_clip": 0.01139033, "auxiliary_loss_mlp": 0.01027704, "balance_loss_clip": 1.04322028, "balance_loss_mlp": 1.02091825, "epoch": 0.9235856429988577, "flos": 24750750976800.0, "grad_norm": 1.7405524805832193, "language_loss": 0.80917454, "learning_rate": 6.085774474409322e-08, "loss": 0.8308419, "num_input_tokens_seen": 166082425, "step": 7681, "time_per_iteration": 2.551063299179077 }, { "auxiliary_loss_clip": 0.01140848, "auxiliary_loss_mlp": 0.01027874, "balance_loss_clip": 1.0480243, "balance_loss_mlp": 1.02152359, "epoch": 0.9237058858894968, "flos": 14099897448960.0, "grad_norm": 1.8867813267990399, "language_loss": 0.70138502, "learning_rate": 6.066719195434267e-08, "loss": 0.72307229, "num_input_tokens_seen": 166100225, "step": 7682, "time_per_iteration": 2.4928176403045654 }, { "auxiliary_loss_clip": 0.01156338, "auxiliary_loss_mlp": 0.01026826, "balance_loss_clip": 1.04627573, "balance_loss_mlp": 1.01922703, "epoch": 0.9238261287801359, "flos": 28694533419360.0, "grad_norm": 2.4012301103145934, "language_loss": 0.66772735, "learning_rate": 6.047693335983717e-08, "loss": 0.68955904, "num_input_tokens_seen": 166122570, "step": 7683, "time_per_iteration": 2.53010892868042 }, { "auxiliary_loss_clip": 0.01157778, "auxiliary_loss_mlp": 0.01022161, "balance_loss_clip": 1.04635239, "balance_loss_mlp": 1.01550651, "epoch": 0.923946371670775, "flos": 23111860230720.0, "grad_norm": 5.10404428641318, "language_loss": 0.82087201, "learning_rate": 6.028696898943853e-08, "loss": 0.84267133, "num_input_tokens_seen": 166141630, "step": 7684, "time_per_iteration": 2.494941234588623 }, { "auxiliary_loss_clip": 0.01139753, "auxiliary_loss_mlp": 0.00762422, "balance_loss_clip": 1.04278851, "balance_loss_mlp": 1.00057733, "epoch": 0.924066614561414, "flos": 21867123766080.0, "grad_norm": 1.9068665246591807, "language_loss": 0.70480245, "learning_rate": 6.00972988719648e-08, "loss": 0.7238242, "num_input_tokens_seen": 166159865, "step": 7685, "time_per_iteration": 2.512624740600586 }, { "auxiliary_loss_clip": 0.01129077, "auxiliary_loss_mlp": 0.00762149, "balance_loss_clip": 1.04340672, "balance_loss_mlp": 1.00055814, "epoch": 0.9241868574520532, "flos": 28511891702400.0, "grad_norm": 2.239872089776826, "language_loss": 0.70655847, "learning_rate": 5.990792303618807e-08, "loss": 0.72547078, "num_input_tokens_seen": 166179445, "step": 7686, "time_per_iteration": 2.6010992527008057 }, { "auxiliary_loss_clip": 0.01127406, "auxiliary_loss_mlp": 0.01025765, "balance_loss_clip": 1.04553008, "balance_loss_mlp": 1.01882362, "epoch": 0.9243071003426923, "flos": 30518328650880.0, "grad_norm": 1.6808536861615078, "language_loss": 0.69451195, "learning_rate": 5.971884151083695e-08, "loss": 0.71604365, "num_input_tokens_seen": 166201855, "step": 7687, "time_per_iteration": 2.6236133575439453 }, { "auxiliary_loss_clip": 0.01141821, "auxiliary_loss_mlp": 0.01025256, "balance_loss_clip": 1.04444516, "balance_loss_mlp": 1.01889622, "epoch": 0.9244273432333313, "flos": 28658335512000.0, "grad_norm": 1.6600867457561117, "language_loss": 0.74094349, "learning_rate": 5.9530054324595124e-08, "loss": 0.76261425, "num_input_tokens_seen": 166221970, "step": 7688, "time_per_iteration": 2.594221591949463 }, { "auxiliary_loss_clip": 0.01045185, "auxiliary_loss_mlp": 0.00752785, "balance_loss_clip": 1.00514698, "balance_loss_mlp": 1.00010478, "epoch": 0.9245475861239704, "flos": 66230599669920.0, "grad_norm": 0.7149608435034035, "language_loss": 0.5756526, "learning_rate": 5.934156150610103e-08, "loss": 0.59363228, "num_input_tokens_seen": 166279335, "step": 7689, "time_per_iteration": 3.128746509552002 }, { "auxiliary_loss_clip": 0.01138316, "auxiliary_loss_mlp": 0.01032634, "balance_loss_clip": 1.04452324, "balance_loss_mlp": 1.02543378, "epoch": 0.9246678290146095, "flos": 24239921657280.0, "grad_norm": 2.6113386237843614, "language_loss": 0.79340792, "learning_rate": 5.915336308394914e-08, "loss": 0.81511748, "num_input_tokens_seen": 166298170, "step": 7690, "time_per_iteration": 2.5471689701080322 }, { "auxiliary_loss_clip": 0.01146302, "auxiliary_loss_mlp": 0.01024769, "balance_loss_clip": 1.04371953, "balance_loss_mlp": 1.01833749, "epoch": 0.9247880719052486, "flos": 18988812265440.0, "grad_norm": 1.5777890557317669, "language_loss": 0.76812768, "learning_rate": 5.89654590866886e-08, "loss": 0.78983843, "num_input_tokens_seen": 166317670, "step": 7691, "time_per_iteration": 2.4795279502868652 }, { "auxiliary_loss_clip": 0.01106188, "auxiliary_loss_mlp": 0.01030499, "balance_loss_clip": 1.04614115, "balance_loss_mlp": 1.0229646, "epoch": 0.9249083147958876, "flos": 24024099057600.0, "grad_norm": 1.928402370660995, "language_loss": 0.88116407, "learning_rate": 5.877784954282483e-08, "loss": 0.90253091, "num_input_tokens_seen": 166337010, "step": 7692, "time_per_iteration": 3.3901326656341553 }, { "auxiliary_loss_clip": 0.0115682, "auxiliary_loss_mlp": 0.01024901, "balance_loss_clip": 1.04619575, "balance_loss_mlp": 1.01774192, "epoch": 0.9250285576865268, "flos": 30773976770880.0, "grad_norm": 2.195574929398914, "language_loss": 0.72541767, "learning_rate": 5.8590534480817963e-08, "loss": 0.74723482, "num_input_tokens_seen": 166358735, "step": 7693, "time_per_iteration": 2.5642454624176025 }, { "auxiliary_loss_clip": 0.01170228, "auxiliary_loss_mlp": 0.01023477, "balance_loss_clip": 1.04923701, "balance_loss_mlp": 1.01647949, "epoch": 0.9251488005771659, "flos": 10633583858400.0, "grad_norm": 2.2210359122978462, "language_loss": 0.72450531, "learning_rate": 5.840351392908349e-08, "loss": 0.74644226, "num_input_tokens_seen": 166374455, "step": 7694, "time_per_iteration": 2.4446258544921875 }, { "auxiliary_loss_clip": 0.01146383, "auxiliary_loss_mlp": 0.00761824, "balance_loss_clip": 1.04498303, "balance_loss_mlp": 1.00054705, "epoch": 0.9252690434678049, "flos": 23586419808960.0, "grad_norm": 2.5954246795895304, "language_loss": 0.70857346, "learning_rate": 5.821678791599205e-08, "loss": 0.72765553, "num_input_tokens_seen": 166393900, "step": 7695, "time_per_iteration": 2.548945903778076 }, { "auxiliary_loss_clip": 0.01138641, "auxiliary_loss_mlp": 0.01026531, "balance_loss_clip": 1.04493761, "balance_loss_mlp": 1.01984978, "epoch": 0.9253892863584441, "flos": 21469162286880.0, "grad_norm": 3.1560123704969962, "language_loss": 0.80715716, "learning_rate": 5.803035646986965e-08, "loss": 0.8288089, "num_input_tokens_seen": 166413235, "step": 7696, "time_per_iteration": 2.5463860034942627 }, { "auxiliary_loss_clip": 0.01170707, "auxiliary_loss_mlp": 0.01028041, "balance_loss_clip": 1.04852176, "balance_loss_mlp": 1.02064419, "epoch": 0.9255095292490831, "flos": 17456683476480.0, "grad_norm": 2.2098651949899635, "language_loss": 0.67227095, "learning_rate": 5.7844219618998766e-08, "loss": 0.69425845, "num_input_tokens_seen": 166427560, "step": 7697, "time_per_iteration": 2.4300432205200195 }, { "auxiliary_loss_clip": 0.01112092, "auxiliary_loss_mlp": 0.01026038, "balance_loss_clip": 1.03957224, "balance_loss_mlp": 1.01877832, "epoch": 0.9256297721397222, "flos": 24750679142880.0, "grad_norm": 1.9579336926497435, "language_loss": 0.71598685, "learning_rate": 5.765837739161505e-08, "loss": 0.73736817, "num_input_tokens_seen": 166446680, "step": 7698, "time_per_iteration": 3.370522975921631 }, { "auxiliary_loss_clip": 0.01125159, "auxiliary_loss_mlp": 0.01022629, "balance_loss_clip": 1.04262996, "balance_loss_mlp": 1.01610279, "epoch": 0.9257500150303614, "flos": 23112219400320.0, "grad_norm": 1.6257448561432872, "language_loss": 0.74449742, "learning_rate": 5.7472829815911504e-08, "loss": 0.7659753, "num_input_tokens_seen": 166465505, "step": 7699, "time_per_iteration": 2.585333824157715 }, { "auxiliary_loss_clip": 0.01135438, "auxiliary_loss_mlp": 0.01026095, "balance_loss_clip": 1.04523623, "balance_loss_mlp": 1.019315, "epoch": 0.9258702579210004, "flos": 22564689336000.0, "grad_norm": 1.8254238250619461, "language_loss": 0.81281978, "learning_rate": 5.7287576920035164e-08, "loss": 0.83443516, "num_input_tokens_seen": 166484520, "step": 7700, "time_per_iteration": 2.5214169025421143 }, { "auxiliary_loss_clip": 0.01123629, "auxiliary_loss_mlp": 0.01025371, "balance_loss_clip": 1.04453504, "balance_loss_mlp": 1.01849866, "epoch": 0.9259905008116395, "flos": 30004302721920.0, "grad_norm": 2.0022394758077526, "language_loss": 0.76748133, "learning_rate": 5.7102618732088435e-08, "loss": 0.7889713, "num_input_tokens_seen": 166503850, "step": 7701, "time_per_iteration": 2.6097049713134766 }, { "auxiliary_loss_clip": 0.01142689, "auxiliary_loss_mlp": 0.01022112, "balance_loss_clip": 1.04627299, "balance_loss_mlp": 1.01559472, "epoch": 0.9261107437022786, "flos": 24572131959360.0, "grad_norm": 1.7208433017716789, "language_loss": 0.74459255, "learning_rate": 5.6917955280130216e-08, "loss": 0.76624054, "num_input_tokens_seen": 166525330, "step": 7702, "time_per_iteration": 2.5505104064941406 }, { "auxiliary_loss_clip": 0.01152277, "auxiliary_loss_mlp": 0.01022502, "balance_loss_clip": 1.04748189, "balance_loss_mlp": 1.01533747, "epoch": 0.9262309865929177, "flos": 22018452282240.0, "grad_norm": 2.4885921638548094, "language_loss": 0.72055936, "learning_rate": 5.6733586592172755e-08, "loss": 0.74230713, "num_input_tokens_seen": 166544825, "step": 7703, "time_per_iteration": 2.488698720932007 }, { "auxiliary_loss_clip": 0.01133145, "auxiliary_loss_mlp": 0.00760982, "balance_loss_clip": 1.04138267, "balance_loss_mlp": 1.0004859, "epoch": 0.9263512294835567, "flos": 20339484597120.0, "grad_norm": 2.0537439916060296, "language_loss": 0.79964697, "learning_rate": 5.6549512696185244e-08, "loss": 0.81858826, "num_input_tokens_seen": 166563325, "step": 7704, "time_per_iteration": 2.512514352798462 }, { "auxiliary_loss_clip": 0.01166333, "auxiliary_loss_mlp": 0.01030529, "balance_loss_clip": 1.04719234, "balance_loss_mlp": 1.02358818, "epoch": 0.9264714723741959, "flos": 21215381848800.0, "grad_norm": 1.8914425522029277, "language_loss": 0.68212074, "learning_rate": 5.636573362009156e-08, "loss": 0.7040894, "num_input_tokens_seen": 166583385, "step": 7705, "time_per_iteration": 2.454415798187256 }, { "auxiliary_loss_clip": 0.01169549, "auxiliary_loss_mlp": 0.01026432, "balance_loss_clip": 1.04643273, "balance_loss_mlp": 1.01916337, "epoch": 0.926591715264835, "flos": 18004967796960.0, "grad_norm": 2.045308230376473, "language_loss": 0.77110553, "learning_rate": 5.618224939177074e-08, "loss": 0.79306543, "num_input_tokens_seen": 166601290, "step": 7706, "time_per_iteration": 3.943171977996826 }, { "auxiliary_loss_clip": 0.01128985, "auxiliary_loss_mlp": 0.01027091, "balance_loss_clip": 1.04243469, "balance_loss_mlp": 1.01966143, "epoch": 0.926711958155474, "flos": 36167974193280.0, "grad_norm": 1.9033702244298778, "language_loss": 0.70178974, "learning_rate": 5.599906003905719e-08, "loss": 0.72335052, "num_input_tokens_seen": 166623835, "step": 7707, "time_per_iteration": 2.6859943866729736 }, { "auxiliary_loss_clip": 0.01152283, "auxiliary_loss_mlp": 0.01028649, "balance_loss_clip": 1.04897726, "balance_loss_mlp": 1.02148151, "epoch": 0.9268322010461132, "flos": 21032740131840.0, "grad_norm": 2.1682180572579073, "language_loss": 0.81485856, "learning_rate": 5.581616558974023e-08, "loss": 0.8366679, "num_input_tokens_seen": 166642400, "step": 7708, "time_per_iteration": 2.4863181114196777 }, { "auxiliary_loss_clip": 0.01158339, "auxiliary_loss_mlp": 0.007622, "balance_loss_clip": 1.04610276, "balance_loss_mlp": 1.00052893, "epoch": 0.9269524439367522, "flos": 22964841749760.0, "grad_norm": 1.8214705077449147, "language_loss": 0.79111588, "learning_rate": 5.5633566071565444e-08, "loss": 0.81032127, "num_input_tokens_seen": 166661640, "step": 7709, "time_per_iteration": 2.4898970127105713 }, { "auxiliary_loss_clip": 0.01103751, "auxiliary_loss_mlp": 0.01025101, "balance_loss_clip": 1.04257321, "balance_loss_mlp": 1.01880634, "epoch": 0.9270726868273913, "flos": 41975556972480.0, "grad_norm": 1.9572916779233815, "language_loss": 0.71210921, "learning_rate": 5.5451261512232896e-08, "loss": 0.73339772, "num_input_tokens_seen": 166684320, "step": 7710, "time_per_iteration": 2.790498971939087 }, { "auxiliary_loss_clip": 0.01158742, "auxiliary_loss_mlp": 0.01022273, "balance_loss_clip": 1.04551709, "balance_loss_mlp": 1.01521003, "epoch": 0.9271929297180305, "flos": 19791774948000.0, "grad_norm": 2.134736199347647, "language_loss": 0.6283586, "learning_rate": 5.5269251939397576e-08, "loss": 0.65016878, "num_input_tokens_seen": 166703835, "step": 7711, "time_per_iteration": 2.470869779586792 }, { "auxiliary_loss_clip": 0.01127432, "auxiliary_loss_mlp": 0.01019029, "balance_loss_clip": 1.04208422, "balance_loss_mlp": 1.01223719, "epoch": 0.9273131726086695, "flos": 19968346698720.0, "grad_norm": 2.2560384539937353, "language_loss": 0.76517153, "learning_rate": 5.508753738067073e-08, "loss": 0.78663611, "num_input_tokens_seen": 166723375, "step": 7712, "time_per_iteration": 2.5778658390045166 }, { "auxiliary_loss_clip": 0.01154881, "auxiliary_loss_mlp": 0.01025489, "balance_loss_clip": 1.04413676, "balance_loss_mlp": 1.01869369, "epoch": 0.9274334154993086, "flos": 23258591376000.0, "grad_norm": 1.901980955866734, "language_loss": 0.79213703, "learning_rate": 5.4906117863617875e-08, "loss": 0.81394076, "num_input_tokens_seen": 166742760, "step": 7713, "time_per_iteration": 2.5079119205474854 }, { "auxiliary_loss_clip": 0.01119008, "auxiliary_loss_mlp": 0.01023766, "balance_loss_clip": 1.04028487, "balance_loss_mlp": 1.01721513, "epoch": 0.9275536583899477, "flos": 31795348074240.0, "grad_norm": 1.9673625720787575, "language_loss": 0.78007084, "learning_rate": 5.4724993415760533e-08, "loss": 0.80149859, "num_input_tokens_seen": 166761115, "step": 7714, "time_per_iteration": 2.620251417160034 }, { "auxiliary_loss_clip": 0.0113538, "auxiliary_loss_mlp": 0.00761956, "balance_loss_clip": 1.04309571, "balance_loss_mlp": 1.00051093, "epoch": 0.9276739012805868, "flos": 18697002155040.0, "grad_norm": 2.2760089814515236, "language_loss": 0.74578321, "learning_rate": 5.454416406457496e-08, "loss": 0.76475656, "num_input_tokens_seen": 166780210, "step": 7715, "time_per_iteration": 2.5336954593658447 }, { "auxiliary_loss_clip": 0.01154672, "auxiliary_loss_mlp": 0.01026733, "balance_loss_clip": 1.04639578, "balance_loss_mlp": 1.02035272, "epoch": 0.9277941441712259, "flos": 13879082391840.0, "grad_norm": 2.356760451661816, "language_loss": 0.73671532, "learning_rate": 5.436362983749299e-08, "loss": 0.75852937, "num_input_tokens_seen": 166795380, "step": 7716, "time_per_iteration": 2.4619953632354736 }, { "auxiliary_loss_clip": 0.01119262, "auxiliary_loss_mlp": 0.01023996, "balance_loss_clip": 1.04464447, "balance_loss_mlp": 1.0176456, "epoch": 0.927914387061865, "flos": 23258663209920.0, "grad_norm": 1.9272465134492827, "language_loss": 0.64342225, "learning_rate": 5.418339076190137e-08, "loss": 0.66485488, "num_input_tokens_seen": 166814890, "step": 7717, "time_per_iteration": 3.2645137310028076 }, { "auxiliary_loss_clip": 0.01130695, "auxiliary_loss_mlp": 0.01026647, "balance_loss_clip": 1.04308331, "balance_loss_mlp": 1.01914895, "epoch": 0.9280346299525041, "flos": 18073735998720.0, "grad_norm": 2.067165891945213, "language_loss": 0.88747191, "learning_rate": 5.400344686514202e-08, "loss": 0.90904534, "num_input_tokens_seen": 166832475, "step": 7718, "time_per_iteration": 2.5145068168640137 }, { "auxiliary_loss_clip": 0.01153696, "auxiliary_loss_mlp": 0.01021576, "balance_loss_clip": 1.04737866, "balance_loss_mlp": 1.01479018, "epoch": 0.9281548728431431, "flos": 22342904520960.0, "grad_norm": 1.9924143346662406, "language_loss": 0.66888082, "learning_rate": 5.38237981745131e-08, "loss": 0.69063354, "num_input_tokens_seen": 166850590, "step": 7719, "time_per_iteration": 2.4903321266174316 }, { "auxiliary_loss_clip": 0.01155354, "auxiliary_loss_mlp": 0.00761634, "balance_loss_clip": 1.0450542, "balance_loss_mlp": 1.00049138, "epoch": 0.9282751157337822, "flos": 18843769217280.0, "grad_norm": 1.6156542878415665, "language_loss": 0.81243181, "learning_rate": 5.364444471726592e-08, "loss": 0.83160174, "num_input_tokens_seen": 166869795, "step": 7720, "time_per_iteration": 2.4956629276275635 }, { "auxiliary_loss_clip": 0.01151424, "auxiliary_loss_mlp": 0.01020787, "balance_loss_clip": 1.04408193, "balance_loss_mlp": 1.01370001, "epoch": 0.9283953586244214, "flos": 25556838434880.0, "grad_norm": 1.9100416683576023, "language_loss": 0.80013531, "learning_rate": 5.346538652060939e-08, "loss": 0.82185751, "num_input_tokens_seen": 166891150, "step": 7721, "time_per_iteration": 2.520644187927246 }, { "auxiliary_loss_clip": 0.01137828, "auxiliary_loss_mlp": 0.01022905, "balance_loss_clip": 1.04658616, "balance_loss_mlp": 1.01652682, "epoch": 0.9285156015150604, "flos": 18223484168640.0, "grad_norm": 2.08650852184622, "language_loss": 0.70001137, "learning_rate": 5.3286623611705994e-08, "loss": 0.72161865, "num_input_tokens_seen": 166909195, "step": 7722, "time_per_iteration": 2.5120952129364014 }, { "auxiliary_loss_clip": 0.01059386, "auxiliary_loss_mlp": 0.01002072, "balance_loss_clip": 1.00538516, "balance_loss_mlp": 1.00083792, "epoch": 0.9286358444056995, "flos": 66400030333920.0, "grad_norm": 0.8189736237858773, "language_loss": 0.60640347, "learning_rate": 5.3108156017673824e-08, "loss": 0.62701797, "num_input_tokens_seen": 166970955, "step": 7723, "time_per_iteration": 3.1267573833465576 }, { "auxiliary_loss_clip": 0.01144344, "auxiliary_loss_mlp": 0.01024106, "balance_loss_clip": 1.04674518, "balance_loss_mlp": 1.01668501, "epoch": 0.9287560872963386, "flos": 22345634209920.0, "grad_norm": 1.7761148827311914, "language_loss": 0.71405238, "learning_rate": 5.2929983765586775e-08, "loss": 0.73573685, "num_input_tokens_seen": 166989735, "step": 7724, "time_per_iteration": 2.499898672103882 }, { "auxiliary_loss_clip": 0.01167954, "auxiliary_loss_mlp": 0.01022604, "balance_loss_clip": 1.04859352, "balance_loss_mlp": 1.01587796, "epoch": 0.9288763301869777, "flos": 25700229302880.0, "grad_norm": 1.706734037476395, "language_loss": 0.62215388, "learning_rate": 5.275210688247278e-08, "loss": 0.64405942, "num_input_tokens_seen": 167010060, "step": 7725, "time_per_iteration": 3.2635490894317627 }, { "auxiliary_loss_clip": 0.01114629, "auxiliary_loss_mlp": 0.01026964, "balance_loss_clip": 1.04428482, "balance_loss_mlp": 1.02000844, "epoch": 0.9289965730776167, "flos": 12312048706080.0, "grad_norm": 1.9666889644670473, "language_loss": 0.85104328, "learning_rate": 5.257452539531604e-08, "loss": 0.87245923, "num_input_tokens_seen": 167027130, "step": 7726, "time_per_iteration": 2.543046236038208 }, { "auxiliary_loss_clip": 0.01154804, "auxiliary_loss_mlp": 0.01024262, "balance_loss_clip": 1.04524636, "balance_loss_mlp": 1.01729178, "epoch": 0.9291168159682559, "flos": 26685977370240.0, "grad_norm": 1.566258297304683, "language_loss": 0.68543082, "learning_rate": 5.2397239331055445e-08, "loss": 0.70722151, "num_input_tokens_seen": 167049130, "step": 7727, "time_per_iteration": 2.5472981929779053 }, { "auxiliary_loss_clip": 0.0113891, "auxiliary_loss_mlp": 0.01031503, "balance_loss_clip": 1.04741168, "balance_loss_mlp": 1.02471066, "epoch": 0.929237058858895, "flos": 14538259119840.0, "grad_norm": 2.184720113452486, "language_loss": 0.81093752, "learning_rate": 5.2220248716585036e-08, "loss": 0.83264166, "num_input_tokens_seen": 167066810, "step": 7728, "time_per_iteration": 2.4707181453704834 }, { "auxiliary_loss_clip": 0.01143166, "auxiliary_loss_mlp": 0.01023246, "balance_loss_clip": 1.04195547, "balance_loss_mlp": 1.0161798, "epoch": 0.929357301749534, "flos": 23835459208320.0, "grad_norm": 2.340738846920696, "language_loss": 0.75314116, "learning_rate": 5.204355357875445e-08, "loss": 0.77480525, "num_input_tokens_seen": 167085155, "step": 7729, "time_per_iteration": 2.5064966678619385 }, { "auxiliary_loss_clip": 0.0113681, "auxiliary_loss_mlp": 0.01023228, "balance_loss_clip": 1.04300165, "balance_loss_mlp": 1.01589382, "epoch": 0.9294775446401732, "flos": 12969321835200.0, "grad_norm": 3.350438799432855, "language_loss": 0.70500469, "learning_rate": 5.1867153944367584e-08, "loss": 0.72660506, "num_input_tokens_seen": 167101545, "step": 7730, "time_per_iteration": 2.495370388031006 }, { "auxiliary_loss_clip": 0.01130302, "auxiliary_loss_mlp": 0.01023386, "balance_loss_clip": 1.04388285, "balance_loss_mlp": 1.01645684, "epoch": 0.9295977875308122, "flos": 26211812878560.0, "grad_norm": 1.5316499758987785, "language_loss": 0.73315305, "learning_rate": 5.16910498401848e-08, "loss": 0.75468987, "num_input_tokens_seen": 167120995, "step": 7731, "time_per_iteration": 2.579564094543457 }, { "auxiliary_loss_clip": 0.01166935, "auxiliary_loss_mlp": 0.01028607, "balance_loss_clip": 1.04888344, "balance_loss_mlp": 1.02244425, "epoch": 0.9297180304214513, "flos": 16472300253600.0, "grad_norm": 1.9100069153456698, "language_loss": 0.83492047, "learning_rate": 5.151524129292073e-08, "loss": 0.8568759, "num_input_tokens_seen": 167138890, "step": 7732, "time_per_iteration": 3.2018423080444336 }, { "auxiliary_loss_clip": 0.01150446, "auxiliary_loss_mlp": 0.01023754, "balance_loss_clip": 1.04480207, "balance_loss_mlp": 1.01720095, "epoch": 0.9298382733120905, "flos": 24060440632800.0, "grad_norm": 2.2859363575390557, "language_loss": 0.6623832, "learning_rate": 5.1339728329245155e-08, "loss": 0.68412519, "num_input_tokens_seen": 167159455, "step": 7733, "time_per_iteration": 3.2618629932403564 }, { "auxiliary_loss_clip": 0.01174159, "auxiliary_loss_mlp": 0.01028865, "balance_loss_clip": 1.04956293, "balance_loss_mlp": 1.02135754, "epoch": 0.9299585162027295, "flos": 22127656592640.0, "grad_norm": 2.3502714950050327, "language_loss": 0.79435772, "learning_rate": 5.116451097578367e-08, "loss": 0.81638789, "num_input_tokens_seen": 167178495, "step": 7734, "time_per_iteration": 2.4520974159240723 }, { "auxiliary_loss_clip": 0.01125187, "auxiliary_loss_mlp": 0.01026531, "balance_loss_clip": 1.04313421, "balance_loss_mlp": 1.02012038, "epoch": 0.9300787590933686, "flos": 21471784224960.0, "grad_norm": 1.8885412042391376, "language_loss": 0.74411356, "learning_rate": 5.0989589259115895e-08, "loss": 0.76563072, "num_input_tokens_seen": 167199380, "step": 7735, "time_per_iteration": 2.573976755142212 }, { "auxiliary_loss_clip": 0.01149345, "auxiliary_loss_mlp": 0.01023219, "balance_loss_clip": 1.04310966, "balance_loss_mlp": 1.0159291, "epoch": 0.9301990019840077, "flos": 17779591285920.0, "grad_norm": 1.8598369455789108, "language_loss": 0.71533942, "learning_rate": 5.081496320577816e-08, "loss": 0.73706514, "num_input_tokens_seen": 167216500, "step": 7736, "time_per_iteration": 2.4438273906707764 }, { "auxiliary_loss_clip": 0.0105353, "auxiliary_loss_mlp": 0.01002986, "balance_loss_clip": 1.02415037, "balance_loss_mlp": 1.00171626, "epoch": 0.9303192448746468, "flos": 58896131977920.0, "grad_norm": 0.9095116859623547, "language_loss": 0.61205488, "learning_rate": 5.0640632842260835e-08, "loss": 0.63262004, "num_input_tokens_seen": 167276760, "step": 7737, "time_per_iteration": 3.1960289478302 }, { "auxiliary_loss_clip": 0.01124123, "auxiliary_loss_mlp": 0.00761557, "balance_loss_clip": 1.04613638, "balance_loss_mlp": 1.00047874, "epoch": 0.9304394877652858, "flos": 57663529057440.0, "grad_norm": 1.394784174020986, "language_loss": 0.72237062, "learning_rate": 5.0466598195009426e-08, "loss": 0.74122745, "num_input_tokens_seen": 167303630, "step": 7738, "time_per_iteration": 2.893721580505371 }, { "auxiliary_loss_clip": 0.01124154, "auxiliary_loss_mlp": 0.010229, "balance_loss_clip": 1.04520321, "balance_loss_mlp": 1.01628661, "epoch": 0.930559730655925, "flos": 20996147137920.0, "grad_norm": 1.9361786703164006, "language_loss": 0.70293653, "learning_rate": 5.0292859290425036e-08, "loss": 0.72440708, "num_input_tokens_seen": 167321500, "step": 7739, "time_per_iteration": 2.525653123855591 }, { "auxiliary_loss_clip": 0.01166867, "auxiliary_loss_mlp": 0.01024233, "balance_loss_clip": 1.04862833, "balance_loss_mlp": 1.01820755, "epoch": 0.9306799735465641, "flos": 23258268123360.0, "grad_norm": 1.869649829512178, "language_loss": 0.77836013, "learning_rate": 5.011941615486348e-08, "loss": 0.80027115, "num_input_tokens_seen": 167340615, "step": 7740, "time_per_iteration": 2.4704091548919678 }, { "auxiliary_loss_clip": 0.01166667, "auxiliary_loss_mlp": 0.01026163, "balance_loss_clip": 1.04762042, "balance_loss_mlp": 1.01978588, "epoch": 0.9308002164372031, "flos": 15231550571520.0, "grad_norm": 1.8532117252905986, "language_loss": 0.84429282, "learning_rate": 4.994626881463659e-08, "loss": 0.86622113, "num_input_tokens_seen": 167356870, "step": 7741, "time_per_iteration": 2.413869857788086 }, { "auxiliary_loss_clip": 0.0109599, "auxiliary_loss_mlp": 0.01026024, "balance_loss_clip": 1.0395422, "balance_loss_mlp": 1.01897895, "epoch": 0.9309204593278423, "flos": 30847486011360.0, "grad_norm": 1.669579400132361, "language_loss": 0.71009827, "learning_rate": 4.9773417296009814e-08, "loss": 0.73131847, "num_input_tokens_seen": 167378390, "step": 7742, "time_per_iteration": 2.639141321182251 }, { "auxiliary_loss_clip": 0.01159059, "auxiliary_loss_mlp": 0.01024159, "balance_loss_clip": 1.0478152, "balance_loss_mlp": 1.01721239, "epoch": 0.9310407022184813, "flos": 23037273481440.0, "grad_norm": 2.08027792355187, "language_loss": 0.65431887, "learning_rate": 4.960086162520527e-08, "loss": 0.67615104, "num_input_tokens_seen": 167398480, "step": 7743, "time_per_iteration": 3.200129508972168 }, { "auxiliary_loss_clip": 0.01114033, "auxiliary_loss_mlp": 0.0102191, "balance_loss_clip": 1.04234648, "balance_loss_mlp": 1.01572347, "epoch": 0.9311609451091204, "flos": 22127979845280.0, "grad_norm": 2.2117643058055543, "language_loss": 0.82348245, "learning_rate": 4.942860182839936e-08, "loss": 0.84484184, "num_input_tokens_seen": 167416825, "step": 7744, "time_per_iteration": 2.5908405780792236 }, { "auxiliary_loss_clip": 0.01138058, "auxiliary_loss_mlp": 0.01023316, "balance_loss_clip": 1.0445776, "balance_loss_mlp": 1.0165391, "epoch": 0.9312811879997596, "flos": 21099209648160.0, "grad_norm": 1.792099612867186, "language_loss": 0.79398167, "learning_rate": 4.925663793172341e-08, "loss": 0.81559545, "num_input_tokens_seen": 167434785, "step": 7745, "time_per_iteration": 2.501943349838257 }, { "auxiliary_loss_clip": 0.01035934, "auxiliary_loss_mlp": 0.00752547, "balance_loss_clip": 1.00682294, "balance_loss_mlp": 1.00007749, "epoch": 0.9314014308903986, "flos": 67148190123840.0, "grad_norm": 0.7890777526740261, "language_loss": 0.56535882, "learning_rate": 4.908496996126477e-08, "loss": 0.58324367, "num_input_tokens_seen": 167498245, "step": 7746, "time_per_iteration": 3.151231288909912 }, { "auxiliary_loss_clip": 0.01154741, "auxiliary_loss_mlp": 0.01026951, "balance_loss_clip": 1.05050993, "balance_loss_mlp": 1.01956034, "epoch": 0.9315216737810377, "flos": 22565587260000.0, "grad_norm": 1.495760714085565, "language_loss": 0.76130974, "learning_rate": 4.89135979430646e-08, "loss": 0.78312671, "num_input_tokens_seen": 167518290, "step": 7747, "time_per_iteration": 2.492698907852173 }, { "auxiliary_loss_clip": 0.01168819, "auxiliary_loss_mlp": 0.01027079, "balance_loss_clip": 1.04898036, "balance_loss_mlp": 1.02005196, "epoch": 0.9316419166716768, "flos": 23984058035520.0, "grad_norm": 1.6941102650128825, "language_loss": 0.85703433, "learning_rate": 4.874252190312078e-08, "loss": 0.87899327, "num_input_tokens_seen": 167538675, "step": 7748, "time_per_iteration": 2.4797346591949463 }, { "auxiliary_loss_clip": 0.01154091, "auxiliary_loss_mlp": 0.01025075, "balance_loss_clip": 1.04757094, "balance_loss_mlp": 1.0178864, "epoch": 0.9317621595623159, "flos": 30230469406080.0, "grad_norm": 1.5686033481824082, "language_loss": 0.64573848, "learning_rate": 4.857174186738477e-08, "loss": 0.66753018, "num_input_tokens_seen": 167562025, "step": 7749, "time_per_iteration": 2.569507598876953 }, { "auxiliary_loss_clip": 0.01169981, "auxiliary_loss_mlp": 0.01027326, "balance_loss_clip": 1.05000544, "balance_loss_mlp": 1.02040339, "epoch": 0.931882402452955, "flos": 15742739060640.0, "grad_norm": 2.2295580564596227, "language_loss": 0.72822636, "learning_rate": 4.840125786176408e-08, "loss": 0.75019944, "num_input_tokens_seen": 167578230, "step": 7750, "time_per_iteration": 2.445335865020752 }, { "auxiliary_loss_clip": 0.01135877, "auxiliary_loss_mlp": 0.01024055, "balance_loss_clip": 1.04491949, "balance_loss_mlp": 1.01752496, "epoch": 0.932002645343594, "flos": 28366525401600.0, "grad_norm": 1.8724212420095085, "language_loss": 0.77198702, "learning_rate": 4.823106991212067e-08, "loss": 0.79358637, "num_input_tokens_seen": 167597470, "step": 7751, "time_per_iteration": 3.3706727027893066 }, { "auxiliary_loss_clip": 0.01153437, "auxiliary_loss_mlp": 0.01024777, "balance_loss_clip": 1.04420161, "balance_loss_mlp": 1.01780343, "epoch": 0.9321228882342332, "flos": 15341149968480.0, "grad_norm": 2.071915336238978, "language_loss": 0.83407593, "learning_rate": 4.806117804427212e-08, "loss": 0.85585803, "num_input_tokens_seen": 167615405, "step": 7752, "time_per_iteration": 2.4835190773010254 }, { "auxiliary_loss_clip": 0.01147087, "auxiliary_loss_mlp": 0.01025735, "balance_loss_clip": 1.04394674, "balance_loss_mlp": 1.01899362, "epoch": 0.9322431311248722, "flos": 17895368400000.0, "grad_norm": 2.087512385703838, "language_loss": 0.64282048, "learning_rate": 4.7891582283990926e-08, "loss": 0.6645487, "num_input_tokens_seen": 167634130, "step": 7753, "time_per_iteration": 2.4821064472198486 }, { "auxiliary_loss_clip": 0.01121324, "auxiliary_loss_mlp": 0.01025336, "balance_loss_clip": 1.04182684, "balance_loss_mlp": 1.01865697, "epoch": 0.9323633740155113, "flos": 24169716777120.0, "grad_norm": 1.5473289734437057, "language_loss": 0.72761297, "learning_rate": 4.772228265700473e-08, "loss": 0.74907953, "num_input_tokens_seen": 167654990, "step": 7754, "time_per_iteration": 2.6007957458496094 }, { "auxiliary_loss_clip": 0.01155555, "auxiliary_loss_mlp": 0.0102971, "balance_loss_clip": 1.04630971, "balance_loss_mlp": 1.0229063, "epoch": 0.9324836169061504, "flos": 15043485393600.0, "grad_norm": 2.0400353660989863, "language_loss": 0.75998569, "learning_rate": 4.75532791889961e-08, "loss": 0.7818383, "num_input_tokens_seen": 167671690, "step": 7755, "time_per_iteration": 2.4846179485321045 }, { "auxiliary_loss_clip": 0.0115256, "auxiliary_loss_mlp": 0.01025836, "balance_loss_clip": 1.04438829, "balance_loss_mlp": 1.01891589, "epoch": 0.9326038597967895, "flos": 18624893676000.0, "grad_norm": 1.958017826756933, "language_loss": 0.6561175, "learning_rate": 4.738457190560252e-08, "loss": 0.67790151, "num_input_tokens_seen": 167690800, "step": 7756, "time_per_iteration": 2.490427255630493 }, { "auxiliary_loss_clip": 0.01111937, "auxiliary_loss_mlp": 0.01024654, "balance_loss_clip": 1.04516745, "balance_loss_mlp": 1.01816607, "epoch": 0.9327241026874286, "flos": 18952650275040.0, "grad_norm": 2.093241047179762, "language_loss": 0.78942066, "learning_rate": 4.721616083241664e-08, "loss": 0.8107866, "num_input_tokens_seen": 167709055, "step": 7757, "time_per_iteration": 2.5778818130493164 }, { "auxiliary_loss_clip": 0.01148191, "auxiliary_loss_mlp": 0.01029313, "balance_loss_clip": 1.04466212, "balance_loss_mlp": 1.02237761, "epoch": 0.9328443455780677, "flos": 29570287003200.0, "grad_norm": 3.132102234427851, "language_loss": 0.77715665, "learning_rate": 4.7048045994986684e-08, "loss": 0.79893172, "num_input_tokens_seen": 167729915, "step": 7758, "time_per_iteration": 4.093000411987305 }, { "auxiliary_loss_clip": 0.01157077, "auxiliary_loss_mlp": 0.01024615, "balance_loss_clip": 1.04896677, "balance_loss_mlp": 1.01770413, "epoch": 0.9329645884687068, "flos": 30081726911040.0, "grad_norm": 2.062865821729519, "language_loss": 0.90871739, "learning_rate": 4.688022741881559e-08, "loss": 0.9305343, "num_input_tokens_seen": 167750440, "step": 7759, "time_per_iteration": 2.5437681674957275 }, { "auxiliary_loss_clip": 0.01151012, "auxiliary_loss_mlp": 0.01026351, "balance_loss_clip": 1.04614115, "balance_loss_mlp": 1.01981211, "epoch": 0.9330848313593458, "flos": 21867985773120.0, "grad_norm": 2.1861286113883516, "language_loss": 0.75011045, "learning_rate": 4.671270512936076e-08, "loss": 0.77188408, "num_input_tokens_seen": 167769600, "step": 7760, "time_per_iteration": 2.500783681869507 }, { "auxiliary_loss_clip": 0.0111592, "auxiliary_loss_mlp": 0.01022771, "balance_loss_clip": 1.04061055, "balance_loss_mlp": 1.01606548, "epoch": 0.933205074249985, "flos": 22127225589120.0, "grad_norm": 1.8313853204475559, "language_loss": 0.82875228, "learning_rate": 4.6545479152035884e-08, "loss": 0.85013926, "num_input_tokens_seen": 167788770, "step": 7761, "time_per_iteration": 2.5423688888549805 }, { "auxiliary_loss_clip": 0.01155872, "auxiliary_loss_mlp": 0.01025198, "balance_loss_clip": 1.0478363, "balance_loss_mlp": 1.01873946, "epoch": 0.9333253171406241, "flos": 15341257719360.0, "grad_norm": 1.7891552860419544, "language_loss": 0.76016021, "learning_rate": 4.637854951220821e-08, "loss": 0.78197086, "num_input_tokens_seen": 167805555, "step": 7762, "time_per_iteration": 2.467984676361084 }, { "auxiliary_loss_clip": 0.01120173, "auxiliary_loss_mlp": 0.01022194, "balance_loss_clip": 1.04169106, "balance_loss_mlp": 1.01528585, "epoch": 0.9334455600312631, "flos": 15706146066720.0, "grad_norm": 1.8496243662760756, "language_loss": 0.75060058, "learning_rate": 4.621191623520171e-08, "loss": 0.77202427, "num_input_tokens_seen": 167823985, "step": 7763, "time_per_iteration": 2.5422232151031494 }, { "auxiliary_loss_clip": 0.01109414, "auxiliary_loss_mlp": 0.01025856, "balance_loss_clip": 1.04548514, "balance_loss_mlp": 1.01914215, "epoch": 0.9335658029219023, "flos": 22163567164320.0, "grad_norm": 2.748490461762646, "language_loss": 0.84809965, "learning_rate": 4.604557934629372e-08, "loss": 0.86945236, "num_input_tokens_seen": 167843060, "step": 7764, "time_per_iteration": 2.6248698234558105 }, { "auxiliary_loss_clip": 0.01133039, "auxiliary_loss_mlp": 0.01018601, "balance_loss_clip": 1.04370511, "balance_loss_mlp": 1.01243174, "epoch": 0.9336860458125413, "flos": 20266837363680.0, "grad_norm": 1.6561706988368627, "language_loss": 0.80284625, "learning_rate": 4.587953887071805e-08, "loss": 0.82436264, "num_input_tokens_seen": 167862880, "step": 7765, "time_per_iteration": 2.5220837593078613 }, { "auxiliary_loss_clip": 0.01132079, "auxiliary_loss_mlp": 0.01030206, "balance_loss_clip": 1.04003656, "balance_loss_mlp": 1.02299988, "epoch": 0.9338062887031804, "flos": 20919692706720.0, "grad_norm": 1.8825867346317122, "language_loss": 0.86084819, "learning_rate": 4.5713794833662554e-08, "loss": 0.88247097, "num_input_tokens_seen": 167882095, "step": 7766, "time_per_iteration": 2.4915995597839355 }, { "auxiliary_loss_clip": 0.01168365, "auxiliary_loss_mlp": 0.01026438, "balance_loss_clip": 1.04741347, "balance_loss_mlp": 1.0192523, "epoch": 0.9339265315938196, "flos": 23221639212480.0, "grad_norm": 2.668730800753698, "language_loss": 0.63843447, "learning_rate": 4.5548347260270236e-08, "loss": 0.66038251, "num_input_tokens_seen": 167901385, "step": 7767, "time_per_iteration": 2.4747023582458496 }, { "auxiliary_loss_clip": 0.01122039, "auxiliary_loss_mlp": 0.01028238, "balance_loss_clip": 1.04296768, "balance_loss_mlp": 1.02135348, "epoch": 0.9340467744844586, "flos": 22820265622080.0, "grad_norm": 2.4723983733944777, "language_loss": 0.6927976, "learning_rate": 4.538319617564012e-08, "loss": 0.71430039, "num_input_tokens_seen": 167920405, "step": 7768, "time_per_iteration": 2.5548856258392334 }, { "auxiliary_loss_clip": 0.01139432, "auxiliary_loss_mlp": 0.01022674, "balance_loss_clip": 1.04346943, "balance_loss_mlp": 1.01558733, "epoch": 0.9341670173750977, "flos": 23660431886880.0, "grad_norm": 2.068696727659266, "language_loss": 0.74329567, "learning_rate": 4.521834160482485e-08, "loss": 0.76491666, "num_input_tokens_seen": 167939145, "step": 7769, "time_per_iteration": 3.29473614692688 }, { "auxiliary_loss_clip": 0.01157423, "auxiliary_loss_mlp": 0.01024483, "balance_loss_clip": 1.04648459, "balance_loss_mlp": 1.01816154, "epoch": 0.9342872602657368, "flos": 24824260217280.0, "grad_norm": 1.4904380726128665, "language_loss": 0.82000506, "learning_rate": 4.5053783572832846e-08, "loss": 0.84182405, "num_input_tokens_seen": 167959325, "step": 7770, "time_per_iteration": 2.5326521396636963 }, { "auxiliary_loss_clip": 0.01155194, "auxiliary_loss_mlp": 0.01024758, "balance_loss_clip": 1.04741812, "balance_loss_mlp": 1.01796603, "epoch": 0.9344075031563759, "flos": 25771834944480.0, "grad_norm": 1.5541342156471036, "language_loss": 0.76243377, "learning_rate": 4.488952210462771e-08, "loss": 0.78423333, "num_input_tokens_seen": 167979530, "step": 7771, "time_per_iteration": 2.503336191177368 }, { "auxiliary_loss_clip": 0.01167414, "auxiliary_loss_mlp": 0.01030047, "balance_loss_clip": 1.04832113, "balance_loss_mlp": 1.02332973, "epoch": 0.9345277460470149, "flos": 25551307223040.0, "grad_norm": 1.7791823985601236, "language_loss": 0.85794193, "learning_rate": 4.4725557225127495e-08, "loss": 0.87991649, "num_input_tokens_seen": 167997870, "step": 7772, "time_per_iteration": 2.507167100906372 }, { "auxiliary_loss_clip": 0.01154614, "auxiliary_loss_mlp": 0.01029644, "balance_loss_clip": 1.04663575, "balance_loss_mlp": 1.0235256, "epoch": 0.9346479889376541, "flos": 34313117179680.0, "grad_norm": 4.298043510290735, "language_loss": 0.79309231, "learning_rate": 4.456188895920565e-08, "loss": 0.81493497, "num_input_tokens_seen": 168019625, "step": 7773, "time_per_iteration": 2.5807852745056152 }, { "auxiliary_loss_clip": 0.01168181, "auxiliary_loss_mlp": 0.01023266, "balance_loss_clip": 1.04778981, "balance_loss_mlp": 1.01596189, "epoch": 0.9347682318282932, "flos": 19093742457600.0, "grad_norm": 1.9270914002904578, "language_loss": 0.85531402, "learning_rate": 4.439851733169031e-08, "loss": 0.8772285, "num_input_tokens_seen": 168037415, "step": 7774, "time_per_iteration": 2.4366002082824707 }, { "auxiliary_loss_clip": 0.01125334, "auxiliary_loss_mlp": 0.01020377, "balance_loss_clip": 1.04313612, "balance_loss_mlp": 1.01364195, "epoch": 0.9348884747189322, "flos": 26249591132160.0, "grad_norm": 2.1074622570281383, "language_loss": 0.69343972, "learning_rate": 4.4235442367365204e-08, "loss": 0.71489686, "num_input_tokens_seen": 168057725, "step": 7775, "time_per_iteration": 2.5908303260803223 }, { "auxiliary_loss_clip": 0.01134953, "auxiliary_loss_mlp": 0.01027252, "balance_loss_clip": 1.04109097, "balance_loss_mlp": 1.0204953, "epoch": 0.9350087176095714, "flos": 18333083565600.0, "grad_norm": 2.1189599512939536, "language_loss": 0.79203236, "learning_rate": 4.4072664090968545e-08, "loss": 0.81365442, "num_input_tokens_seen": 168076110, "step": 7776, "time_per_iteration": 2.5018410682678223 }, { "auxiliary_loss_clip": 0.01137957, "auxiliary_loss_mlp": 0.01025418, "balance_loss_clip": 1.04054856, "balance_loss_mlp": 1.01784778, "epoch": 0.9351289605002104, "flos": 19318256961600.0, "grad_norm": 1.7285097690428777, "language_loss": 0.84786749, "learning_rate": 4.391018252719347e-08, "loss": 0.86950123, "num_input_tokens_seen": 168095905, "step": 7777, "time_per_iteration": 3.294233798980713 }, { "auxiliary_loss_clip": 0.01141777, "auxiliary_loss_mlp": 0.01024537, "balance_loss_clip": 1.04457641, "balance_loss_mlp": 1.01784039, "epoch": 0.9352492033908495, "flos": 18799992831360.0, "grad_norm": 1.6580973887694495, "language_loss": 0.69017601, "learning_rate": 4.374799770068849e-08, "loss": 0.7118392, "num_input_tokens_seen": 168112580, "step": 7778, "time_per_iteration": 2.505404233932495 }, { "auxiliary_loss_clip": 0.01150236, "auxiliary_loss_mlp": 0.01018231, "balance_loss_clip": 1.04568708, "balance_loss_mlp": 1.01160598, "epoch": 0.9353694462814887, "flos": 29530138230240.0, "grad_norm": 1.8465535644837432, "language_loss": 0.74977082, "learning_rate": 4.358610963605658e-08, "loss": 0.77145553, "num_input_tokens_seen": 168133030, "step": 7779, "time_per_iteration": 2.5259389877319336 }, { "auxiliary_loss_clip": 0.01172486, "auxiliary_loss_mlp": 0.0102945, "balance_loss_clip": 1.04977989, "balance_loss_mlp": 1.02264595, "epoch": 0.9354896891721277, "flos": 30665454882720.0, "grad_norm": 2.038386145451177, "language_loss": 0.68915433, "learning_rate": 4.342451835785677e-08, "loss": 0.71117371, "num_input_tokens_seen": 168153940, "step": 7780, "time_per_iteration": 2.5324630737304688 }, { "auxiliary_loss_clip": 0.01137151, "auxiliary_loss_mlp": 0.01024307, "balance_loss_clip": 1.04486847, "balance_loss_mlp": 1.01746488, "epoch": 0.9356099320627668, "flos": 19463910598080.0, "grad_norm": 1.5359350287905784, "language_loss": 0.75215924, "learning_rate": 4.3263223890601665e-08, "loss": 0.77377379, "num_input_tokens_seen": 168172650, "step": 7781, "time_per_iteration": 2.4895215034484863 }, { "auxiliary_loss_clip": 0.01150732, "auxiliary_loss_mlp": 0.00761582, "balance_loss_clip": 1.04888129, "balance_loss_mlp": 1.00048137, "epoch": 0.9357301749534058, "flos": 19098160243680.0, "grad_norm": 1.6983476083383586, "language_loss": 0.79145956, "learning_rate": 4.31022262587597e-08, "loss": 0.81058264, "num_input_tokens_seen": 168191325, "step": 7782, "time_per_iteration": 2.48515248298645 }, { "auxiliary_loss_clip": 0.01156167, "auxiliary_loss_mlp": 0.01026497, "balance_loss_clip": 1.04838514, "balance_loss_mlp": 1.01934767, "epoch": 0.935850417844045, "flos": 23550365569440.0, "grad_norm": 1.5423644370987621, "language_loss": 0.6611619, "learning_rate": 4.2941525486754225e-08, "loss": 0.68298852, "num_input_tokens_seen": 168211645, "step": 7783, "time_per_iteration": 3.267481803894043 }, { "auxiliary_loss_clip": 0.01123014, "auxiliary_loss_mlp": 0.01022568, "balance_loss_clip": 1.04500937, "balance_loss_mlp": 1.01620507, "epoch": 0.935970660734684, "flos": 18588336599040.0, "grad_norm": 1.8836750264961863, "language_loss": 0.7949357, "learning_rate": 4.278112159896286e-08, "loss": 0.81639153, "num_input_tokens_seen": 168229485, "step": 7784, "time_per_iteration": 3.234445810317993 }, { "auxiliary_loss_clip": 0.01132968, "auxiliary_loss_mlp": 0.01024936, "balance_loss_clip": 1.04093266, "balance_loss_mlp": 1.01850426, "epoch": 0.9360909036253231, "flos": 20631258790560.0, "grad_norm": 2.043875315189584, "language_loss": 0.67622864, "learning_rate": 4.2621014619719896e-08, "loss": 0.69780773, "num_input_tokens_seen": 168247250, "step": 7785, "time_per_iteration": 2.5257458686828613 }, { "auxiliary_loss_clip": 0.01038843, "auxiliary_loss_mlp": 0.01000503, "balance_loss_clip": 1.0062933, "balance_loss_mlp": 0.9992156, "epoch": 0.9362111465159623, "flos": 61791432200640.0, "grad_norm": 0.7191122800161643, "language_loss": 0.5865078, "learning_rate": 4.246120457331215e-08, "loss": 0.60690129, "num_input_tokens_seen": 168309425, "step": 7786, "time_per_iteration": 3.1440811157226562 }, { "auxiliary_loss_clip": 0.01132512, "auxiliary_loss_mlp": 0.01025505, "balance_loss_clip": 1.04462218, "balance_loss_mlp": 1.01814103, "epoch": 0.9363313894066013, "flos": 24170399199360.0, "grad_norm": 3.628996815498469, "language_loss": 0.72161365, "learning_rate": 4.2301691483983325e-08, "loss": 0.74319381, "num_input_tokens_seen": 168329545, "step": 7787, "time_per_iteration": 2.5415871143341064 }, { "auxiliary_loss_clip": 0.01157711, "auxiliary_loss_mlp": 0.0102765, "balance_loss_clip": 1.04684556, "balance_loss_mlp": 1.02025628, "epoch": 0.9364516322972404, "flos": 20120357637120.0, "grad_norm": 2.1084868465300954, "language_loss": 0.75577241, "learning_rate": 4.214247537593163e-08, "loss": 0.77762604, "num_input_tokens_seen": 168348795, "step": 7788, "time_per_iteration": 2.478663206100464 }, { "auxiliary_loss_clip": 0.01140839, "auxiliary_loss_mlp": 0.01020413, "balance_loss_clip": 1.04359901, "balance_loss_mlp": 1.01376081, "epoch": 0.9365718751878795, "flos": 20703762356160.0, "grad_norm": 2.4402134185271556, "language_loss": 0.80387038, "learning_rate": 4.1983556273309293e-08, "loss": 0.8254829, "num_input_tokens_seen": 168367545, "step": 7789, "time_per_iteration": 2.542961359024048 }, { "auxiliary_loss_clip": 0.0116924, "auxiliary_loss_mlp": 0.01027754, "balance_loss_clip": 1.04755783, "balance_loss_mlp": 1.02000546, "epoch": 0.9366921180785186, "flos": 18655273035840.0, "grad_norm": 4.79102184299383, "language_loss": 0.68965733, "learning_rate": 4.182493420022526e-08, "loss": 0.7116273, "num_input_tokens_seen": 168383215, "step": 7790, "time_per_iteration": 2.415898561477661 }, { "auxiliary_loss_clip": 0.0112321, "auxiliary_loss_mlp": 0.01026642, "balance_loss_clip": 1.04398298, "balance_loss_mlp": 1.01950741, "epoch": 0.9368123609691577, "flos": 25774959720000.0, "grad_norm": 2.006395940395128, "language_loss": 0.78515321, "learning_rate": 4.166660918074139e-08, "loss": 0.80665171, "num_input_tokens_seen": 168403120, "step": 7791, "time_per_iteration": 2.6117818355560303 }, { "auxiliary_loss_clip": 0.01120518, "auxiliary_loss_mlp": 0.01026772, "balance_loss_clip": 1.04303312, "balance_loss_mlp": 1.0197711, "epoch": 0.9369326038597968, "flos": 25553390406720.0, "grad_norm": 1.440754686721723, "language_loss": 0.73534787, "learning_rate": 4.15085812388758e-08, "loss": 0.7568208, "num_input_tokens_seen": 168425340, "step": 7792, "time_per_iteration": 2.584601879119873 }, { "auxiliary_loss_clip": 0.01139523, "auxiliary_loss_mlp": 0.01024289, "balance_loss_clip": 1.04546678, "balance_loss_mlp": 1.01741624, "epoch": 0.9370528467504359, "flos": 23220022949280.0, "grad_norm": 1.6708755238955448, "language_loss": 0.78747189, "learning_rate": 4.135085039860153e-08, "loss": 0.80911005, "num_input_tokens_seen": 168444740, "step": 7793, "time_per_iteration": 2.542445659637451 }, { "auxiliary_loss_clip": 0.01138582, "auxiliary_loss_mlp": 0.01024099, "balance_loss_clip": 1.04640424, "balance_loss_mlp": 1.01684487, "epoch": 0.9371730896410749, "flos": 24967471500480.0, "grad_norm": 3.030601046978546, "language_loss": 0.78814381, "learning_rate": 4.1193416683845906e-08, "loss": 0.80977058, "num_input_tokens_seen": 168463670, "step": 7794, "time_per_iteration": 2.531111001968384 }, { "auxiliary_loss_clip": 0.0112817, "auxiliary_loss_mlp": 0.01024378, "balance_loss_clip": 1.04553676, "balance_loss_mlp": 1.01850677, "epoch": 0.9372933325317141, "flos": 15553093536480.0, "grad_norm": 2.2080326360575997, "language_loss": 0.83419955, "learning_rate": 4.103628011849136e-08, "loss": 0.85572493, "num_input_tokens_seen": 168479030, "step": 7795, "time_per_iteration": 3.282008171081543 }, { "auxiliary_loss_clip": 0.01142483, "auxiliary_loss_mlp": 0.01026385, "balance_loss_clip": 1.04622507, "balance_loss_mlp": 1.0194118, "epoch": 0.9374135754223532, "flos": 21871864804800.0, "grad_norm": 1.804351190948223, "language_loss": 0.75935906, "learning_rate": 4.0879440726375506e-08, "loss": 0.78104776, "num_input_tokens_seen": 168496815, "step": 7796, "time_per_iteration": 2.5046210289001465 }, { "auxiliary_loss_clip": 0.01135778, "auxiliary_loss_mlp": 0.01023208, "balance_loss_clip": 1.0407002, "balance_loss_mlp": 1.01612973, "epoch": 0.9375338183129922, "flos": 22631051101440.0, "grad_norm": 2.8230340683286244, "language_loss": 0.56554246, "learning_rate": 4.0722898531291074e-08, "loss": 0.58713233, "num_input_tokens_seen": 168514055, "step": 7797, "time_per_iteration": 2.5264687538146973 }, { "auxiliary_loss_clip": 0.01146494, "auxiliary_loss_mlp": 0.01025341, "balance_loss_clip": 1.04690194, "balance_loss_mlp": 1.01860595, "epoch": 0.9376540612036314, "flos": 26104296665280.0, "grad_norm": 2.0283463524799257, "language_loss": 0.76726937, "learning_rate": 4.0566653556985295e-08, "loss": 0.7889877, "num_input_tokens_seen": 168534600, "step": 7798, "time_per_iteration": 2.5465402603149414 }, { "auxiliary_loss_clip": 0.01089579, "auxiliary_loss_mlp": 0.01026321, "balance_loss_clip": 1.04221201, "balance_loss_mlp": 1.01887608, "epoch": 0.9377743040942704, "flos": 19717583285280.0, "grad_norm": 2.144343625076485, "language_loss": 0.81533116, "learning_rate": 4.0410705827159886e-08, "loss": 0.83649015, "num_input_tokens_seen": 168551895, "step": 7799, "time_per_iteration": 2.6398558616638184 }, { "auxiliary_loss_clip": 0.01133941, "auxiliary_loss_mlp": 0.01023287, "balance_loss_clip": 1.04043329, "balance_loss_mlp": 1.01644766, "epoch": 0.9378945469849095, "flos": 15267532977120.0, "grad_norm": 2.16696606671986, "language_loss": 0.70824766, "learning_rate": 4.0255055365472356e-08, "loss": 0.72982001, "num_input_tokens_seen": 168569990, "step": 7800, "time_per_iteration": 2.502209424972534 }, { "auxiliary_loss_clip": 0.01095712, "auxiliary_loss_mlp": 0.01032444, "balance_loss_clip": 1.03647017, "balance_loss_mlp": 1.0255394, "epoch": 0.9380147898755486, "flos": 20591397353280.0, "grad_norm": 2.446650653915767, "language_loss": 0.74927807, "learning_rate": 4.009970219553471e-08, "loss": 0.77055967, "num_input_tokens_seen": 168586940, "step": 7801, "time_per_iteration": 2.6105599403381348 }, { "auxiliary_loss_clip": 0.01157387, "auxiliary_loss_mlp": 0.01025485, "balance_loss_clip": 1.04629254, "balance_loss_mlp": 1.01829994, "epoch": 0.9381350327661877, "flos": 26281119834720.0, "grad_norm": 2.436703691467478, "language_loss": 0.76811266, "learning_rate": 3.99446463409141e-08, "loss": 0.78994137, "num_input_tokens_seen": 168604795, "step": 7802, "time_per_iteration": 2.540095329284668 }, { "auxiliary_loss_clip": 0.01156871, "auxiliary_loss_mlp": 0.01026034, "balance_loss_clip": 1.04414356, "balance_loss_mlp": 1.01900649, "epoch": 0.9382552756568268, "flos": 23586347975040.0, "grad_norm": 2.1374421131681554, "language_loss": 0.68953145, "learning_rate": 3.978988782513215e-08, "loss": 0.71136051, "num_input_tokens_seen": 168622290, "step": 7803, "time_per_iteration": 3.264599323272705 }, { "auxiliary_loss_clip": 0.01157805, "auxiliary_loss_mlp": 0.01023175, "balance_loss_clip": 1.04599941, "balance_loss_mlp": 1.01593876, "epoch": 0.9383755185474659, "flos": 28438813465440.0, "grad_norm": 1.6908957163546978, "language_loss": 0.76393104, "learning_rate": 3.963542667166586e-08, "loss": 0.78574085, "num_input_tokens_seen": 168642395, "step": 7804, "time_per_iteration": 2.546901226043701 }, { "auxiliary_loss_clip": 0.01124206, "auxiliary_loss_mlp": 0.01024155, "balance_loss_clip": 1.04618347, "balance_loss_mlp": 1.01767576, "epoch": 0.938495761438105, "flos": 20449586831520.0, "grad_norm": 1.6534791963019153, "language_loss": 0.68273139, "learning_rate": 3.9481262903946486e-08, "loss": 0.70421505, "num_input_tokens_seen": 168661840, "step": 7805, "time_per_iteration": 2.5301895141601562 }, { "auxiliary_loss_clip": 0.01021039, "auxiliary_loss_mlp": 0.0100297, "balance_loss_clip": 1.00536633, "balance_loss_mlp": 1.00173593, "epoch": 0.938616004328744, "flos": 69302723062080.0, "grad_norm": 0.7800678742064493, "language_loss": 0.54559875, "learning_rate": 3.932739654536066e-08, "loss": 0.56583881, "num_input_tokens_seen": 168724540, "step": 7806, "time_per_iteration": 3.16611385345459 }, { "auxiliary_loss_clip": 0.01151253, "auxiliary_loss_mlp": 0.01023242, "balance_loss_clip": 1.04687047, "balance_loss_mlp": 1.01651561, "epoch": 0.9387362472193832, "flos": 18911639495040.0, "grad_norm": 4.350420290435778, "language_loss": 0.73993242, "learning_rate": 3.917382761925014e-08, "loss": 0.76167738, "num_input_tokens_seen": 168740375, "step": 7807, "time_per_iteration": 2.463046073913574 }, { "auxiliary_loss_clip": 0.01148382, "auxiliary_loss_mlp": 0.01022693, "balance_loss_clip": 1.04534769, "balance_loss_mlp": 1.01573122, "epoch": 0.9388564901100223, "flos": 26501970808800.0, "grad_norm": 1.7460628290680542, "language_loss": 0.79088008, "learning_rate": 3.9020556148910754e-08, "loss": 0.81259084, "num_input_tokens_seen": 168759730, "step": 7808, "time_per_iteration": 2.534616231918335 }, { "auxiliary_loss_clip": 0.01042664, "auxiliary_loss_mlp": 0.01000865, "balance_loss_clip": 1.00599957, "balance_loss_mlp": 0.99960703, "epoch": 0.9389767330006613, "flos": 58941093623520.0, "grad_norm": 0.7074721767315046, "language_loss": 0.56694627, "learning_rate": 3.8867582157593895e-08, "loss": 0.58738154, "num_input_tokens_seen": 168813935, "step": 7809, "time_per_iteration": 3.7183427810668945 }, { "auxiliary_loss_clip": 0.01153647, "auxiliary_loss_mlp": 0.01026504, "balance_loss_clip": 1.04820395, "balance_loss_mlp": 1.01972127, "epoch": 0.9390969758913005, "flos": 31102559460000.0, "grad_norm": 1.8404552787312354, "language_loss": 0.76590347, "learning_rate": 3.871490566850544e-08, "loss": 0.787705, "num_input_tokens_seen": 168838145, "step": 7810, "time_per_iteration": 3.270310163497925 }, { "auxiliary_loss_clip": 0.01133815, "auxiliary_loss_mlp": 0.01023186, "balance_loss_clip": 1.04296601, "balance_loss_mlp": 1.01620674, "epoch": 0.9392172187819395, "flos": 22419394869120.0, "grad_norm": 1.7234236780574275, "language_loss": 0.70618773, "learning_rate": 3.856252670480642e-08, "loss": 0.72775781, "num_input_tokens_seen": 168856805, "step": 7811, "time_per_iteration": 2.5640029907226562 }, { "auxiliary_loss_clip": 0.01134547, "auxiliary_loss_mlp": 0.01025717, "balance_loss_clip": 1.04006147, "balance_loss_mlp": 1.01875544, "epoch": 0.9393374616725786, "flos": 19719486884160.0, "grad_norm": 1.7706709713980018, "language_loss": 0.81293154, "learning_rate": 3.841044528961279e-08, "loss": 0.83453417, "num_input_tokens_seen": 168874600, "step": 7812, "time_per_iteration": 2.5139617919921875 }, { "auxiliary_loss_clip": 0.01167395, "auxiliary_loss_mlp": 0.01029564, "balance_loss_clip": 1.04507947, "balance_loss_mlp": 1.02259278, "epoch": 0.9394577045632178, "flos": 24170219614560.0, "grad_norm": 2.0105877650453423, "language_loss": 0.78816873, "learning_rate": 3.825866144599477e-08, "loss": 0.81013834, "num_input_tokens_seen": 168893655, "step": 7813, "time_per_iteration": 2.503263473510742 }, { "auxiliary_loss_clip": 0.01138134, "auxiliary_loss_mlp": 0.0102329, "balance_loss_clip": 1.04233122, "balance_loss_mlp": 1.01607513, "epoch": 0.9395779474538568, "flos": 19023932664000.0, "grad_norm": 6.330766061128162, "language_loss": 0.75583756, "learning_rate": 3.8107175196978145e-08, "loss": 0.77745175, "num_input_tokens_seen": 168909960, "step": 7814, "time_per_iteration": 2.474400281906128 }, { "auxiliary_loss_clip": 0.01125033, "auxiliary_loss_mlp": 0.0102194, "balance_loss_clip": 1.04438114, "balance_loss_mlp": 1.01558316, "epoch": 0.9396981903444959, "flos": 14319132159840.0, "grad_norm": 1.9427114607871216, "language_loss": 0.76437122, "learning_rate": 3.7955986565542996e-08, "loss": 0.78584099, "num_input_tokens_seen": 168928040, "step": 7815, "time_per_iteration": 2.538655996322632 }, { "auxiliary_loss_clip": 0.01124836, "auxiliary_loss_mlp": 0.01022287, "balance_loss_clip": 1.04226375, "balance_loss_mlp": 1.0156436, "epoch": 0.9398184332351349, "flos": 34787569007040.0, "grad_norm": 1.9588329692327733, "language_loss": 0.68623799, "learning_rate": 3.780509557462497e-08, "loss": 0.70770919, "num_input_tokens_seen": 168948240, "step": 7816, "time_per_iteration": 2.653881549835205 }, { "auxiliary_loss_clip": 0.01139651, "auxiliary_loss_mlp": 0.01026533, "balance_loss_clip": 1.04439902, "balance_loss_mlp": 1.01913261, "epoch": 0.9399386761257741, "flos": 25372257202080.0, "grad_norm": 5.8282561910851145, "language_loss": 0.75824004, "learning_rate": 3.765450224711375e-08, "loss": 0.77990186, "num_input_tokens_seen": 168968745, "step": 7817, "time_per_iteration": 2.578463315963745 }, { "auxiliary_loss_clip": 0.01138845, "auxiliary_loss_mlp": 0.01025469, "balance_loss_clip": 1.04745257, "balance_loss_mlp": 1.01826322, "epoch": 0.9400589190164131, "flos": 27304969408320.0, "grad_norm": 1.875425004461104, "language_loss": 0.79610932, "learning_rate": 3.750420660585396e-08, "loss": 0.81775248, "num_input_tokens_seen": 168990685, "step": 7818, "time_per_iteration": 2.565119504928589 }, { "auxiliary_loss_clip": 0.01165929, "auxiliary_loss_mlp": 0.01026341, "balance_loss_clip": 1.04844141, "balance_loss_mlp": 1.02005553, "epoch": 0.9401791619070522, "flos": 23399863143360.0, "grad_norm": 1.9689015400094199, "language_loss": 0.79722047, "learning_rate": 3.735420867364603e-08, "loss": 0.81914324, "num_input_tokens_seen": 169011665, "step": 7819, "time_per_iteration": 2.4741058349609375 }, { "auxiliary_loss_clip": 0.01091156, "auxiliary_loss_mlp": 0.01019085, "balance_loss_clip": 1.03848279, "balance_loss_mlp": 1.01257658, "epoch": 0.9402994047976914, "flos": 35881407959040.0, "grad_norm": 3.771955664090144, "language_loss": 0.61583853, "learning_rate": 3.7204508473244186e-08, "loss": 0.63694096, "num_input_tokens_seen": 169035290, "step": 7820, "time_per_iteration": 2.738109588623047 }, { "auxiliary_loss_clip": 0.01079673, "auxiliary_loss_mlp": 0.01023326, "balance_loss_clip": 1.04046607, "balance_loss_mlp": 1.01664114, "epoch": 0.9404196476883304, "flos": 22236824986080.0, "grad_norm": 1.5836049382895694, "language_loss": 0.69095767, "learning_rate": 3.7055106027357395e-08, "loss": 0.71198761, "num_input_tokens_seen": 169055155, "step": 7821, "time_per_iteration": 3.5198941230773926 }, { "auxiliary_loss_clip": 0.01149047, "auxiliary_loss_mlp": 0.01025079, "balance_loss_clip": 1.04503822, "balance_loss_mlp": 1.01753008, "epoch": 0.9405398905789695, "flos": 18915805862400.0, "grad_norm": 2.3890052437715688, "language_loss": 0.71679282, "learning_rate": 3.690600135865063e-08, "loss": 0.73853409, "num_input_tokens_seen": 169072080, "step": 7822, "time_per_iteration": 2.851017951965332 }, { "auxiliary_loss_clip": 0.01014813, "auxiliary_loss_mlp": 0.0100159, "balance_loss_clip": 1.00626433, "balance_loss_mlp": 1.00031471, "epoch": 0.9406601334696086, "flos": 70274140262400.0, "grad_norm": 0.783545808874998, "language_loss": 0.58107674, "learning_rate": 3.675719448974246e-08, "loss": 0.60124075, "num_input_tokens_seen": 169137170, "step": 7823, "time_per_iteration": 3.252366542816162 }, { "auxiliary_loss_clip": 0.01107607, "auxiliary_loss_mlp": 0.00761872, "balance_loss_clip": 1.04258096, "balance_loss_mlp": 1.00042391, "epoch": 0.9407803763602477, "flos": 22165075676640.0, "grad_norm": 2.97164316629536, "language_loss": 0.6014477, "learning_rate": 3.6608685443207054e-08, "loss": 0.62014246, "num_input_tokens_seen": 169156320, "step": 7824, "time_per_iteration": 2.6294548511505127 }, { "auxiliary_loss_clip": 0.01127667, "auxiliary_loss_mlp": 0.01023799, "balance_loss_clip": 1.04296112, "balance_loss_mlp": 1.01721859, "epoch": 0.9409006192508867, "flos": 18879500204160.0, "grad_norm": 2.384856651962812, "language_loss": 0.66758001, "learning_rate": 3.646047424157306e-08, "loss": 0.68909466, "num_input_tokens_seen": 169173295, "step": 7825, "time_per_iteration": 2.537014961242676 }, { "auxiliary_loss_clip": 0.01140728, "auxiliary_loss_mlp": 0.0102732, "balance_loss_clip": 1.04730678, "balance_loss_mlp": 1.01995313, "epoch": 0.9410208621415259, "flos": 23368262606880.0, "grad_norm": 2.276716982232451, "language_loss": 0.68670017, "learning_rate": 3.631256090732382e-08, "loss": 0.70838058, "num_input_tokens_seen": 169193755, "step": 7826, "time_per_iteration": 2.536181926727295 }, { "auxiliary_loss_clip": 0.01126618, "auxiliary_loss_mlp": 0.01022096, "balance_loss_clip": 1.04400277, "balance_loss_mlp": 1.01564622, "epoch": 0.941141105032165, "flos": 22742230844640.0, "grad_norm": 1.562388758895508, "language_loss": 0.82882881, "learning_rate": 3.6164945462897833e-08, "loss": 0.85031599, "num_input_tokens_seen": 169213045, "step": 7827, "time_per_iteration": 2.558441400527954 }, { "auxiliary_loss_clip": 0.01149552, "auxiliary_loss_mlp": 0.00761422, "balance_loss_clip": 1.04537356, "balance_loss_mlp": 1.00049078, "epoch": 0.941261347922804, "flos": 20704911698880.0, "grad_norm": 1.6219404648615772, "language_loss": 0.75492704, "learning_rate": 3.6017627930687856e-08, "loss": 0.77403677, "num_input_tokens_seen": 169232870, "step": 7828, "time_per_iteration": 2.519761323928833 }, { "auxiliary_loss_clip": 0.01106157, "auxiliary_loss_mlp": 0.01022722, "balance_loss_clip": 1.03872907, "balance_loss_mlp": 1.01576638, "epoch": 0.9413815908134432, "flos": 19422001894080.0, "grad_norm": 2.0081291375443717, "language_loss": 0.7722466, "learning_rate": 3.587060833304267e-08, "loss": 0.79353535, "num_input_tokens_seen": 169251060, "step": 7829, "time_per_iteration": 3.3537259101867676 }, { "auxiliary_loss_clip": 0.01155691, "auxiliary_loss_mlp": 0.01024247, "balance_loss_clip": 1.04639316, "balance_loss_mlp": 1.01729155, "epoch": 0.9415018337040822, "flos": 17493456055200.0, "grad_norm": 1.9814679018364159, "language_loss": 0.63837552, "learning_rate": 3.5723886692264225e-08, "loss": 0.66017485, "num_input_tokens_seen": 169268600, "step": 7830, "time_per_iteration": 2.482638120651245 }, { "auxiliary_loss_clip": 0.01136298, "auxiliary_loss_mlp": 0.01025455, "balance_loss_clip": 1.0419476, "balance_loss_mlp": 1.01815367, "epoch": 0.9416220765947213, "flos": 31831617815520.0, "grad_norm": 2.3035967575804683, "language_loss": 0.61613542, "learning_rate": 3.557746303061071e-08, "loss": 0.63775301, "num_input_tokens_seen": 169290355, "step": 7831, "time_per_iteration": 2.6345913410186768 }, { "auxiliary_loss_clip": 0.01134902, "auxiliary_loss_mlp": 0.01025853, "balance_loss_clip": 1.04197192, "balance_loss_mlp": 1.01935053, "epoch": 0.9417423194853605, "flos": 23511976727520.0, "grad_norm": 1.6298560811695715, "language_loss": 0.72380292, "learning_rate": 3.543133737029391e-08, "loss": 0.74541044, "num_input_tokens_seen": 169310865, "step": 7832, "time_per_iteration": 2.590444326400757 }, { "auxiliary_loss_clip": 0.01153777, "auxiliary_loss_mlp": 0.01025718, "balance_loss_clip": 1.04439747, "balance_loss_mlp": 1.01887786, "epoch": 0.9418625623759995, "flos": 23915110248960.0, "grad_norm": 1.87667999972396, "language_loss": 0.6910094, "learning_rate": 3.5285509733481214e-08, "loss": 0.71280438, "num_input_tokens_seen": 169330590, "step": 7833, "time_per_iteration": 2.5126800537109375 }, { "auxiliary_loss_clip": 0.01148901, "auxiliary_loss_mlp": 0.01026161, "balance_loss_clip": 1.0442158, "balance_loss_mlp": 1.01942825, "epoch": 0.9419828052666386, "flos": 18076968525120.0, "grad_norm": 1.659277980297444, "language_loss": 0.76576203, "learning_rate": 3.513998014229469e-08, "loss": 0.78751266, "num_input_tokens_seen": 169349540, "step": 7834, "time_per_iteration": 3.4574954509735107 }, { "auxiliary_loss_clip": 0.01137276, "auxiliary_loss_mlp": 0.0102347, "balance_loss_clip": 1.04636574, "balance_loss_mlp": 1.01726198, "epoch": 0.9421030481572777, "flos": 17712331596480.0, "grad_norm": 2.3348281232100314, "language_loss": 0.86268127, "learning_rate": 3.499474861881069e-08, "loss": 0.88428873, "num_input_tokens_seen": 169366765, "step": 7835, "time_per_iteration": 2.503145456314087 }, { "auxiliary_loss_clip": 0.01098638, "auxiliary_loss_mlp": 0.01026636, "balance_loss_clip": 1.04217172, "balance_loss_mlp": 1.0198884, "epoch": 0.9422232910479168, "flos": 20194118296320.0, "grad_norm": 1.892183328401754, "language_loss": 0.67668301, "learning_rate": 3.4849815185061136e-08, "loss": 0.69793576, "num_input_tokens_seen": 169386655, "step": 7836, "time_per_iteration": 3.3876895904541016 }, { "auxiliary_loss_clip": 0.01150857, "auxiliary_loss_mlp": 0.01022995, "balance_loss_clip": 1.04374909, "balance_loss_mlp": 1.01669455, "epoch": 0.9423435339385559, "flos": 18442575211680.0, "grad_norm": 2.1118169522183434, "language_loss": 0.76117575, "learning_rate": 3.470517986303223e-08, "loss": 0.78291428, "num_input_tokens_seen": 169405640, "step": 7837, "time_per_iteration": 2.4898364543914795 }, { "auxiliary_loss_clip": 0.01125893, "auxiliary_loss_mlp": 0.01027364, "balance_loss_clip": 1.04589105, "balance_loss_mlp": 1.02080178, "epoch": 0.942463776829195, "flos": 20080639867680.0, "grad_norm": 1.9012085204189744, "language_loss": 0.79450041, "learning_rate": 3.4560842674664856e-08, "loss": 0.81603301, "num_input_tokens_seen": 169424155, "step": 7838, "time_per_iteration": 2.5382330417633057 }, { "auxiliary_loss_clip": 0.01151979, "auxiliary_loss_mlp": 0.01022535, "balance_loss_clip": 1.04259253, "balance_loss_mlp": 1.01594257, "epoch": 0.9425840197198341, "flos": 22636259060640.0, "grad_norm": 1.7974966441706075, "language_loss": 0.7565152, "learning_rate": 3.441680364185506e-08, "loss": 0.77826029, "num_input_tokens_seen": 169444025, "step": 7839, "time_per_iteration": 2.5313498973846436 }, { "auxiliary_loss_clip": 0.0114264, "auxiliary_loss_mlp": 0.01029281, "balance_loss_clip": 1.04609823, "balance_loss_mlp": 1.02200353, "epoch": 0.9427042626104731, "flos": 19937895504960.0, "grad_norm": 2.018358619439503, "language_loss": 0.74658108, "learning_rate": 3.427306278645314e-08, "loss": 0.76830029, "num_input_tokens_seen": 169462480, "step": 7840, "time_per_iteration": 2.4956820011138916 }, { "auxiliary_loss_clip": 0.01112585, "auxiliary_loss_mlp": 0.01026653, "balance_loss_clip": 1.04374909, "balance_loss_mlp": 1.01976275, "epoch": 0.9428245055011123, "flos": 22857002283840.0, "grad_norm": 1.9064472803594217, "language_loss": 0.7270391, "learning_rate": 3.4129620130264767e-08, "loss": 0.7484315, "num_input_tokens_seen": 169480840, "step": 7841, "time_per_iteration": 2.5866901874542236 }, { "auxiliary_loss_clip": 0.01142949, "auxiliary_loss_mlp": 0.00761343, "balance_loss_clip": 1.04593575, "balance_loss_mlp": 1.0005275, "epoch": 0.9429447483917514, "flos": 20951760163680.0, "grad_norm": 2.8116321860304256, "language_loss": 0.7780993, "learning_rate": 3.398647569505009e-08, "loss": 0.79714227, "num_input_tokens_seen": 169498265, "step": 7842, "time_per_iteration": 2.501697301864624 }, { "auxiliary_loss_clip": 0.01131499, "auxiliary_loss_mlp": 0.01023943, "balance_loss_clip": 1.04542434, "balance_loss_mlp": 1.01657856, "epoch": 0.9430649912823904, "flos": 18843661466400.0, "grad_norm": 2.465645778824866, "language_loss": 0.74864268, "learning_rate": 3.384362950252373e-08, "loss": 0.77019715, "num_input_tokens_seen": 169515235, "step": 7843, "time_per_iteration": 2.5456676483154297 }, { "auxiliary_loss_clip": 0.01135652, "auxiliary_loss_mlp": 0.01018446, "balance_loss_clip": 1.04218268, "balance_loss_mlp": 1.01165986, "epoch": 0.9431852341730296, "flos": 32556042883200.0, "grad_norm": 2.147435034507851, "language_loss": 0.56425691, "learning_rate": 3.3701081574355473e-08, "loss": 0.58579791, "num_input_tokens_seen": 169537195, "step": 7844, "time_per_iteration": 2.60250186920166 }, { "auxiliary_loss_clip": 0.01037121, "auxiliary_loss_mlp": 0.01000698, "balance_loss_clip": 1.00577152, "balance_loss_mlp": 0.99944609, "epoch": 0.9433054770636686, "flos": 66904502351520.0, "grad_norm": 0.6401227261861729, "language_loss": 0.51703632, "learning_rate": 3.3558831932169796e-08, "loss": 0.53741449, "num_input_tokens_seen": 169605865, "step": 7845, "time_per_iteration": 3.192800283432007 }, { "auxiliary_loss_clip": 0.01153133, "auxiliary_loss_mlp": 0.01024559, "balance_loss_clip": 1.04708934, "balance_loss_mlp": 1.01772499, "epoch": 0.9434257199543077, "flos": 26140350904800.0, "grad_norm": 1.8613845606992663, "language_loss": 0.88687009, "learning_rate": 3.341688059754588e-08, "loss": 0.90864694, "num_input_tokens_seen": 169621520, "step": 7846, "time_per_iteration": 2.556575298309326 }, { "auxiliary_loss_clip": 0.01132672, "auxiliary_loss_mlp": 0.00761675, "balance_loss_clip": 1.04191113, "balance_loss_mlp": 1.00047183, "epoch": 0.9435459628449467, "flos": 25003489823040.0, "grad_norm": 2.0911113333876794, "language_loss": 0.77756166, "learning_rate": 3.327522759201762e-08, "loss": 0.79650509, "num_input_tokens_seen": 169641390, "step": 7847, "time_per_iteration": 3.2986602783203125 }, { "auxiliary_loss_clip": 0.01127012, "auxiliary_loss_mlp": 0.0102422, "balance_loss_clip": 1.04533458, "balance_loss_mlp": 1.01701629, "epoch": 0.9436662057355859, "flos": 22163243911680.0, "grad_norm": 2.3384194494802935, "language_loss": 0.66594529, "learning_rate": 3.313387293707359e-08, "loss": 0.68745756, "num_input_tokens_seen": 169660095, "step": 7848, "time_per_iteration": 2.5776846408843994 }, { "auxiliary_loss_clip": 0.01122719, "auxiliary_loss_mlp": 0.01030396, "balance_loss_clip": 1.04466927, "balance_loss_mlp": 1.02303505, "epoch": 0.943786448626225, "flos": 20118525872160.0, "grad_norm": 1.8288774044618508, "language_loss": 0.68317628, "learning_rate": 3.29928166541571e-08, "loss": 0.70470738, "num_input_tokens_seen": 169679050, "step": 7849, "time_per_iteration": 2.5743703842163086 }, { "auxiliary_loss_clip": 0.01130336, "auxiliary_loss_mlp": 0.01024535, "balance_loss_clip": 1.04372334, "balance_loss_mlp": 1.01780283, "epoch": 0.943906691516864, "flos": 22090812180000.0, "grad_norm": 3.017799575717904, "language_loss": 0.80104214, "learning_rate": 3.2852058764666346e-08, "loss": 0.82259089, "num_input_tokens_seen": 169698150, "step": 7850, "time_per_iteration": 2.5296790599823 }, { "auxiliary_loss_clip": 0.01115672, "auxiliary_loss_mlp": 0.01026901, "balance_loss_clip": 1.04534435, "balance_loss_mlp": 1.02035618, "epoch": 0.9440269344075032, "flos": 35298506077440.0, "grad_norm": 1.745394289878013, "language_loss": 0.68516964, "learning_rate": 3.2711599289954264e-08, "loss": 0.70659542, "num_input_tokens_seen": 169722185, "step": 7851, "time_per_iteration": 2.6601579189300537 }, { "auxiliary_loss_clip": 0.01100894, "auxiliary_loss_mlp": 0.01025476, "balance_loss_clip": 1.04316306, "balance_loss_mlp": 1.01881194, "epoch": 0.9441471772981422, "flos": 19238139000480.0, "grad_norm": 1.9142121126653604, "language_loss": 0.77613235, "learning_rate": 3.257143825132847e-08, "loss": 0.79739606, "num_input_tokens_seen": 169740355, "step": 7852, "time_per_iteration": 2.622337818145752 }, { "auxiliary_loss_clip": 0.01138838, "auxiliary_loss_mlp": 0.01022821, "balance_loss_clip": 1.04417431, "balance_loss_mlp": 1.01602936, "epoch": 0.9442674201887813, "flos": 25739803404480.0, "grad_norm": 2.047979827698591, "language_loss": 0.76024437, "learning_rate": 3.243157567005106e-08, "loss": 0.78186095, "num_input_tokens_seen": 169758535, "step": 7853, "time_per_iteration": 2.5529911518096924 }, { "auxiliary_loss_clip": 0.01171793, "auxiliary_loss_mlp": 0.01031233, "balance_loss_clip": 1.05024898, "balance_loss_mlp": 1.0238806, "epoch": 0.9443876630794205, "flos": 15523324764960.0, "grad_norm": 2.236211432373456, "language_loss": 0.63990128, "learning_rate": 3.2292011567339296e-08, "loss": 0.66193151, "num_input_tokens_seen": 169776340, "step": 7854, "time_per_iteration": 2.450807809829712 }, { "auxiliary_loss_clip": 0.01154178, "auxiliary_loss_mlp": 0.00761492, "balance_loss_clip": 1.04490447, "balance_loss_mlp": 1.00055599, "epoch": 0.9445079059700595, "flos": 13400823366720.0, "grad_norm": 2.119864940305021, "language_loss": 0.55374932, "learning_rate": 3.21527459643649e-08, "loss": 0.57290596, "num_input_tokens_seen": 169793225, "step": 7855, "time_per_iteration": 3.241607427597046 }, { "auxiliary_loss_clip": 0.01156605, "auxiliary_loss_mlp": 0.01022978, "balance_loss_clip": 1.04720521, "balance_loss_mlp": 1.01564682, "epoch": 0.9446281488606986, "flos": 23659246627200.0, "grad_norm": 1.9136788825665971, "language_loss": 0.74032211, "learning_rate": 3.2013778882254536e-08, "loss": 0.76211786, "num_input_tokens_seen": 169812020, "step": 7856, "time_per_iteration": 2.4892168045043945 }, { "auxiliary_loss_clip": 0.01142693, "auxiliary_loss_mlp": 0.01025384, "balance_loss_clip": 1.04284489, "balance_loss_mlp": 1.01849699, "epoch": 0.9447483917513377, "flos": 25557341272320.0, "grad_norm": 4.606491967590348, "language_loss": 0.75547743, "learning_rate": 3.1875110342088676e-08, "loss": 0.7771582, "num_input_tokens_seen": 169833470, "step": 7857, "time_per_iteration": 2.5281686782836914 }, { "auxiliary_loss_clip": 0.01131469, "auxiliary_loss_mlp": 0.01027982, "balance_loss_clip": 1.04317331, "balance_loss_mlp": 1.0213542, "epoch": 0.9448686346419768, "flos": 24535467131520.0, "grad_norm": 1.8232158471640265, "language_loss": 0.65657705, "learning_rate": 3.1736740364904035e-08, "loss": 0.67817163, "num_input_tokens_seen": 169854000, "step": 7858, "time_per_iteration": 2.5363125801086426 }, { "auxiliary_loss_clip": 0.01107514, "auxiliary_loss_mlp": 0.00761542, "balance_loss_clip": 1.04132295, "balance_loss_mlp": 1.00050247, "epoch": 0.9449888775326158, "flos": 14721260006400.0, "grad_norm": 2.1764315319657834, "language_loss": 0.77497792, "learning_rate": 3.159866897169094e-08, "loss": 0.79366851, "num_input_tokens_seen": 169872200, "step": 7859, "time_per_iteration": 2.578937530517578 }, { "auxiliary_loss_clip": 0.01129146, "auxiliary_loss_mlp": 0.0102288, "balance_loss_clip": 1.04443014, "balance_loss_mlp": 1.01569772, "epoch": 0.945109120423255, "flos": 15447876008640.0, "grad_norm": 1.7291348993513744, "language_loss": 0.75465775, "learning_rate": 3.146089618339487e-08, "loss": 0.776178, "num_input_tokens_seen": 169889055, "step": 7860, "time_per_iteration": 3.255439043045044 }, { "auxiliary_loss_clip": 0.01127055, "auxiliary_loss_mlp": 0.01027671, "balance_loss_clip": 1.04361868, "balance_loss_mlp": 1.02051306, "epoch": 0.9452293633138941, "flos": 25448101044960.0, "grad_norm": 1.8683732817759406, "language_loss": 0.67733431, "learning_rate": 3.132342202091554e-08, "loss": 0.69888157, "num_input_tokens_seen": 169909280, "step": 7861, "time_per_iteration": 2.592453956604004 }, { "auxiliary_loss_clip": 0.01168712, "auxiliary_loss_mlp": 0.01028363, "balance_loss_clip": 1.04738736, "balance_loss_mlp": 1.02091789, "epoch": 0.9453496062045331, "flos": 21215345931840.0, "grad_norm": 1.9855621817174516, "language_loss": 0.68341255, "learning_rate": 3.1186246505107595e-08, "loss": 0.70538336, "num_input_tokens_seen": 169928420, "step": 7862, "time_per_iteration": 3.1993494033813477 }, { "auxiliary_loss_clip": 0.01154229, "auxiliary_loss_mlp": 0.01026347, "balance_loss_clip": 1.04821539, "balance_loss_mlp": 1.01916718, "epoch": 0.9454698490951723, "flos": 20010901908000.0, "grad_norm": 1.9699300544757634, "language_loss": 0.83745962, "learning_rate": 3.104936965678084e-08, "loss": 0.85926533, "num_input_tokens_seen": 169946750, "step": 7863, "time_per_iteration": 2.4813995361328125 }, { "auxiliary_loss_clip": 0.01152916, "auxiliary_loss_mlp": 0.01028555, "balance_loss_clip": 1.04437041, "balance_loss_mlp": 1.02182591, "epoch": 0.9455900919858113, "flos": 21069656378400.0, "grad_norm": 2.022382345311575, "language_loss": 0.81778234, "learning_rate": 3.091279149669956e-08, "loss": 0.83959711, "num_input_tokens_seen": 169965540, "step": 7864, "time_per_iteration": 2.477106809616089 }, { "auxiliary_loss_clip": 0.01152271, "auxiliary_loss_mlp": 0.0076174, "balance_loss_clip": 1.04621291, "balance_loss_mlp": 1.00055182, "epoch": 0.9457103348764504, "flos": 20740858187520.0, "grad_norm": 1.995606459497886, "language_loss": 0.73790264, "learning_rate": 3.0776512045581624e-08, "loss": 0.75704271, "num_input_tokens_seen": 169984330, "step": 7865, "time_per_iteration": 2.5033388137817383 }, { "auxiliary_loss_clip": 0.01132646, "auxiliary_loss_mlp": 0.01029066, "balance_loss_clip": 1.04358816, "balance_loss_mlp": 1.02167797, "epoch": 0.9458305777670896, "flos": 21428367008640.0, "grad_norm": 1.938000839033622, "language_loss": 0.77875066, "learning_rate": 3.0640531324101384e-08, "loss": 0.80036777, "num_input_tokens_seen": 170002095, "step": 7866, "time_per_iteration": 2.4966325759887695 }, { "auxiliary_loss_clip": 0.01159015, "auxiliary_loss_mlp": 0.01026975, "balance_loss_clip": 1.05109763, "balance_loss_mlp": 1.01965594, "epoch": 0.9459508206577286, "flos": 20011189243680.0, "grad_norm": 1.6292667261686904, "language_loss": 0.76008725, "learning_rate": 3.0504849352886554e-08, "loss": 0.78194726, "num_input_tokens_seen": 170020240, "step": 7867, "time_per_iteration": 2.4770596027374268 }, { "auxiliary_loss_clip": 0.01153742, "auxiliary_loss_mlp": 0.01025501, "balance_loss_clip": 1.0471127, "balance_loss_mlp": 1.01834261, "epoch": 0.9460710635483677, "flos": 12166430986560.0, "grad_norm": 6.407092196852913, "language_loss": 0.71492094, "learning_rate": 3.036946615252023e-08, "loss": 0.73671335, "num_input_tokens_seen": 170035770, "step": 7868, "time_per_iteration": 2.44279408454895 }, { "auxiliary_loss_clip": 0.01140959, "auxiliary_loss_mlp": 0.01030766, "balance_loss_clip": 1.04468417, "balance_loss_mlp": 1.0242157, "epoch": 0.9461913064390068, "flos": 34276200933120.0, "grad_norm": 2.668716116715759, "language_loss": 0.6674701, "learning_rate": 3.0234381743539984e-08, "loss": 0.68918729, "num_input_tokens_seen": 170053385, "step": 7869, "time_per_iteration": 2.6256299018859863 }, { "auxiliary_loss_clip": 0.01144767, "auxiliary_loss_mlp": 0.01025934, "balance_loss_clip": 1.04368687, "balance_loss_mlp": 1.01911545, "epoch": 0.9463115493296459, "flos": 19463766930240.0, "grad_norm": 2.1480257021821325, "language_loss": 0.80197477, "learning_rate": 3.0099596146437863e-08, "loss": 0.82368183, "num_input_tokens_seen": 170070490, "step": 7870, "time_per_iteration": 2.544541120529175 }, { "auxiliary_loss_clip": 0.01059569, "auxiliary_loss_mlp": 0.01000705, "balance_loss_clip": 1.00556386, "balance_loss_mlp": 0.99944752, "epoch": 0.946431792220285, "flos": 70570835079360.0, "grad_norm": 0.7710394095427763, "language_loss": 0.60132498, "learning_rate": 2.996510938166086e-08, "loss": 0.62192768, "num_input_tokens_seen": 170133465, "step": 7871, "time_per_iteration": 3.130401372909546 }, { "auxiliary_loss_clip": 0.01151297, "auxiliary_loss_mlp": 0.01020956, "balance_loss_clip": 1.0463618, "balance_loss_mlp": 1.01439965, "epoch": 0.9465520351109241, "flos": 18947909236320.0, "grad_norm": 1.8308183447815405, "language_loss": 0.73585671, "learning_rate": 2.983092146960997e-08, "loss": 0.75757927, "num_input_tokens_seen": 170150810, "step": 7872, "time_per_iteration": 3.2199337482452393 }, { "auxiliary_loss_clip": 0.01141542, "auxiliary_loss_mlp": 0.01025148, "balance_loss_clip": 1.0443058, "balance_loss_mlp": 1.01785553, "epoch": 0.9466722780015632, "flos": 19135651161600.0, "grad_norm": 2.0960732828368007, "language_loss": 0.79976606, "learning_rate": 2.9697032430642256e-08, "loss": 0.82143301, "num_input_tokens_seen": 170169025, "step": 7873, "time_per_iteration": 2.503824234008789 }, { "auxiliary_loss_clip": 0.01164487, "auxiliary_loss_mlp": 0.01020418, "balance_loss_clip": 1.04655671, "balance_loss_mlp": 1.01427031, "epoch": 0.9467925208922022, "flos": 17237915686080.0, "grad_norm": 2.2225691305044433, "language_loss": 0.73585212, "learning_rate": 2.9563442285067906e-08, "loss": 0.75770122, "num_input_tokens_seen": 170186070, "step": 7874, "time_per_iteration": 2.4383544921875 }, { "auxiliary_loss_clip": 0.01155931, "auxiliary_loss_mlp": 0.0102598, "balance_loss_clip": 1.04695868, "balance_loss_mlp": 1.0188334, "epoch": 0.9469127637828414, "flos": 29169021163680.0, "grad_norm": 1.8839562792390765, "language_loss": 0.79668403, "learning_rate": 2.943015105315294e-08, "loss": 0.81850314, "num_input_tokens_seen": 170206265, "step": 7875, "time_per_iteration": 2.5303657054901123 }, { "auxiliary_loss_clip": 0.01115173, "auxiliary_loss_mlp": 0.0102434, "balance_loss_clip": 1.0429554, "balance_loss_mlp": 1.01723528, "epoch": 0.9470330066734804, "flos": 26030464172160.0, "grad_norm": 2.116135626812667, "language_loss": 0.66471773, "learning_rate": 2.929715875511718e-08, "loss": 0.68611282, "num_input_tokens_seen": 170225300, "step": 7876, "time_per_iteration": 2.6132519245147705 }, { "auxiliary_loss_clip": 0.01153661, "auxiliary_loss_mlp": 0.01023662, "balance_loss_clip": 1.04377031, "balance_loss_mlp": 1.01691484, "epoch": 0.9471532495641195, "flos": 23440909840320.0, "grad_norm": 2.032485782875067, "language_loss": 0.69727606, "learning_rate": 2.9164465411135375e-08, "loss": 0.71904933, "num_input_tokens_seen": 170245070, "step": 7877, "time_per_iteration": 2.508275032043457 }, { "auxiliary_loss_clip": 0.01158427, "auxiliary_loss_mlp": 0.01028128, "balance_loss_clip": 1.05076277, "balance_loss_mlp": 1.02127957, "epoch": 0.9472734924547586, "flos": 15815853214560.0, "grad_norm": 1.9560527905380118, "language_loss": 0.80839586, "learning_rate": 2.9032071041337426e-08, "loss": 0.83026147, "num_input_tokens_seen": 170263305, "step": 7878, "time_per_iteration": 2.5025250911712646 }, { "auxiliary_loss_clip": 0.01131216, "auxiliary_loss_mlp": 0.01021983, "balance_loss_clip": 1.04376984, "balance_loss_mlp": 1.01536393, "epoch": 0.9473937353453977, "flos": 11181796344960.0, "grad_norm": 2.0800256503600894, "language_loss": 0.73197967, "learning_rate": 2.889997566580704e-08, "loss": 0.75351167, "num_input_tokens_seen": 170281460, "step": 7879, "time_per_iteration": 2.4869320392608643 }, { "auxiliary_loss_clip": 0.01168813, "auxiliary_loss_mlp": 0.01030281, "balance_loss_clip": 1.04774344, "balance_loss_mlp": 1.02311063, "epoch": 0.9475139782360368, "flos": 25775534391360.0, "grad_norm": 2.0143312004425513, "language_loss": 0.70313746, "learning_rate": 2.8768179304583086e-08, "loss": 0.72512841, "num_input_tokens_seen": 170303515, "step": 7880, "time_per_iteration": 2.511976957321167 }, { "auxiliary_loss_clip": 0.01127591, "auxiliary_loss_mlp": 0.01022777, "balance_loss_clip": 1.04637098, "balance_loss_mlp": 1.01627743, "epoch": 0.9476342211266758, "flos": 22820050120320.0, "grad_norm": 1.6706427866941702, "language_loss": 0.73788047, "learning_rate": 2.8636681977659117e-08, "loss": 0.75938404, "num_input_tokens_seen": 170323165, "step": 7881, "time_per_iteration": 3.3768210411071777 }, { "auxiliary_loss_clip": 0.01108065, "auxiliary_loss_mlp": 0.01026369, "balance_loss_clip": 1.04315317, "balance_loss_mlp": 1.01916277, "epoch": 0.947754464017315, "flos": 20193615458880.0, "grad_norm": 2.661189648050548, "language_loss": 0.77761978, "learning_rate": 2.850548370498318e-08, "loss": 0.79896414, "num_input_tokens_seen": 170341005, "step": 7882, "time_per_iteration": 2.5825424194335938 }, { "auxiliary_loss_clip": 0.01149884, "auxiliary_loss_mlp": 0.01019465, "balance_loss_clip": 1.04292083, "balance_loss_mlp": 1.01296484, "epoch": 0.9478747069079541, "flos": 24717929263680.0, "grad_norm": 1.69933290459375, "language_loss": 0.71342587, "learning_rate": 2.8374584506457798e-08, "loss": 0.73511934, "num_input_tokens_seen": 170362280, "step": 7883, "time_per_iteration": 2.5094401836395264 }, { "auxiliary_loss_clip": 0.01137749, "auxiliary_loss_mlp": 0.01024227, "balance_loss_clip": 1.04562712, "balance_loss_mlp": 1.01735497, "epoch": 0.9479949497985931, "flos": 21361358737920.0, "grad_norm": 2.361876620444591, "language_loss": 0.67370903, "learning_rate": 2.824398440193998e-08, "loss": 0.69532883, "num_input_tokens_seen": 170381080, "step": 7884, "time_per_iteration": 2.5238547325134277 }, { "auxiliary_loss_clip": 0.0110586, "auxiliary_loss_mlp": 0.01023658, "balance_loss_clip": 1.04062557, "balance_loss_mlp": 1.01642764, "epoch": 0.9481151926892323, "flos": 18148610083680.0, "grad_norm": 1.938895420116199, "language_loss": 0.71342123, "learning_rate": 2.811368341124232e-08, "loss": 0.73471642, "num_input_tokens_seen": 170400150, "step": 7885, "time_per_iteration": 2.56762433052063 }, { "auxiliary_loss_clip": 0.01138626, "auxiliary_loss_mlp": 0.0102423, "balance_loss_clip": 1.04452705, "balance_loss_mlp": 1.01772141, "epoch": 0.9482354355798713, "flos": 22128015762240.0, "grad_norm": 2.379443745995745, "language_loss": 0.68519104, "learning_rate": 2.7983681554131222e-08, "loss": 0.70681959, "num_input_tokens_seen": 170420410, "step": 7886, "time_per_iteration": 3.298358917236328 }, { "auxiliary_loss_clip": 0.011386, "auxiliary_loss_mlp": 0.01025479, "balance_loss_clip": 1.0440613, "balance_loss_mlp": 1.01799905, "epoch": 0.9483556784705104, "flos": 19063075762080.0, "grad_norm": 2.063556920003456, "language_loss": 0.70287788, "learning_rate": 2.7853978850327365e-08, "loss": 0.72451866, "num_input_tokens_seen": 170439580, "step": 7887, "time_per_iteration": 2.5512235164642334 }, { "auxiliary_loss_clip": 0.01126627, "auxiliary_loss_mlp": 0.01029594, "balance_loss_clip": 1.04829574, "balance_loss_mlp": 1.02323461, "epoch": 0.9484759213611496, "flos": 25777114737600.0, "grad_norm": 1.799312609185661, "language_loss": 0.87162113, "learning_rate": 2.7724575319507225e-08, "loss": 0.89318335, "num_input_tokens_seen": 170459290, "step": 7888, "time_per_iteration": 3.404021978378296 }, { "auxiliary_loss_clip": 0.01151694, "auxiliary_loss_mlp": 0.01021793, "balance_loss_clip": 1.04295588, "balance_loss_mlp": 1.01565707, "epoch": 0.9485961642517886, "flos": 20667744033600.0, "grad_norm": 1.906343395567177, "language_loss": 0.77384412, "learning_rate": 2.759547098130044e-08, "loss": 0.79557908, "num_input_tokens_seen": 170478020, "step": 7889, "time_per_iteration": 2.4980363845825195 }, { "auxiliary_loss_clip": 0.01164249, "auxiliary_loss_mlp": 0.01025751, "balance_loss_clip": 1.04598999, "balance_loss_mlp": 1.01889348, "epoch": 0.9487164071424277, "flos": 22674073231200.0, "grad_norm": 1.6340835393351676, "language_loss": 0.76620603, "learning_rate": 2.746666585529267e-08, "loss": 0.78810596, "num_input_tokens_seen": 170498295, "step": 7890, "time_per_iteration": 2.4676849842071533 }, { "auxiliary_loss_clip": 0.01145525, "auxiliary_loss_mlp": 0.01026854, "balance_loss_clip": 1.04465365, "balance_loss_mlp": 1.01978171, "epoch": 0.9488366500330668, "flos": 38726466743040.0, "grad_norm": 2.2947298229139643, "language_loss": 0.7419343, "learning_rate": 2.73381599610234e-08, "loss": 0.76365811, "num_input_tokens_seen": 170518695, "step": 7891, "time_per_iteration": 2.651179790496826 }, { "auxiliary_loss_clip": 0.01147193, "auxiliary_loss_mlp": 0.01025157, "balance_loss_clip": 1.04279304, "balance_loss_mlp": 1.01816487, "epoch": 0.9489568929237059, "flos": 27890924231520.0, "grad_norm": 1.7104925608795947, "language_loss": 0.71824324, "learning_rate": 2.7209953317987033e-08, "loss": 0.73996675, "num_input_tokens_seen": 170539735, "step": 7892, "time_per_iteration": 2.5134947299957275 }, { "auxiliary_loss_clip": 0.01153753, "auxiliary_loss_mlp": 0.01026943, "balance_loss_clip": 1.04672694, "balance_loss_mlp": 1.02025807, "epoch": 0.9490771358143449, "flos": 33580646712960.0, "grad_norm": 1.934301522110241, "language_loss": 0.77870399, "learning_rate": 2.7082045945631793e-08, "loss": 0.80051094, "num_input_tokens_seen": 170561950, "step": 7893, "time_per_iteration": 2.6060047149658203 }, { "auxiliary_loss_clip": 0.01115063, "auxiliary_loss_mlp": 0.0101975, "balance_loss_clip": 1.04110169, "balance_loss_mlp": 1.01290417, "epoch": 0.9491973787049841, "flos": 14793799488960.0, "grad_norm": 2.1302727212767687, "language_loss": 0.69037104, "learning_rate": 2.6954437863361712e-08, "loss": 0.71171921, "num_input_tokens_seen": 170579865, "step": 7894, "time_per_iteration": 2.5198400020599365 }, { "auxiliary_loss_clip": 0.01099768, "auxiliary_loss_mlp": 0.01019865, "balance_loss_clip": 1.0411284, "balance_loss_mlp": 1.01315653, "epoch": 0.9493176215956232, "flos": 25332539432640.0, "grad_norm": 4.442428652234587, "language_loss": 0.71126139, "learning_rate": 2.6827129090534862e-08, "loss": 0.73245776, "num_input_tokens_seen": 170600165, "step": 7895, "time_per_iteration": 2.6384329795837402 }, { "auxiliary_loss_clip": 0.01139963, "auxiliary_loss_mlp": 0.01024656, "balance_loss_clip": 1.04707134, "balance_loss_mlp": 1.01763749, "epoch": 0.9494378644862622, "flos": 21029974525920.0, "grad_norm": 1.9403146800003572, "language_loss": 0.77693981, "learning_rate": 2.670011964646335e-08, "loss": 0.79858601, "num_input_tokens_seen": 170618845, "step": 7896, "time_per_iteration": 2.50339412689209 }, { "auxiliary_loss_clip": 0.01086897, "auxiliary_loss_mlp": 0.01025828, "balance_loss_clip": 1.03509116, "balance_loss_mlp": 1.01859164, "epoch": 0.9495581073769014, "flos": 15195137162400.0, "grad_norm": 2.0003832258608765, "language_loss": 0.675883, "learning_rate": 2.657340955041487e-08, "loss": 0.69701022, "num_input_tokens_seen": 170637620, "step": 7897, "time_per_iteration": 2.6291069984436035 }, { "auxiliary_loss_clip": 0.01141663, "auxiliary_loss_mlp": 0.01028451, "balance_loss_clip": 1.04902899, "balance_loss_mlp": 1.02110744, "epoch": 0.9496783502675404, "flos": 28616570475840.0, "grad_norm": 2.0245936242346168, "language_loss": 0.71727288, "learning_rate": 2.6446998821611167e-08, "loss": 0.73897398, "num_input_tokens_seen": 170657815, "step": 7898, "time_per_iteration": 3.3490560054779053 }, { "auxiliary_loss_clip": 0.01113603, "auxiliary_loss_mlp": 0.01029003, "balance_loss_clip": 1.04356158, "balance_loss_mlp": 1.02246094, "epoch": 0.9497985931581795, "flos": 14866877725920.0, "grad_norm": 2.483368144127661, "language_loss": 0.7165029, "learning_rate": 2.6320887479228228e-08, "loss": 0.73792899, "num_input_tokens_seen": 170674415, "step": 7899, "time_per_iteration": 2.5296132564544678 }, { "auxiliary_loss_clip": 0.01142182, "auxiliary_loss_mlp": 0.0102617, "balance_loss_clip": 1.0439992, "balance_loss_mlp": 1.01921976, "epoch": 0.9499188360488187, "flos": 27193322744640.0, "grad_norm": 3.101300638933684, "language_loss": 0.72419935, "learning_rate": 2.619507554239786e-08, "loss": 0.74588287, "num_input_tokens_seen": 170692975, "step": 7900, "time_per_iteration": 2.5661776065826416 }, { "auxiliary_loss_clip": 0.01141604, "auxiliary_loss_mlp": 0.01027498, "balance_loss_clip": 1.04652274, "balance_loss_mlp": 1.02050638, "epoch": 0.9500390789394577, "flos": 24316483839360.0, "grad_norm": 1.5129527359552002, "language_loss": 0.69802916, "learning_rate": 2.606956303020502e-08, "loss": 0.71972018, "num_input_tokens_seen": 170713780, "step": 7901, "time_per_iteration": 2.5465214252471924 }, { "auxiliary_loss_clip": 0.01153316, "auxiliary_loss_mlp": 0.01027081, "balance_loss_clip": 1.04681647, "balance_loss_mlp": 1.02009201, "epoch": 0.9501593218300968, "flos": 14354755395840.0, "grad_norm": 1.8155051834500031, "language_loss": 0.84204334, "learning_rate": 2.5944349961690036e-08, "loss": 0.86384726, "num_input_tokens_seen": 170730800, "step": 7902, "time_per_iteration": 2.4797403812408447 }, { "auxiliary_loss_clip": 0.01125577, "auxiliary_loss_mlp": 0.01022898, "balance_loss_clip": 1.04451966, "balance_loss_mlp": 1.01574218, "epoch": 0.9502795647207359, "flos": 38728118923200.0, "grad_norm": 1.6403467998909245, "language_loss": 0.73381442, "learning_rate": 2.581943635584749e-08, "loss": 0.75529909, "num_input_tokens_seen": 170753630, "step": 7903, "time_per_iteration": 2.6859264373779297 }, { "auxiliary_loss_clip": 0.01130385, "auxiliary_loss_mlp": 0.01023982, "balance_loss_clip": 1.04332376, "balance_loss_mlp": 1.01709795, "epoch": 0.950399807611375, "flos": 40808029195200.0, "grad_norm": 1.5140045224102234, "language_loss": 0.65142393, "learning_rate": 2.569482223162689e-08, "loss": 0.67296755, "num_input_tokens_seen": 170777605, "step": 7904, "time_per_iteration": 2.6949336528778076 }, { "auxiliary_loss_clip": 0.01153834, "auxiliary_loss_mlp": 0.01029121, "balance_loss_clip": 1.04561913, "balance_loss_mlp": 1.02231741, "epoch": 0.950520050502014, "flos": 23440407002880.0, "grad_norm": 2.7325106649249578, "language_loss": 0.72173965, "learning_rate": 2.5570507607932e-08, "loss": 0.7435692, "num_input_tokens_seen": 170797520, "step": 7905, "time_per_iteration": 2.49159836769104 }, { "auxiliary_loss_clip": 0.01157611, "auxiliary_loss_mlp": 0.01025373, "balance_loss_clip": 1.04562116, "balance_loss_mlp": 1.01833963, "epoch": 0.9506402933926532, "flos": 17783721736320.0, "grad_norm": 7.1810509843506605, "language_loss": 0.63727701, "learning_rate": 2.54464925036213e-08, "loss": 0.65910679, "num_input_tokens_seen": 170814810, "step": 7906, "time_per_iteration": 2.456681728363037 }, { "auxiliary_loss_clip": 0.01155295, "auxiliary_loss_mlp": 0.0102171, "balance_loss_clip": 1.04847455, "balance_loss_mlp": 1.01518643, "epoch": 0.9507605362832923, "flos": 32561933264640.0, "grad_norm": 1.8481348861592048, "language_loss": 0.60867524, "learning_rate": 2.532277693750773e-08, "loss": 0.63044536, "num_input_tokens_seen": 170835735, "step": 7907, "time_per_iteration": 3.3266797065734863 }, { "auxiliary_loss_clip": 0.01111488, "auxiliary_loss_mlp": 0.01027224, "balance_loss_clip": 1.04542291, "balance_loss_mlp": 1.02006292, "epoch": 0.9508807791739313, "flos": 19602057589920.0, "grad_norm": 1.8673455730236563, "language_loss": 0.76090991, "learning_rate": 2.5199360928358948e-08, "loss": 0.78229702, "num_input_tokens_seen": 170852970, "step": 7908, "time_per_iteration": 2.585221767425537 }, { "auxiliary_loss_clip": 0.01141965, "auxiliary_loss_mlp": 0.0076077, "balance_loss_clip": 1.04261255, "balance_loss_mlp": 1.00053, "epoch": 0.9510010220645704, "flos": 21471856058880.0, "grad_norm": 1.6162697156862984, "language_loss": 0.87224257, "learning_rate": 2.507624449489665e-08, "loss": 0.89126998, "num_input_tokens_seen": 170871600, "step": 7909, "time_per_iteration": 2.4863786697387695 }, { "auxiliary_loss_clip": 0.01139759, "auxiliary_loss_mlp": 0.01025952, "balance_loss_clip": 1.04578829, "balance_loss_mlp": 1.01897252, "epoch": 0.9511212649552095, "flos": 18880003041600.0, "grad_norm": 2.068997987052488, "language_loss": 0.64787173, "learning_rate": 2.495342765579811e-08, "loss": 0.66952878, "num_input_tokens_seen": 170890260, "step": 7910, "time_per_iteration": 2.4989542961120605 }, { "auxiliary_loss_clip": 0.01113023, "auxiliary_loss_mlp": 0.01023978, "balance_loss_clip": 1.04663014, "balance_loss_mlp": 1.01734722, "epoch": 0.9512415078458486, "flos": 20810524313280.0, "grad_norm": 1.8438114050434837, "language_loss": 0.70913577, "learning_rate": 2.4830910429693984e-08, "loss": 0.73050582, "num_input_tokens_seen": 170910220, "step": 7911, "time_per_iteration": 2.5896401405334473 }, { "auxiliary_loss_clip": 0.01167344, "auxiliary_loss_mlp": 0.01027673, "balance_loss_clip": 1.04727352, "balance_loss_mlp": 1.02073789, "epoch": 0.9513617507364877, "flos": 18369568808640.0, "grad_norm": 2.063380871094882, "language_loss": 0.79700673, "learning_rate": 2.470869283517052e-08, "loss": 0.81895697, "num_input_tokens_seen": 170928255, "step": 7912, "time_per_iteration": 3.2001707553863525 }, { "auxiliary_loss_clip": 0.0114505, "auxiliary_loss_mlp": 0.01034979, "balance_loss_clip": 1.04353392, "balance_loss_mlp": 1.02850246, "epoch": 0.9514819936271268, "flos": 25010170377600.0, "grad_norm": 1.5462344862695465, "language_loss": 0.7708801, "learning_rate": 2.458677489076777e-08, "loss": 0.79268044, "num_input_tokens_seen": 170949265, "step": 7913, "time_per_iteration": 2.6076509952545166 }, { "auxiliary_loss_clip": 0.01140989, "auxiliary_loss_mlp": 0.01023916, "balance_loss_clip": 1.04152656, "balance_loss_mlp": 1.01720464, "epoch": 0.9516022365177659, "flos": 18662133175200.0, "grad_norm": 1.6309760342622814, "language_loss": 0.83002788, "learning_rate": 2.446515661498072e-08, "loss": 0.85167694, "num_input_tokens_seen": 170968595, "step": 7914, "time_per_iteration": 3.1818816661834717 }, { "auxiliary_loss_clip": 0.01097808, "auxiliary_loss_mlp": 0.01026361, "balance_loss_clip": 1.04188299, "balance_loss_mlp": 1.0197804, "epoch": 0.9517224794084049, "flos": 25372113534240.0, "grad_norm": 2.7462144713421113, "language_loss": 0.74588919, "learning_rate": 2.434383802625861e-08, "loss": 0.76713085, "num_input_tokens_seen": 170987550, "step": 7915, "time_per_iteration": 2.6657493114471436 }, { "auxiliary_loss_clip": 0.01124769, "auxiliary_loss_mlp": 0.01025946, "balance_loss_clip": 1.0414021, "balance_loss_mlp": 1.0195595, "epoch": 0.9518427222990441, "flos": 21470922217920.0, "grad_norm": 1.9212709399079522, "language_loss": 0.73907423, "learning_rate": 2.4222819143005168e-08, "loss": 0.76058137, "num_input_tokens_seen": 171007145, "step": 7916, "time_per_iteration": 2.5535826683044434 }, { "auxiliary_loss_clip": 0.01166982, "auxiliary_loss_mlp": 0.01029249, "balance_loss_clip": 1.04833555, "balance_loss_mlp": 1.0222218, "epoch": 0.9519629651896832, "flos": 21033638055840.0, "grad_norm": 1.8785043689932106, "language_loss": 0.80846465, "learning_rate": 2.4102099983579706e-08, "loss": 0.83042693, "num_input_tokens_seen": 171026295, "step": 7917, "time_per_iteration": 2.468003988265991 }, { "auxiliary_loss_clip": 0.01152032, "auxiliary_loss_mlp": 0.01022071, "balance_loss_clip": 1.04373789, "balance_loss_mlp": 1.01465058, "epoch": 0.9520832080803222, "flos": 21689223087840.0, "grad_norm": 1.8109685038562944, "language_loss": 0.7726953, "learning_rate": 2.3981680566294236e-08, "loss": 0.7944364, "num_input_tokens_seen": 171045895, "step": 7918, "time_per_iteration": 2.569347381591797 }, { "auxiliary_loss_clip": 0.0116675, "auxiliary_loss_mlp": 0.01026608, "balance_loss_clip": 1.0486083, "balance_loss_mlp": 1.01979232, "epoch": 0.9522034509709614, "flos": 23145292532160.0, "grad_norm": 1.790628818791085, "language_loss": 0.73475039, "learning_rate": 2.3861560909416822e-08, "loss": 0.75668395, "num_input_tokens_seen": 171065445, "step": 7919, "time_per_iteration": 2.5108680725097656 }, { "auxiliary_loss_clip": 0.01108088, "auxiliary_loss_mlp": 0.01028553, "balance_loss_clip": 1.04355311, "balance_loss_mlp": 1.02163291, "epoch": 0.9523236938616004, "flos": 24679432670880.0, "grad_norm": 1.7234356425716884, "language_loss": 0.8246969, "learning_rate": 2.3741741031169325e-08, "loss": 0.84606332, "num_input_tokens_seen": 171085015, "step": 7920, "time_per_iteration": 2.607175827026367 }, { "auxiliary_loss_clip": 0.0110518, "auxiliary_loss_mlp": 0.01024196, "balance_loss_clip": 1.04058099, "balance_loss_mlp": 1.01730561, "epoch": 0.9524439367522395, "flos": 22672313300160.0, "grad_norm": 1.8624859673423553, "language_loss": 0.71712524, "learning_rate": 2.3622220949728544e-08, "loss": 0.738419, "num_input_tokens_seen": 171103900, "step": 7921, "time_per_iteration": 2.5815160274505615 }, { "auxiliary_loss_clip": 0.01144878, "auxiliary_loss_mlp": 0.01025198, "balance_loss_clip": 1.04320312, "balance_loss_mlp": 1.01794434, "epoch": 0.9525641796428787, "flos": 34055529543840.0, "grad_norm": 4.291480652842178, "language_loss": 0.61464906, "learning_rate": 2.3503000683225526e-08, "loss": 0.63634986, "num_input_tokens_seen": 171121615, "step": 7922, "time_per_iteration": 2.5664026737213135 }, { "auxiliary_loss_clip": 0.01166697, "auxiliary_loss_mlp": 0.01022671, "balance_loss_clip": 1.04564476, "balance_loss_mlp": 1.01564646, "epoch": 0.9526844225335177, "flos": 16727086366560.0, "grad_norm": 1.9217190426100226, "language_loss": 0.8432709, "learning_rate": 2.3384080249745585e-08, "loss": 0.86516464, "num_input_tokens_seen": 171139505, "step": 7923, "time_per_iteration": 2.426018476486206 }, { "auxiliary_loss_clip": 0.01115027, "auxiliary_loss_mlp": 0.01027127, "balance_loss_clip": 1.04102492, "balance_loss_mlp": 1.02082348, "epoch": 0.9528046654241568, "flos": 36939372256320.0, "grad_norm": 2.1703942609083673, "language_loss": 0.82869637, "learning_rate": 2.3265459667329178e-08, "loss": 0.85011792, "num_input_tokens_seen": 171158995, "step": 7924, "time_per_iteration": 3.422579288482666 }, { "auxiliary_loss_clip": 0.01140989, "auxiliary_loss_mlp": 0.01022677, "balance_loss_clip": 1.0460161, "balance_loss_mlp": 1.01611137, "epoch": 0.9529249083147959, "flos": 18255012871200.0, "grad_norm": 2.1191929750176235, "language_loss": 0.86416632, "learning_rate": 2.31471389539708e-08, "loss": 0.88580298, "num_input_tokens_seen": 171176120, "step": 7925, "time_per_iteration": 2.5004897117614746 }, { "auxiliary_loss_clip": 0.01156716, "auxiliary_loss_mlp": 0.00761224, "balance_loss_clip": 1.04789007, "balance_loss_mlp": 1.00044739, "epoch": 0.953045151205435, "flos": 28658443262880.0, "grad_norm": 2.0557998134706907, "language_loss": 0.73007548, "learning_rate": 2.3029118127619872e-08, "loss": 0.74925494, "num_input_tokens_seen": 171195835, "step": 7926, "time_per_iteration": 2.545212745666504 }, { "auxiliary_loss_clip": 0.01132354, "auxiliary_loss_mlp": 0.01026545, "balance_loss_clip": 1.04398131, "balance_loss_mlp": 1.01929164, "epoch": 0.953165394096074, "flos": 21835235893920.0, "grad_norm": 2.0712152418955876, "language_loss": 0.86473298, "learning_rate": 2.2911397206179628e-08, "loss": 0.88632202, "num_input_tokens_seen": 171212585, "step": 7927, "time_per_iteration": 2.520781993865967 }, { "auxiliary_loss_clip": 0.01168391, "auxiliary_loss_mlp": 0.01023815, "balance_loss_clip": 1.04954624, "balance_loss_mlp": 1.01729393, "epoch": 0.9532856369867132, "flos": 19975063170240.0, "grad_norm": 1.915088859465088, "language_loss": 0.62777305, "learning_rate": 2.279397620750845e-08, "loss": 0.6496951, "num_input_tokens_seen": 171231630, "step": 7928, "time_per_iteration": 2.4713752269744873 }, { "auxiliary_loss_clip": 0.0113497, "auxiliary_loss_mlp": 0.0102979, "balance_loss_clip": 1.04343343, "balance_loss_mlp": 1.02344513, "epoch": 0.9534058798773523, "flos": 15049591276800.0, "grad_norm": 1.9247522527833518, "language_loss": 0.78927273, "learning_rate": 2.2676855149419195e-08, "loss": 0.8109203, "num_input_tokens_seen": 171248800, "step": 7929, "time_per_iteration": 2.5098612308502197 }, { "auxiliary_loss_clip": 0.01141083, "auxiliary_loss_mlp": 0.010233, "balance_loss_clip": 1.05127072, "balance_loss_mlp": 1.01677036, "epoch": 0.9535261227679913, "flos": 17602804033440.0, "grad_norm": 3.3429547542679376, "language_loss": 0.75075245, "learning_rate": 2.2560034049678988e-08, "loss": 0.77239627, "num_input_tokens_seen": 171263150, "step": 7930, "time_per_iteration": 2.484261989593506 }, { "auxiliary_loss_clip": 0.01170645, "auxiliary_loss_mlp": 0.01025994, "balance_loss_clip": 1.04855227, "balance_loss_mlp": 1.01881194, "epoch": 0.9536463656586305, "flos": 23142957929760.0, "grad_norm": 3.2282985153248442, "language_loss": 0.75112933, "learning_rate": 2.2443512926008988e-08, "loss": 0.77309573, "num_input_tokens_seen": 171282480, "step": 7931, "time_per_iteration": 2.4862637519836426 }, { "auxiliary_loss_clip": 0.01127584, "auxiliary_loss_mlp": 0.01026354, "balance_loss_clip": 1.04268527, "balance_loss_mlp": 1.01981282, "epoch": 0.9537666085492695, "flos": 18625037343840.0, "grad_norm": 2.7394187502935052, "language_loss": 0.69863266, "learning_rate": 2.2327291796085946e-08, "loss": 0.72017205, "num_input_tokens_seen": 171300840, "step": 7932, "time_per_iteration": 2.518266201019287 }, { "auxiliary_loss_clip": 0.01166066, "auxiliary_loss_mlp": 0.01022801, "balance_loss_clip": 1.04530978, "balance_loss_mlp": 1.01525497, "epoch": 0.9538868514399086, "flos": 18989351019840.0, "grad_norm": 5.475887587380631, "language_loss": 0.77286267, "learning_rate": 2.2211370677540197e-08, "loss": 0.79475141, "num_input_tokens_seen": 171317365, "step": 7933, "time_per_iteration": 3.179276704788208 }, { "auxiliary_loss_clip": 0.01169909, "auxiliary_loss_mlp": 0.01027472, "balance_loss_clip": 1.04884028, "balance_loss_mlp": 1.02082586, "epoch": 0.9540070943305478, "flos": 16800559690080.0, "grad_norm": 3.6145608227968142, "language_loss": 0.78050566, "learning_rate": 2.2095749587957012e-08, "loss": 0.80247945, "num_input_tokens_seen": 171335270, "step": 7934, "time_per_iteration": 2.424013614654541 }, { "auxiliary_loss_clip": 0.01133871, "auxiliary_loss_mlp": 0.01021388, "balance_loss_clip": 1.04033756, "balance_loss_mlp": 1.01417303, "epoch": 0.9541273372211868, "flos": 20156914714080.0, "grad_norm": 1.8653119096120836, "language_loss": 0.69479179, "learning_rate": 2.1980428544876138e-08, "loss": 0.71634442, "num_input_tokens_seen": 171353910, "step": 7935, "time_per_iteration": 2.494166851043701 }, { "auxiliary_loss_clip": 0.01102954, "auxiliary_loss_mlp": 0.01023726, "balance_loss_clip": 1.03560996, "balance_loss_mlp": 1.01691043, "epoch": 0.9542475801118259, "flos": 26725515554880.0, "grad_norm": 1.6245533937297914, "language_loss": 0.74418318, "learning_rate": 2.1865407565791584e-08, "loss": 0.76544994, "num_input_tokens_seen": 171375480, "step": 7936, "time_per_iteration": 2.6310431957244873 }, { "auxiliary_loss_clip": 0.01139743, "auxiliary_loss_mlp": 0.01025887, "balance_loss_clip": 1.04352319, "balance_loss_mlp": 1.01856434, "epoch": 0.954367823002465, "flos": 23330915356800.0, "grad_norm": 1.9644958922210842, "language_loss": 0.77253115, "learning_rate": 2.175068666815183e-08, "loss": 0.79418743, "num_input_tokens_seen": 171396320, "step": 7937, "time_per_iteration": 2.518202066421509 }, { "auxiliary_loss_clip": 0.01126143, "auxiliary_loss_mlp": 0.01029362, "balance_loss_clip": 1.04270363, "balance_loss_mlp": 1.02231336, "epoch": 0.9544880658931041, "flos": 14902716463680.0, "grad_norm": 2.5581197350194196, "language_loss": 0.79245853, "learning_rate": 2.163626586935985e-08, "loss": 0.8140136, "num_input_tokens_seen": 171412860, "step": 7938, "time_per_iteration": 3.2772810459136963 }, { "auxiliary_loss_clip": 0.0114953, "auxiliary_loss_mlp": 0.01025728, "balance_loss_clip": 1.0443579, "balance_loss_mlp": 1.01844406, "epoch": 0.9546083087837431, "flos": 29095907009760.0, "grad_norm": 2.1663700166780875, "language_loss": 0.62929195, "learning_rate": 2.1522145186773755e-08, "loss": 0.65104449, "num_input_tokens_seen": 171431780, "step": 7939, "time_per_iteration": 3.2446224689483643 }, { "auxiliary_loss_clip": 0.0113966, "auxiliary_loss_mlp": 0.0102761, "balance_loss_clip": 1.0464251, "balance_loss_mlp": 1.02054119, "epoch": 0.9547285516743822, "flos": 21142339528800.0, "grad_norm": 1.6775295832665194, "language_loss": 0.85275078, "learning_rate": 2.140832463770481e-08, "loss": 0.8744235, "num_input_tokens_seen": 171450975, "step": 7940, "time_per_iteration": 2.5361201763153076 }, { "auxiliary_loss_clip": 0.01143498, "auxiliary_loss_mlp": 0.01022835, "balance_loss_clip": 1.04398537, "balance_loss_mlp": 1.01599836, "epoch": 0.9548487945650214, "flos": 27490161229440.0, "grad_norm": 2.456350176764915, "language_loss": 0.75678355, "learning_rate": 2.129480423941987e-08, "loss": 0.77844691, "num_input_tokens_seen": 171467645, "step": 7941, "time_per_iteration": 2.5556042194366455 }, { "auxiliary_loss_clip": 0.01140275, "auxiliary_loss_mlp": 0.01028837, "balance_loss_clip": 1.04592788, "balance_loss_mlp": 1.02240562, "epoch": 0.9549690374556604, "flos": 22273202478240.0, "grad_norm": 1.609453234669024, "language_loss": 0.80285156, "learning_rate": 2.1181584009140052e-08, "loss": 0.82454264, "num_input_tokens_seen": 171487185, "step": 7942, "time_per_iteration": 2.5199599266052246 }, { "auxiliary_loss_clip": 0.01132445, "auxiliary_loss_mlp": 0.01020933, "balance_loss_clip": 1.04225302, "balance_loss_mlp": 1.01500559, "epoch": 0.9550892803462995, "flos": 17595297388800.0, "grad_norm": 2.0352789154442954, "language_loss": 0.83869612, "learning_rate": 2.10686639640405e-08, "loss": 0.86022997, "num_input_tokens_seen": 171501275, "step": 7943, "time_per_iteration": 2.4948630332946777 }, { "auxiliary_loss_clip": 0.0115374, "auxiliary_loss_mlp": 0.01023356, "balance_loss_clip": 1.04624033, "balance_loss_mlp": 1.01682019, "epoch": 0.9552095232369386, "flos": 24353148667200.0, "grad_norm": 1.7077741962383255, "language_loss": 0.8122654, "learning_rate": 2.0956044121251294e-08, "loss": 0.83403635, "num_input_tokens_seen": 171520060, "step": 7944, "time_per_iteration": 2.5026357173919678 }, { "auxiliary_loss_clip": 0.01127051, "auxiliary_loss_mlp": 0.0102503, "balance_loss_clip": 1.04679275, "balance_loss_mlp": 1.01799369, "epoch": 0.9553297661275777, "flos": 22746864132480.0, "grad_norm": 1.7926727193957543, "language_loss": 0.80999076, "learning_rate": 2.084372449785654e-08, "loss": 0.83151162, "num_input_tokens_seen": 171539895, "step": 7945, "time_per_iteration": 2.5447816848754883 }, { "auxiliary_loss_clip": 0.01133748, "auxiliary_loss_mlp": 0.01023793, "balance_loss_clip": 1.04111028, "balance_loss_mlp": 1.01690006, "epoch": 0.9554500090182168, "flos": 15413869035840.0, "grad_norm": 1.668660062161609, "language_loss": 0.68640846, "learning_rate": 2.0731705110895282e-08, "loss": 0.70798385, "num_input_tokens_seen": 171557385, "step": 7946, "time_per_iteration": 2.5237019062042236 }, { "auxiliary_loss_clip": 0.01158042, "auxiliary_loss_mlp": 0.01025972, "balance_loss_clip": 1.04944313, "balance_loss_mlp": 1.01817608, "epoch": 0.9555702519088559, "flos": 23513521156800.0, "grad_norm": 1.69906843751868, "language_loss": 0.86378622, "learning_rate": 2.0619985977360587e-08, "loss": 0.88562644, "num_input_tokens_seen": 171575705, "step": 7947, "time_per_iteration": 2.497403383255005 }, { "auxiliary_loss_clip": 0.0111254, "auxiliary_loss_mlp": 0.01021797, "balance_loss_clip": 1.03935099, "balance_loss_mlp": 1.01510644, "epoch": 0.955690494799495, "flos": 22962076143840.0, "grad_norm": 2.1875606674210943, "language_loss": 0.76934826, "learning_rate": 2.0508567114200237e-08, "loss": 0.79069161, "num_input_tokens_seen": 171595620, "step": 7948, "time_per_iteration": 2.6037991046905518 }, { "auxiliary_loss_clip": 0.01140482, "auxiliary_loss_mlp": 0.01027215, "balance_loss_clip": 1.04423892, "balance_loss_mlp": 1.02030683, "epoch": 0.955810737690134, "flos": 26031254345280.0, "grad_norm": 2.1546660219101827, "language_loss": 0.78673214, "learning_rate": 2.0397448538316485e-08, "loss": 0.80840915, "num_input_tokens_seen": 171616660, "step": 7949, "time_per_iteration": 2.5908970832824707 }, { "auxiliary_loss_clip": 0.0111971, "auxiliary_loss_mlp": 0.0102603, "balance_loss_clip": 1.04182267, "balance_loss_mlp": 1.01958919, "epoch": 0.9559309805807732, "flos": 20849954747040.0, "grad_norm": 2.0734624337204077, "language_loss": 0.66594428, "learning_rate": 2.028663026656563e-08, "loss": 0.68740171, "num_input_tokens_seen": 171635515, "step": 7950, "time_per_iteration": 3.4925546646118164 }, { "auxiliary_loss_clip": 0.0116616, "auxiliary_loss_mlp": 0.00761612, "balance_loss_clip": 1.04826677, "balance_loss_mlp": 1.00048101, "epoch": 0.9560512234714122, "flos": 21578222929440.0, "grad_norm": 2.6654619543394524, "language_loss": 0.72103953, "learning_rate": 2.0176112315758885e-08, "loss": 0.74031734, "num_input_tokens_seen": 171653305, "step": 7951, "time_per_iteration": 2.4699175357818604 }, { "auxiliary_loss_clip": 0.01117251, "auxiliary_loss_mlp": 0.01023159, "balance_loss_clip": 1.04204535, "balance_loss_mlp": 1.01674247, "epoch": 0.9561714663620513, "flos": 17450146589760.0, "grad_norm": 41.69438110934833, "language_loss": 0.69234288, "learning_rate": 2.0065894702661957e-08, "loss": 0.71374702, "num_input_tokens_seen": 171669980, "step": 7952, "time_per_iteration": 2.5535755157470703 }, { "auxiliary_loss_clip": 0.01115727, "auxiliary_loss_mlp": 0.00761828, "balance_loss_clip": 1.03851533, "balance_loss_mlp": 1.00049579, "epoch": 0.9562917092526905, "flos": 26098514034720.0, "grad_norm": 2.04066486671106, "language_loss": 0.77881354, "learning_rate": 1.9955977443994577e-08, "loss": 0.79758906, "num_input_tokens_seen": 171689970, "step": 7953, "time_per_iteration": 2.615462064743042 }, { "auxiliary_loss_clip": 0.01138567, "auxiliary_loss_mlp": 0.0102591, "balance_loss_clip": 1.04458046, "balance_loss_mlp": 1.01788139, "epoch": 0.9564119521433295, "flos": 24096746291040.0, "grad_norm": 2.4188185023789646, "language_loss": 0.62132561, "learning_rate": 1.9846360556430965e-08, "loss": 0.64297038, "num_input_tokens_seen": 171708270, "step": 7954, "time_per_iteration": 2.5343711376190186 }, { "auxiliary_loss_clip": 0.01163736, "auxiliary_loss_mlp": 0.01026898, "balance_loss_clip": 1.04530311, "balance_loss_mlp": 1.0200789, "epoch": 0.9565321950339686, "flos": 32008907905440.0, "grad_norm": 2.759242369194684, "language_loss": 0.61496443, "learning_rate": 1.973704405660004e-08, "loss": 0.63687074, "num_input_tokens_seen": 171729385, "step": 7955, "time_per_iteration": 2.558000326156616 }, { "auxiliary_loss_clip": 0.01097529, "auxiliary_loss_mlp": 0.01020903, "balance_loss_clip": 1.04273438, "balance_loss_mlp": 1.01474881, "epoch": 0.9566524379246077, "flos": 23588646660480.0, "grad_norm": 1.6448344249303033, "language_loss": 0.77995741, "learning_rate": 1.9628027961085203e-08, "loss": 0.80114174, "num_input_tokens_seen": 171752615, "step": 7956, "time_per_iteration": 2.6403543949127197 }, { "auxiliary_loss_clip": 0.01111819, "auxiliary_loss_mlp": 0.01027487, "balance_loss_clip": 1.03916216, "balance_loss_mlp": 1.02074599, "epoch": 0.9567726808152468, "flos": 38067721018560.0, "grad_norm": 2.1913822178625093, "language_loss": 0.84157032, "learning_rate": 1.9519312286423894e-08, "loss": 0.86296344, "num_input_tokens_seen": 171775810, "step": 7957, "time_per_iteration": 2.7126011848449707 }, { "auxiliary_loss_clip": 0.01152373, "auxiliary_loss_mlp": 0.01030913, "balance_loss_clip": 1.04713535, "balance_loss_mlp": 1.02450824, "epoch": 0.9568929237058859, "flos": 22744062609600.0, "grad_norm": 1.6860417659861207, "language_loss": 0.77522564, "learning_rate": 1.9410897049108255e-08, "loss": 0.79705846, "num_input_tokens_seen": 171795090, "step": 7958, "time_per_iteration": 2.4892325401306152 }, { "auxiliary_loss_clip": 0.01173835, "auxiliary_loss_mlp": 0.01030061, "balance_loss_clip": 1.05099607, "balance_loss_mlp": 1.02303672, "epoch": 0.957013166596525, "flos": 23841636925440.0, "grad_norm": 1.7379192367301193, "language_loss": 0.90979588, "learning_rate": 1.9302782265584905e-08, "loss": 0.93183482, "num_input_tokens_seen": 171815755, "step": 7959, "time_per_iteration": 3.240030288696289 }, { "auxiliary_loss_clip": 0.01103472, "auxiliary_loss_mlp": 0.01026243, "balance_loss_clip": 1.04375398, "balance_loss_mlp": 1.01953149, "epoch": 0.9571334094871641, "flos": 17639289276480.0, "grad_norm": 2.0160661756429428, "language_loss": 0.8702473, "learning_rate": 1.9194967952254282e-08, "loss": 0.89154446, "num_input_tokens_seen": 171834330, "step": 7960, "time_per_iteration": 2.5437138080596924 }, { "auxiliary_loss_clip": 0.0115423, "auxiliary_loss_mlp": 0.01024701, "balance_loss_clip": 1.04831946, "balance_loss_mlp": 1.01783776, "epoch": 0.9572536523778031, "flos": 15369625729440.0, "grad_norm": 2.720536745291313, "language_loss": 0.80748737, "learning_rate": 1.9087454125472635e-08, "loss": 0.82927668, "num_input_tokens_seen": 171848805, "step": 7961, "time_per_iteration": 2.4634900093078613 }, { "auxiliary_loss_clip": 0.01168622, "auxiliary_loss_mlp": 0.01029557, "balance_loss_clip": 1.04845035, "balance_loss_mlp": 1.02295256, "epoch": 0.9573738952684423, "flos": 24969842019840.0, "grad_norm": 1.7616559263619014, "language_loss": 0.78390598, "learning_rate": 1.8980240801548696e-08, "loss": 0.8058877, "num_input_tokens_seen": 171867995, "step": 7962, "time_per_iteration": 2.4693500995635986 }, { "auxiliary_loss_clip": 0.01136299, "auxiliary_loss_mlp": 0.01029212, "balance_loss_clip": 1.04621077, "balance_loss_mlp": 1.02230978, "epoch": 0.9574941381590814, "flos": 25769464425120.0, "grad_norm": 2.768281786722228, "language_loss": 0.74219877, "learning_rate": 1.8873327996747458e-08, "loss": 0.76385385, "num_input_tokens_seen": 171886495, "step": 7963, "time_per_iteration": 2.539682149887085 }, { "auxiliary_loss_clip": 0.01155968, "auxiliary_loss_mlp": 0.01026754, "balance_loss_clip": 1.04504299, "balance_loss_mlp": 1.01966071, "epoch": 0.9576143810497204, "flos": 32307183068640.0, "grad_norm": 2.806794879358432, "language_loss": 0.65697771, "learning_rate": 1.8766715727287053e-08, "loss": 0.67880493, "num_input_tokens_seen": 171908200, "step": 7964, "time_per_iteration": 3.3441407680511475 }, { "auxiliary_loss_clip": 0.01157585, "auxiliary_loss_mlp": 0.00762268, "balance_loss_clip": 1.04578829, "balance_loss_mlp": 1.00051332, "epoch": 0.9577346239403596, "flos": 27745737515520.0, "grad_norm": 2.208200304224164, "language_loss": 0.79635376, "learning_rate": 1.8660404009340546e-08, "loss": 0.81555223, "num_input_tokens_seen": 171928650, "step": 7965, "time_per_iteration": 3.3079981803894043 }, { "auxiliary_loss_clip": 0.01051132, "auxiliary_loss_mlp": 0.0100091, "balance_loss_clip": 1.00540757, "balance_loss_mlp": 0.99959898, "epoch": 0.9578548668309986, "flos": 57468322792800.0, "grad_norm": 0.899980001782397, "language_loss": 0.59550762, "learning_rate": 1.8554392859035485e-08, "loss": 0.61602807, "num_input_tokens_seen": 171986400, "step": 7966, "time_per_iteration": 3.0985302925109863 }, { "auxiliary_loss_clip": 0.01089603, "auxiliary_loss_mlp": 0.01025059, "balance_loss_clip": 1.04149747, "balance_loss_mlp": 1.01781094, "epoch": 0.9579751097216377, "flos": 19756043961120.0, "grad_norm": 2.0281006047233854, "language_loss": 0.78780729, "learning_rate": 1.8448682292453444e-08, "loss": 0.80895388, "num_input_tokens_seen": 172005475, "step": 7967, "time_per_iteration": 2.6712164878845215 }, { "auxiliary_loss_clip": 0.01168052, "auxiliary_loss_mlp": 0.01026292, "balance_loss_clip": 1.04812574, "balance_loss_mlp": 1.01966977, "epoch": 0.9580953526122769, "flos": 18041273455200.0, "grad_norm": 2.085506839919104, "language_loss": 0.65962082, "learning_rate": 1.8343272325631154e-08, "loss": 0.68156427, "num_input_tokens_seen": 172024420, "step": 7968, "time_per_iteration": 2.676762104034424 }, { "auxiliary_loss_clip": 0.01088886, "auxiliary_loss_mlp": 0.00762415, "balance_loss_clip": 1.04065633, "balance_loss_mlp": 1.00058389, "epoch": 0.9582155955029159, "flos": 24270157349280.0, "grad_norm": 2.275073486678153, "language_loss": 0.78203773, "learning_rate": 1.8238162974558492e-08, "loss": 0.80055076, "num_input_tokens_seen": 172038350, "step": 7969, "time_per_iteration": 2.6146514415740967 }, { "auxiliary_loss_clip": 0.01139689, "auxiliary_loss_mlp": 0.01024554, "balance_loss_clip": 1.04752886, "balance_loss_mlp": 1.01782465, "epoch": 0.958335838393555, "flos": 22783313458560.0, "grad_norm": 1.9352654143333734, "language_loss": 0.74509174, "learning_rate": 1.8133354255181144e-08, "loss": 0.76673412, "num_input_tokens_seen": 172058665, "step": 7970, "time_per_iteration": 2.5248143672943115 }, { "auxiliary_loss_clip": 0.01147653, "auxiliary_loss_mlp": 0.01022916, "balance_loss_clip": 1.04483175, "balance_loss_mlp": 1.01645458, "epoch": 0.958456081284194, "flos": 16911488014560.0, "grad_norm": 5.366717123989161, "language_loss": 0.74492025, "learning_rate": 1.802884618339795e-08, "loss": 0.76662588, "num_input_tokens_seen": 172077470, "step": 7971, "time_per_iteration": 2.4682159423828125 }, { "auxiliary_loss_clip": 0.01157046, "auxiliary_loss_mlp": 0.01028173, "balance_loss_clip": 1.04877198, "balance_loss_mlp": 1.02141714, "epoch": 0.9585763241748332, "flos": 19974955419360.0, "grad_norm": 2.097864014897335, "language_loss": 0.81112731, "learning_rate": 1.7924638775062894e-08, "loss": 0.83297956, "num_input_tokens_seen": 172096590, "step": 7972, "time_per_iteration": 2.4833083152770996 }, { "auxiliary_loss_clip": 0.01120541, "auxiliary_loss_mlp": 0.01029534, "balance_loss_clip": 1.04418468, "balance_loss_mlp": 1.02263141, "epoch": 0.9586965670654722, "flos": 21395653046400.0, "grad_norm": 2.148696888404443, "language_loss": 0.8154766, "learning_rate": 1.7820732045984444e-08, "loss": 0.83697736, "num_input_tokens_seen": 172116735, "step": 7973, "time_per_iteration": 2.5436089038848877 }, { "auxiliary_loss_clip": 0.01150189, "auxiliary_loss_mlp": 0.010248, "balance_loss_clip": 1.04530239, "balance_loss_mlp": 1.01764488, "epoch": 0.9588168099561113, "flos": 21435119397120.0, "grad_norm": 2.654636887392148, "language_loss": 0.73874927, "learning_rate": 1.7717126011924655e-08, "loss": 0.76049918, "num_input_tokens_seen": 172138320, "step": 7974, "time_per_iteration": 2.541977882385254 }, { "auxiliary_loss_clip": 0.0110375, "auxiliary_loss_mlp": 0.01025157, "balance_loss_clip": 1.03749704, "balance_loss_mlp": 1.01821637, "epoch": 0.9589370528467505, "flos": 11763764385600.0, "grad_norm": 2.300552942535351, "language_loss": 0.76525372, "learning_rate": 1.7613820688600957e-08, "loss": 0.78654277, "num_input_tokens_seen": 172154225, "step": 7975, "time_per_iteration": 3.2285687923431396 }, { "auxiliary_loss_clip": 0.01139395, "auxiliary_loss_mlp": 0.01023751, "balance_loss_clip": 1.04323292, "balance_loss_mlp": 1.01678061, "epoch": 0.9590572957373895, "flos": 23441520428640.0, "grad_norm": 1.8028675422003064, "language_loss": 0.78492594, "learning_rate": 1.7510816091684588e-08, "loss": 0.80655742, "num_input_tokens_seen": 172174150, "step": 7976, "time_per_iteration": 2.5228700637817383 }, { "auxiliary_loss_clip": 0.01142635, "auxiliary_loss_mlp": 0.01029307, "balance_loss_clip": 1.04726279, "balance_loss_mlp": 1.02129626, "epoch": 0.9591775386280286, "flos": 22528275926880.0, "grad_norm": 2.33387743476561, "language_loss": 0.78789163, "learning_rate": 1.740811223680083e-08, "loss": 0.80961108, "num_input_tokens_seen": 172191005, "step": 7977, "time_per_iteration": 2.5255422592163086 }, { "auxiliary_loss_clip": 0.01168315, "auxiliary_loss_mlp": 0.01025103, "balance_loss_clip": 1.04829693, "balance_loss_mlp": 1.01843953, "epoch": 0.9592977815186677, "flos": 18186963008640.0, "grad_norm": 2.1937613584485365, "language_loss": 0.74090958, "learning_rate": 1.7305709139530334e-08, "loss": 0.76284379, "num_input_tokens_seen": 172209785, "step": 7978, "time_per_iteration": 2.425623655319214 }, { "auxiliary_loss_clip": 0.01147531, "auxiliary_loss_mlp": 0.01024766, "balance_loss_clip": 1.04386067, "balance_loss_mlp": 1.01786053, "epoch": 0.9594180244093068, "flos": 16537800012000.0, "grad_norm": 2.430675620233765, "language_loss": 0.74657792, "learning_rate": 1.7203606815407334e-08, "loss": 0.76830089, "num_input_tokens_seen": 172224380, "step": 7979, "time_per_iteration": 2.460272789001465 }, { "auxiliary_loss_clip": 0.01144113, "auxiliary_loss_mlp": 0.01027859, "balance_loss_clip": 1.04583979, "balance_loss_mlp": 1.02060235, "epoch": 0.9595382672999458, "flos": 20554337438880.0, "grad_norm": 1.6305144317181048, "language_loss": 0.7913745, "learning_rate": 1.7101805279920557e-08, "loss": 0.81309426, "num_input_tokens_seen": 172242540, "step": 7980, "time_per_iteration": 2.509115695953369 }, { "auxiliary_loss_clip": 0.01171484, "auxiliary_loss_mlp": 0.01026715, "balance_loss_clip": 1.05029857, "balance_loss_mlp": 1.01902628, "epoch": 0.959658510190585, "flos": 22638270410400.0, "grad_norm": 1.9667920032948307, "language_loss": 0.81101024, "learning_rate": 1.7000304548513643e-08, "loss": 0.83299226, "num_input_tokens_seen": 172262645, "step": 7981, "time_per_iteration": 2.4701132774353027 }, { "auxiliary_loss_clip": 0.01122644, "auxiliary_loss_mlp": 0.0102692, "balance_loss_clip": 1.04150784, "balance_loss_mlp": 1.01985121, "epoch": 0.9597787530812241, "flos": 19135256075040.0, "grad_norm": 1.934496532392516, "language_loss": 0.8307817, "learning_rate": 1.6899104636583394e-08, "loss": 0.8522774, "num_input_tokens_seen": 172280695, "step": 7982, "time_per_iteration": 2.530574083328247 }, { "auxiliary_loss_clip": 0.01051025, "auxiliary_loss_mlp": 0.01000639, "balance_loss_clip": 1.00508332, "balance_loss_mlp": 0.99930984, "epoch": 0.9598989959718631, "flos": 60098133648480.0, "grad_norm": 0.7250664247126525, "language_loss": 0.61935401, "learning_rate": 1.6798205559482638e-08, "loss": 0.63987064, "num_input_tokens_seen": 172343075, "step": 7983, "time_per_iteration": 3.2376770973205566 }, { "auxiliary_loss_clip": 0.01130631, "auxiliary_loss_mlp": 0.01028538, "balance_loss_clip": 1.04655027, "balance_loss_mlp": 1.02126074, "epoch": 0.9600192388625023, "flos": 20886799159680.0, "grad_norm": 4.9831397214324635, "language_loss": 0.76631773, "learning_rate": 1.669760733251713e-08, "loss": 0.78790951, "num_input_tokens_seen": 172361950, "step": 7984, "time_per_iteration": 2.563833475112915 }, { "auxiliary_loss_clip": 0.01098873, "auxiliary_loss_mlp": 0.01019142, "balance_loss_clip": 1.03985524, "balance_loss_mlp": 1.01243377, "epoch": 0.9601394817531413, "flos": 20445743716800.0, "grad_norm": 1.6114905765451333, "language_loss": 0.82561648, "learning_rate": 1.659730997094755e-08, "loss": 0.84679663, "num_input_tokens_seen": 172380440, "step": 7985, "time_per_iteration": 3.357604503631592 }, { "auxiliary_loss_clip": 0.01145379, "auxiliary_loss_mlp": 0.01023333, "balance_loss_clip": 1.04426348, "balance_loss_mlp": 1.01640701, "epoch": 0.9602597246437804, "flos": 21507156042240.0, "grad_norm": 1.7336070795318366, "language_loss": 0.62503684, "learning_rate": 1.6497313489989283e-08, "loss": 0.64672399, "num_input_tokens_seen": 172400265, "step": 7986, "time_per_iteration": 2.499124050140381 }, { "auxiliary_loss_clip": 0.01108731, "auxiliary_loss_mlp": 0.01022583, "balance_loss_clip": 1.03536415, "balance_loss_mlp": 1.01585388, "epoch": 0.9603799675344196, "flos": 29935103516640.0, "grad_norm": 2.163959965889606, "language_loss": 0.69831854, "learning_rate": 1.639761790481131e-08, "loss": 0.71963167, "num_input_tokens_seen": 172421145, "step": 7987, "time_per_iteration": 2.6480114459991455 }, { "auxiliary_loss_clip": 0.01152016, "auxiliary_loss_mlp": 0.01030789, "balance_loss_clip": 1.0474118, "balance_loss_mlp": 1.02415848, "epoch": 0.9605002104250586, "flos": 28001529303360.0, "grad_norm": 2.976390987901465, "language_loss": 0.79520142, "learning_rate": 1.6298223230537754e-08, "loss": 0.81702954, "num_input_tokens_seen": 172438945, "step": 7988, "time_per_iteration": 2.52268385887146 }, { "auxiliary_loss_clip": 0.011387, "auxiliary_loss_mlp": 0.00762313, "balance_loss_clip": 1.04585683, "balance_loss_mlp": 1.00053406, "epoch": 0.9606204533156977, "flos": 35590495772640.0, "grad_norm": 2.0438681195077875, "language_loss": 0.69671482, "learning_rate": 1.619912948224611e-08, "loss": 0.71572495, "num_input_tokens_seen": 172460150, "step": 7989, "time_per_iteration": 2.636651039123535 }, { "auxiliary_loss_clip": 0.0112066, "auxiliary_loss_mlp": 0.0102705, "balance_loss_clip": 1.04275799, "balance_loss_mlp": 1.02008212, "epoch": 0.9607406962063368, "flos": 26574618042240.0, "grad_norm": 3.370979335885444, "language_loss": 0.61399525, "learning_rate": 1.6100336674969682e-08, "loss": 0.6354723, "num_input_tokens_seen": 172478990, "step": 7990, "time_per_iteration": 2.5876107215881348 }, { "auxiliary_loss_clip": 0.01110177, "auxiliary_loss_mlp": 0.01028361, "balance_loss_clip": 1.03927624, "balance_loss_mlp": 1.02178681, "epoch": 0.9608609390969759, "flos": 25331785176480.0, "grad_norm": 1.6685889650573942, "language_loss": 0.76607811, "learning_rate": 1.600184482369449e-08, "loss": 0.78746355, "num_input_tokens_seen": 172498905, "step": 7991, "time_per_iteration": 3.3618032932281494 }, { "auxiliary_loss_clip": 0.01129149, "auxiliary_loss_mlp": 0.01024541, "balance_loss_clip": 1.04332519, "balance_loss_mlp": 1.0176568, "epoch": 0.960981181987615, "flos": 21069117624000.0, "grad_norm": 2.374984971928248, "language_loss": 0.89157802, "learning_rate": 1.5903653943362126e-08, "loss": 0.91311491, "num_input_tokens_seen": 172517900, "step": 7992, "time_per_iteration": 2.563352108001709 }, { "auxiliary_loss_clip": 0.01140901, "auxiliary_loss_mlp": 0.01022909, "balance_loss_clip": 1.04571128, "balance_loss_mlp": 1.01625085, "epoch": 0.9611014248782541, "flos": 17823259920960.0, "grad_norm": 2.5415973578087896, "language_loss": 0.76607895, "learning_rate": 1.580576404886802e-08, "loss": 0.78771704, "num_input_tokens_seen": 172536430, "step": 7993, "time_per_iteration": 2.4987823963165283 }, { "auxiliary_loss_clip": 0.01153333, "auxiliary_loss_mlp": 0.01020364, "balance_loss_clip": 1.04579306, "balance_loss_mlp": 1.01438856, "epoch": 0.9612216677688932, "flos": 19354634453760.0, "grad_norm": 2.039385726108984, "language_loss": 0.79794443, "learning_rate": 1.570817515506162e-08, "loss": 0.81968141, "num_input_tokens_seen": 172555120, "step": 7994, "time_per_iteration": 2.4761037826538086 }, { "auxiliary_loss_clip": 0.0116554, "auxiliary_loss_mlp": 0.01023173, "balance_loss_clip": 1.04821098, "balance_loss_mlp": 1.01692629, "epoch": 0.9613419106595322, "flos": 15808741656480.0, "grad_norm": 1.988041905032371, "language_loss": 0.81540585, "learning_rate": 1.561088727674753e-08, "loss": 0.83729297, "num_input_tokens_seen": 172569330, "step": 7995, "time_per_iteration": 2.404966115951538 }, { "auxiliary_loss_clip": 0.01126357, "auxiliary_loss_mlp": 0.01025283, "balance_loss_clip": 1.04483676, "balance_loss_mlp": 1.01766562, "epoch": 0.9614621535501714, "flos": 25702492071360.0, "grad_norm": 2.061523842998044, "language_loss": 0.71322083, "learning_rate": 1.551390042868417e-08, "loss": 0.73473722, "num_input_tokens_seen": 172591100, "step": 7996, "time_per_iteration": 2.662372350692749 }, { "auxiliary_loss_clip": 0.01155548, "auxiliary_loss_mlp": 0.01027434, "balance_loss_clip": 1.04813361, "balance_loss_mlp": 1.02028775, "epoch": 0.9615823964408104, "flos": 17819057636640.0, "grad_norm": 2.6139765782036704, "language_loss": 0.71080315, "learning_rate": 1.5417214625584207e-08, "loss": 0.73263294, "num_input_tokens_seen": 172608755, "step": 7997, "time_per_iteration": 2.470884084701538 }, { "auxiliary_loss_clip": 0.01146691, "auxiliary_loss_mlp": 0.01028263, "balance_loss_clip": 1.04299402, "balance_loss_mlp": 1.02134562, "epoch": 0.9617026393314495, "flos": 20190023762880.0, "grad_norm": 1.7332662183633465, "language_loss": 0.85109508, "learning_rate": 1.5320829882114806e-08, "loss": 0.87284458, "num_input_tokens_seen": 172626830, "step": 7998, "time_per_iteration": 2.4952199459075928 }, { "auxiliary_loss_clip": 0.01164873, "auxiliary_loss_mlp": 0.01021623, "balance_loss_clip": 1.0449115, "balance_loss_mlp": 1.01510215, "epoch": 0.9618228822220887, "flos": 20267016948480.0, "grad_norm": 1.9954001873815936, "language_loss": 0.78833604, "learning_rate": 1.5224746212897378e-08, "loss": 0.81020105, "num_input_tokens_seen": 172646125, "step": 7999, "time_per_iteration": 2.4437382221221924 }, { "auxiliary_loss_clip": 0.01162656, "auxiliary_loss_mlp": 0.01024586, "balance_loss_clip": 1.04556274, "balance_loss_mlp": 1.01774931, "epoch": 0.9619431251127277, "flos": 21031303453440.0, "grad_norm": 1.794360052920439, "language_loss": 0.77202284, "learning_rate": 1.512896363250804e-08, "loss": 0.7938953, "num_input_tokens_seen": 172666235, "step": 8000, "time_per_iteration": 2.4795830249786377 }, { "auxiliary_loss_clip": 0.0115546, "auxiliary_loss_mlp": 0.0102753, "balance_loss_clip": 1.04528093, "balance_loss_mlp": 1.02076745, "epoch": 0.9620633680033668, "flos": 22382658207360.0, "grad_norm": 1.8504940801574252, "language_loss": 0.75150013, "learning_rate": 1.503348215547673e-08, "loss": 0.77332997, "num_input_tokens_seen": 172687325, "step": 8001, "time_per_iteration": 2.518918514251709 }, { "auxiliary_loss_clip": 0.01137244, "auxiliary_loss_mlp": 0.01023918, "balance_loss_clip": 1.04524493, "balance_loss_mlp": 1.0168159, "epoch": 0.962183610894006, "flos": 18471733394880.0, "grad_norm": 1.9146883625519184, "language_loss": 0.80867374, "learning_rate": 1.4938301796288078e-08, "loss": 0.83028531, "num_input_tokens_seen": 172703895, "step": 8002, "time_per_iteration": 3.2464351654052734 }, { "auxiliary_loss_clip": 0.01168773, "auxiliary_loss_mlp": 0.01025716, "balance_loss_clip": 1.0476048, "balance_loss_mlp": 1.01837587, "epoch": 0.962303853784645, "flos": 18435248151840.0, "grad_norm": 2.583173936434123, "language_loss": 0.82463777, "learning_rate": 1.4843422569380537e-08, "loss": 0.84658271, "num_input_tokens_seen": 172720650, "step": 8003, "time_per_iteration": 2.423888683319092 }, { "auxiliary_loss_clip": 0.01106667, "auxiliary_loss_mlp": 0.01023327, "balance_loss_clip": 1.04127622, "balance_loss_mlp": 1.0162158, "epoch": 0.9624240966752841, "flos": 26391078401280.0, "grad_norm": 1.7603751076788856, "language_loss": 0.82573932, "learning_rate": 1.4748844489147483e-08, "loss": 0.84703922, "num_input_tokens_seen": 172737640, "step": 8004, "time_per_iteration": 2.6214139461517334 }, { "auxiliary_loss_clip": 0.01135452, "auxiliary_loss_mlp": 0.01024247, "balance_loss_clip": 1.0406512, "balance_loss_mlp": 1.01779485, "epoch": 0.9625443395659231, "flos": 14647678932000.0, "grad_norm": 1.9663210422440853, "language_loss": 0.70844674, "learning_rate": 1.4654567569936326e-08, "loss": 0.73004377, "num_input_tokens_seen": 172755215, "step": 8005, "time_per_iteration": 2.4942715167999268 }, { "auxiliary_loss_clip": 0.01105563, "auxiliary_loss_mlp": 0.01025933, "balance_loss_clip": 1.04033685, "balance_loss_mlp": 1.01900363, "epoch": 0.9626645824565623, "flos": 18367629292800.0, "grad_norm": 2.005786813713208, "language_loss": 0.83005154, "learning_rate": 1.456059182604874e-08, "loss": 0.85136652, "num_input_tokens_seen": 172774020, "step": 8006, "time_per_iteration": 2.5655386447906494 }, { "auxiliary_loss_clip": 0.01169364, "auxiliary_loss_mlp": 0.01025218, "balance_loss_clip": 1.04966784, "balance_loss_mlp": 1.01776481, "epoch": 0.9627848253472013, "flos": 16580427055200.0, "grad_norm": 2.7156827289656946, "language_loss": 0.76250058, "learning_rate": 1.4466917271740653e-08, "loss": 0.78444636, "num_input_tokens_seen": 172792220, "step": 8007, "time_per_iteration": 2.4226784706115723 }, { "auxiliary_loss_clip": 0.011351, "auxiliary_loss_mlp": 0.01026759, "balance_loss_clip": 1.0435127, "balance_loss_mlp": 1.01923728, "epoch": 0.9629050682378404, "flos": 20886870993600.0, "grad_norm": 2.1176527734806294, "language_loss": 0.67705178, "learning_rate": 1.4373543921222697e-08, "loss": 0.69867039, "num_input_tokens_seen": 172811805, "step": 8008, "time_per_iteration": 2.5186069011688232 }, { "auxiliary_loss_clip": 0.01139318, "auxiliary_loss_mlp": 0.0102951, "balance_loss_clip": 1.0476737, "balance_loss_mlp": 1.02234507, "epoch": 0.9630253111284796, "flos": 17019255646560.0, "grad_norm": 1.8282603904070533, "language_loss": 0.78040528, "learning_rate": 1.428047178865932e-08, "loss": 0.80209351, "num_input_tokens_seen": 172828595, "step": 8009, "time_per_iteration": 2.495936632156372 }, { "auxiliary_loss_clip": 0.01136345, "auxiliary_loss_mlp": 0.01023702, "balance_loss_clip": 1.04166067, "balance_loss_mlp": 1.01696968, "epoch": 0.9631455540191186, "flos": 20338945842720.0, "grad_norm": 5.4094816790651, "language_loss": 0.74255228, "learning_rate": 1.4187700888169451e-08, "loss": 0.76415271, "num_input_tokens_seen": 172847770, "step": 8010, "time_per_iteration": 2.52931547164917 }, { "auxiliary_loss_clip": 0.01048892, "auxiliary_loss_mlp": 0.01002494, "balance_loss_clip": 1.00618434, "balance_loss_mlp": 1.00125372, "epoch": 0.9632657969097577, "flos": 65956712122560.0, "grad_norm": 0.7506536358665313, "language_loss": 0.56999993, "learning_rate": 1.40952312338265e-08, "loss": 0.59051383, "num_input_tokens_seen": 172912415, "step": 8011, "time_per_iteration": 3.873385429382324 }, { "auxiliary_loss_clip": 0.0112648, "auxiliary_loss_mlp": 0.01026732, "balance_loss_clip": 1.04195547, "balance_loss_mlp": 1.02028561, "epoch": 0.9633860398003968, "flos": 44419529501760.0, "grad_norm": 1.7756386293204223, "language_loss": 0.68669242, "learning_rate": 1.4003062839657909e-08, "loss": 0.70822453, "num_input_tokens_seen": 172934895, "step": 8012, "time_per_iteration": 2.727745532989502 }, { "auxiliary_loss_clip": 0.01122008, "auxiliary_loss_mlp": 0.01024466, "balance_loss_clip": 1.04206514, "balance_loss_mlp": 1.01843691, "epoch": 0.9635062826910359, "flos": 24827708245440.0, "grad_norm": 1.7043567366692356, "language_loss": 0.79820013, "learning_rate": 1.391119571964583e-08, "loss": 0.81966484, "num_input_tokens_seen": 172955835, "step": 8013, "time_per_iteration": 2.5729763507843018 }, { "auxiliary_loss_clip": 0.01151756, "auxiliary_loss_mlp": 0.01021987, "balance_loss_clip": 1.04642987, "balance_loss_mlp": 1.01539516, "epoch": 0.9636265255816749, "flos": 15961363183200.0, "grad_norm": 2.004712569656372, "language_loss": 0.73120546, "learning_rate": 1.3819629887726225e-08, "loss": 0.75294292, "num_input_tokens_seen": 172973925, "step": 8014, "time_per_iteration": 2.4493796825408936 }, { "auxiliary_loss_clip": 0.01143937, "auxiliary_loss_mlp": 0.01022611, "balance_loss_clip": 1.04632807, "balance_loss_mlp": 1.01602125, "epoch": 0.9637467684723141, "flos": 22601785167360.0, "grad_norm": 2.2622104913825716, "language_loss": 0.76272839, "learning_rate": 1.3728365357789317e-08, "loss": 0.78439391, "num_input_tokens_seen": 172993290, "step": 8015, "time_per_iteration": 2.533036470413208 }, { "auxiliary_loss_clip": 0.01082541, "auxiliary_loss_mlp": 0.01024386, "balance_loss_clip": 1.03863788, "balance_loss_mlp": 1.01719809, "epoch": 0.9638670113629532, "flos": 17565815952960.0, "grad_norm": 3.1595475285698997, "language_loss": 0.76234508, "learning_rate": 1.3637402143680254e-08, "loss": 0.78341436, "num_input_tokens_seen": 173008190, "step": 8016, "time_per_iteration": 2.560457468032837 }, { "auxiliary_loss_clip": 0.01025346, "auxiliary_loss_mlp": 0.00999405, "balance_loss_clip": 1.01056266, "balance_loss_mlp": 0.99828482, "epoch": 0.9639872542535922, "flos": 55072149348960.0, "grad_norm": 0.7251204426813496, "language_loss": 0.5508855, "learning_rate": 1.3546740259197998e-08, "loss": 0.57113302, "num_input_tokens_seen": 173061000, "step": 8017, "time_per_iteration": 4.557470083236694 }, { "auxiliary_loss_clip": 0.01140685, "auxiliary_loss_mlp": 0.01024986, "balance_loss_clip": 1.04549146, "balance_loss_mlp": 1.01780665, "epoch": 0.9641074971442314, "flos": 24134488627680.0, "grad_norm": 2.2156400507817424, "language_loss": 0.70212835, "learning_rate": 1.3456379718095989e-08, "loss": 0.72378504, "num_input_tokens_seen": 173081415, "step": 8018, "time_per_iteration": 2.5523715019226074 }, { "auxiliary_loss_clip": 0.01036039, "auxiliary_loss_mlp": 0.01000538, "balance_loss_clip": 1.00473523, "balance_loss_mlp": 0.99928015, "epoch": 0.9642277400348704, "flos": 66747427121760.0, "grad_norm": 0.8463543819291002, "language_loss": 0.62004137, "learning_rate": 1.3366320534081487e-08, "loss": 0.64040715, "num_input_tokens_seen": 173144095, "step": 8019, "time_per_iteration": 3.1440205574035645 }, { "auxiliary_loss_clip": 0.01153539, "auxiliary_loss_mlp": 0.01028453, "balance_loss_clip": 1.04668117, "balance_loss_mlp": 1.02170885, "epoch": 0.9643479829255095, "flos": 30920276912640.0, "grad_norm": 2.4697423322230043, "language_loss": 0.76125115, "learning_rate": 1.3276562720816675e-08, "loss": 0.7830711, "num_input_tokens_seen": 173165605, "step": 8020, "time_per_iteration": 2.5872080326080322 }, { "auxiliary_loss_clip": 0.01167249, "auxiliary_loss_mlp": 0.01026042, "balance_loss_clip": 1.04674482, "balance_loss_mlp": 1.0193541, "epoch": 0.9644682258161487, "flos": 20048249158080.0, "grad_norm": 2.1789829994573533, "language_loss": 0.82402813, "learning_rate": 1.3187106291917549e-08, "loss": 0.84596103, "num_input_tokens_seen": 173182595, "step": 8021, "time_per_iteration": 2.441877603530884 }, { "auxiliary_loss_clip": 0.01147749, "auxiliary_loss_mlp": 0.01021842, "balance_loss_clip": 1.04391646, "balance_loss_mlp": 1.01540184, "epoch": 0.9645884687067877, "flos": 21178717020960.0, "grad_norm": 1.8015290029446054, "language_loss": 0.70441699, "learning_rate": 1.309795126095503e-08, "loss": 0.7261129, "num_input_tokens_seen": 173200895, "step": 8022, "time_per_iteration": 2.491445302963257 }, { "auxiliary_loss_clip": 0.01079823, "auxiliary_loss_mlp": 0.01026756, "balance_loss_clip": 1.03778887, "balance_loss_mlp": 1.01973462, "epoch": 0.9647087115974268, "flos": 18945969720480.0, "grad_norm": 2.8301388498797633, "language_loss": 0.80481482, "learning_rate": 1.3009097641453192e-08, "loss": 0.82588053, "num_input_tokens_seen": 173218745, "step": 8023, "time_per_iteration": 2.6639280319213867 }, { "auxiliary_loss_clip": 0.01139113, "auxiliary_loss_mlp": 0.01026955, "balance_loss_clip": 1.0457629, "balance_loss_mlp": 1.02012682, "epoch": 0.9648289544880659, "flos": 16545091154880.0, "grad_norm": 1.6586093396454331, "language_loss": 0.75783205, "learning_rate": 1.2920545446891474e-08, "loss": 0.77949274, "num_input_tokens_seen": 173235465, "step": 8024, "time_per_iteration": 2.511528253555298 }, { "auxiliary_loss_clip": 0.0114381, "auxiliary_loss_mlp": 0.01027919, "balance_loss_clip": 1.04907465, "balance_loss_mlp": 1.02085268, "epoch": 0.964949197378705, "flos": 24057531359040.0, "grad_norm": 2.374579769705091, "language_loss": 0.70859373, "learning_rate": 1.2832294690703127e-08, "loss": 0.73031098, "num_input_tokens_seen": 173254440, "step": 8025, "time_per_iteration": 2.5302207469940186 }, { "auxiliary_loss_clip": 0.01153248, "auxiliary_loss_mlp": 0.0102058, "balance_loss_clip": 1.04613864, "balance_loss_mlp": 1.01414871, "epoch": 0.965069440269344, "flos": 23365568834880.0, "grad_norm": 2.1634243925829306, "language_loss": 0.77538824, "learning_rate": 1.2744345386275668e-08, "loss": 0.79712653, "num_input_tokens_seen": 173273980, "step": 8026, "time_per_iteration": 2.5050156116485596 }, { "auxiliary_loss_clip": 0.01144566, "auxiliary_loss_mlp": 0.01026737, "balance_loss_clip": 1.04868412, "balance_loss_mlp": 1.01974237, "epoch": 0.9651896831599832, "flos": 25374879140160.0, "grad_norm": 1.6537433305370635, "language_loss": 0.78842276, "learning_rate": 1.265669754695109e-08, "loss": 0.81013578, "num_input_tokens_seen": 173293550, "step": 8027, "time_per_iteration": 2.536374807357788 }, { "auxiliary_loss_clip": 0.01100539, "auxiliary_loss_mlp": 0.01027355, "balance_loss_clip": 1.03949869, "balance_loss_mlp": 1.02019668, "epoch": 0.9653099260506223, "flos": 22272879225600.0, "grad_norm": 1.8554328011956724, "language_loss": 0.81796575, "learning_rate": 1.2569351186025201e-08, "loss": 0.83924472, "num_input_tokens_seen": 173312005, "step": 8028, "time_per_iteration": 3.3708295822143555 }, { "auxiliary_loss_clip": 0.01113847, "auxiliary_loss_mlp": 0.0102106, "balance_loss_clip": 1.04080868, "balance_loss_mlp": 1.0145514, "epoch": 0.9654301689412613, "flos": 26760851455200.0, "grad_norm": 1.5043835274367254, "language_loss": 0.75298584, "learning_rate": 1.2482306316748737e-08, "loss": 0.77433491, "num_input_tokens_seen": 173332450, "step": 8029, "time_per_iteration": 2.583169937133789 }, { "auxiliary_loss_clip": 0.01158768, "auxiliary_loss_mlp": 0.01022403, "balance_loss_clip": 1.04567778, "balance_loss_mlp": 1.01574183, "epoch": 0.9655504118319005, "flos": 17412691588800.0, "grad_norm": 2.9527590418277483, "language_loss": 0.7872923, "learning_rate": 1.2395562952326021e-08, "loss": 0.80910397, "num_input_tokens_seen": 173349610, "step": 8030, "time_per_iteration": 2.466265916824341 }, { "auxiliary_loss_clip": 0.01150416, "auxiliary_loss_mlp": 0.0102719, "balance_loss_clip": 1.04671001, "balance_loss_mlp": 1.02003169, "epoch": 0.9656706547225395, "flos": 22126974170400.0, "grad_norm": 2.6884179829319224, "language_loss": 0.81150091, "learning_rate": 1.2309121105916309e-08, "loss": 0.83327699, "num_input_tokens_seen": 173367900, "step": 8031, "time_per_iteration": 2.5266284942626953 }, { "auxiliary_loss_clip": 0.0115709, "auxiliary_loss_mlp": 0.01021848, "balance_loss_clip": 1.0471096, "balance_loss_mlp": 1.01447487, "epoch": 0.9657908976131786, "flos": 37049294905920.0, "grad_norm": 3.5504388118845935, "language_loss": 0.69109738, "learning_rate": 1.222298079063222e-08, "loss": 0.71288675, "num_input_tokens_seen": 173389040, "step": 8032, "time_per_iteration": 2.605119228363037 }, { "auxiliary_loss_clip": 0.01152334, "auxiliary_loss_mlp": 0.01022841, "balance_loss_clip": 1.04623163, "balance_loss_mlp": 1.01629972, "epoch": 0.9659111405038178, "flos": 24389813495040.0, "grad_norm": 1.8861030338569051, "language_loss": 0.72279286, "learning_rate": 1.2137142019541524e-08, "loss": 0.74454463, "num_input_tokens_seen": 173407595, "step": 8033, "time_per_iteration": 2.4993574619293213 }, { "auxiliary_loss_clip": 0.01142503, "auxiliary_loss_mlp": 0.01027704, "balance_loss_clip": 1.04448509, "balance_loss_mlp": 1.0205754, "epoch": 0.9660313833944568, "flos": 25009416121440.0, "grad_norm": 2.354413470244886, "language_loss": 0.73486352, "learning_rate": 1.2051604805666027e-08, "loss": 0.75656557, "num_input_tokens_seen": 173424720, "step": 8034, "time_per_iteration": 2.537566661834717 }, { "auxiliary_loss_clip": 0.01166653, "auxiliary_loss_mlp": 0.00761678, "balance_loss_clip": 1.04754436, "balance_loss_mlp": 1.00051749, "epoch": 0.9661516262850959, "flos": 11801578556160.0, "grad_norm": 3.020581074537436, "language_loss": 0.78182983, "learning_rate": 1.196636916198135e-08, "loss": 0.80111313, "num_input_tokens_seen": 173442260, "step": 8035, "time_per_iteration": 2.4392828941345215 }, { "auxiliary_loss_clip": 0.0116868, "auxiliary_loss_mlp": 0.01022477, "balance_loss_clip": 1.04760838, "balance_loss_mlp": 1.01558721, "epoch": 0.9662718691757349, "flos": 20047782237600.0, "grad_norm": 1.944248099110461, "language_loss": 0.76728547, "learning_rate": 1.1881435101418036e-08, "loss": 0.78919703, "num_input_tokens_seen": 173461675, "step": 8036, "time_per_iteration": 2.448986053466797 }, { "auxiliary_loss_clip": 0.01038966, "auxiliary_loss_mlp": 0.01002722, "balance_loss_clip": 1.0058639, "balance_loss_mlp": 1.00156009, "epoch": 0.9663921120663741, "flos": 68027714988480.0, "grad_norm": 0.7228776711398983, "language_loss": 0.65568912, "learning_rate": 1.1796802636860003e-08, "loss": 0.67610604, "num_input_tokens_seen": 173530205, "step": 8037, "time_per_iteration": 3.9235892295837402 }, { "auxiliary_loss_clip": 0.01169653, "auxiliary_loss_mlp": 0.01026037, "balance_loss_clip": 1.04839659, "balance_loss_mlp": 1.01903009, "epoch": 0.9665123549570132, "flos": 26322920787840.0, "grad_norm": 1.9966806138752256, "language_loss": 0.73850691, "learning_rate": 1.1712471781146316e-08, "loss": 0.76046383, "num_input_tokens_seen": 173549540, "step": 8038, "time_per_iteration": 2.4760639667510986 }, { "auxiliary_loss_clip": 0.01163539, "auxiliary_loss_mlp": 0.01026132, "balance_loss_clip": 1.04438055, "balance_loss_mlp": 1.01915574, "epoch": 0.9666325978476522, "flos": 43941126808800.0, "grad_norm": 1.8064310082266573, "language_loss": 0.66924512, "learning_rate": 1.1628442547069628e-08, "loss": 0.69114184, "num_input_tokens_seen": 173571740, "step": 8039, "time_per_iteration": 2.654331922531128 }, { "auxiliary_loss_clip": 0.0115669, "auxiliary_loss_mlp": 0.0076188, "balance_loss_clip": 1.04617095, "balance_loss_mlp": 1.00048554, "epoch": 0.9667528407382914, "flos": 21543425783520.0, "grad_norm": 1.9032412453881213, "language_loss": 0.77212644, "learning_rate": 1.1544714947377521e-08, "loss": 0.79131216, "num_input_tokens_seen": 173589425, "step": 8040, "time_per_iteration": 2.4857523441314697 }, { "auxiliary_loss_clip": 0.01170974, "auxiliary_loss_mlp": 0.01028164, "balance_loss_clip": 1.04975486, "balance_loss_mlp": 1.02092206, "epoch": 0.9668730836289304, "flos": 23878589088960.0, "grad_norm": 1.776438258227877, "language_loss": 0.70230085, "learning_rate": 1.1461288994770945e-08, "loss": 0.72429228, "num_input_tokens_seen": 173608500, "step": 8041, "time_per_iteration": 2.475555419921875 }, { "auxiliary_loss_clip": 0.01172018, "auxiliary_loss_mlp": 0.0102052, "balance_loss_clip": 1.04859865, "balance_loss_mlp": 1.01302218, "epoch": 0.9669933265195695, "flos": 28293016161120.0, "grad_norm": 1.7428854603888855, "language_loss": 0.77026367, "learning_rate": 1.1378164701906002e-08, "loss": 0.792189, "num_input_tokens_seen": 173630265, "step": 8042, "time_per_iteration": 2.491874933242798 }, { "auxiliary_loss_clip": 0.01169338, "auxiliary_loss_mlp": 0.01026004, "balance_loss_clip": 1.04706836, "balance_loss_mlp": 1.01898575, "epoch": 0.9671135694102087, "flos": 22454766686400.0, "grad_norm": 1.741148037824072, "language_loss": 0.66838694, "learning_rate": 1.1295342081392156e-08, "loss": 0.6903404, "num_input_tokens_seen": 173649625, "step": 8043, "time_per_iteration": 3.249126672744751 }, { "auxiliary_loss_clip": 0.01136532, "auxiliary_loss_mlp": 0.01025448, "balance_loss_clip": 1.04411161, "balance_loss_mlp": 1.0184921, "epoch": 0.9672338123008477, "flos": 20155944956160.0, "grad_norm": 1.763959216068757, "language_loss": 0.69482505, "learning_rate": 1.1212821145793804e-08, "loss": 0.71644485, "num_input_tokens_seen": 173669240, "step": 8044, "time_per_iteration": 3.274325132369995 }, { "auxiliary_loss_clip": 0.01138912, "auxiliary_loss_mlp": 0.01029743, "balance_loss_clip": 1.04286861, "balance_loss_mlp": 1.02252519, "epoch": 0.9673540551914868, "flos": 16977490610400.0, "grad_norm": 2.0225218579491617, "language_loss": 0.78546011, "learning_rate": 1.1130601907629156e-08, "loss": 0.80714667, "num_input_tokens_seen": 173686970, "step": 8045, "time_per_iteration": 2.5103981494903564 }, { "auxiliary_loss_clip": 0.01050345, "auxiliary_loss_mlp": 0.01000801, "balance_loss_clip": 1.0050844, "balance_loss_mlp": 0.99956077, "epoch": 0.9674742980821259, "flos": 61892914364640.0, "grad_norm": 0.8187355699479757, "language_loss": 0.64857268, "learning_rate": 1.1048684379370899e-08, "loss": 0.66908413, "num_input_tokens_seen": 173747655, "step": 8046, "time_per_iteration": 3.061122417449951 }, { "auxiliary_loss_clip": 0.01127857, "auxiliary_loss_mlp": 0.01022738, "balance_loss_clip": 1.04246688, "balance_loss_mlp": 1.01607704, "epoch": 0.967594540972765, "flos": 18697828245120.0, "grad_norm": 1.8150790362236788, "language_loss": 0.74795848, "learning_rate": 1.0967068573445759e-08, "loss": 0.76946443, "num_input_tokens_seen": 173765140, "step": 8047, "time_per_iteration": 2.4802725315093994 }, { "auxiliary_loss_clip": 0.0113617, "auxiliary_loss_mlp": 0.01023616, "balance_loss_clip": 1.04307318, "balance_loss_mlp": 1.01660335, "epoch": 0.967714783863404, "flos": 20777415264480.0, "grad_norm": 4.718352427721217, "language_loss": 0.65200895, "learning_rate": 1.0885754502234945e-08, "loss": 0.67360681, "num_input_tokens_seen": 173784800, "step": 8048, "time_per_iteration": 2.528869390487671 }, { "auxiliary_loss_clip": 0.01122078, "auxiliary_loss_mlp": 0.01025932, "balance_loss_clip": 1.04538476, "balance_loss_mlp": 1.0188446, "epoch": 0.9678350267540432, "flos": 23185477222080.0, "grad_norm": 1.8180333378864648, "language_loss": 0.78061116, "learning_rate": 1.08047421780737e-08, "loss": 0.80209124, "num_input_tokens_seen": 173803990, "step": 8049, "time_per_iteration": 2.5253348350524902 }, { "auxiliary_loss_clip": 0.01146488, "auxiliary_loss_mlp": 0.00761404, "balance_loss_clip": 1.04470885, "balance_loss_mlp": 1.00053442, "epoch": 0.9679552696446823, "flos": 21726067500480.0, "grad_norm": 2.8358129319207754, "language_loss": 0.74028516, "learning_rate": 1.0724031613251305e-08, "loss": 0.75936407, "num_input_tokens_seen": 173821890, "step": 8050, "time_per_iteration": 2.539203643798828 }, { "auxiliary_loss_clip": 0.01159153, "auxiliary_loss_mlp": 0.01028266, "balance_loss_clip": 1.04643023, "balance_loss_mlp": 1.02113104, "epoch": 0.9680755125353213, "flos": 26869050090720.0, "grad_norm": 1.9095974646163825, "language_loss": 0.66646826, "learning_rate": 1.0643622820011744e-08, "loss": 0.68834245, "num_input_tokens_seen": 173842945, "step": 8051, "time_per_iteration": 2.516176223754883 }, { "auxiliary_loss_clip": 0.01172494, "auxiliary_loss_mlp": 0.01028543, "balance_loss_clip": 1.049335, "balance_loss_mlp": 1.02096128, "epoch": 0.9681957554259605, "flos": 28325011784160.0, "grad_norm": 2.346322200312972, "language_loss": 0.67987871, "learning_rate": 1.0563515810552814e-08, "loss": 0.70188904, "num_input_tokens_seen": 173859915, "step": 8052, "time_per_iteration": 2.4974822998046875 }, { "auxiliary_loss_clip": 0.01167084, "auxiliary_loss_mlp": 0.01030188, "balance_loss_clip": 1.047979, "balance_loss_mlp": 1.02305937, "epoch": 0.9683159983165995, "flos": 20557677716160.0, "grad_norm": 1.4902389742767008, "language_loss": 0.73058873, "learning_rate": 1.0483710597026795e-08, "loss": 0.75256151, "num_input_tokens_seen": 173879775, "step": 8053, "time_per_iteration": 2.4506402015686035 }, { "auxiliary_loss_clip": 0.01126748, "auxiliary_loss_mlp": 0.01027124, "balance_loss_clip": 1.04363692, "balance_loss_mlp": 1.02034426, "epoch": 0.9684362412072386, "flos": 24207961951200.0, "grad_norm": 2.023686682327001, "language_loss": 0.73972416, "learning_rate": 1.0404207191540227e-08, "loss": 0.76126295, "num_input_tokens_seen": 173900230, "step": 8054, "time_per_iteration": 3.3607773780822754 }, { "auxiliary_loss_clip": 0.0116422, "auxiliary_loss_mlp": 0.01028276, "balance_loss_clip": 1.04509044, "balance_loss_mlp": 1.02160954, "epoch": 0.9685564840978778, "flos": 22346244798240.0, "grad_norm": 2.026319818590963, "language_loss": 0.74847847, "learning_rate": 1.0325005606153236e-08, "loss": 0.7704035, "num_input_tokens_seen": 173919690, "step": 8055, "time_per_iteration": 2.4590988159179688 }, { "auxiliary_loss_clip": 0.01111539, "auxiliary_loss_mlp": 0.010287, "balance_loss_clip": 1.04186463, "balance_loss_mlp": 1.02153277, "epoch": 0.9686767269885168, "flos": 14386391849280.0, "grad_norm": 2.9129556312813074, "language_loss": 0.79342526, "learning_rate": 1.0246105852881104e-08, "loss": 0.81482768, "num_input_tokens_seen": 173934790, "step": 8056, "time_per_iteration": 2.5521063804626465 }, { "auxiliary_loss_clip": 0.01170173, "auxiliary_loss_mlp": 0.01025513, "balance_loss_clip": 1.04804087, "balance_loss_mlp": 1.01873326, "epoch": 0.9687969698791559, "flos": 21287633995680.0, "grad_norm": 2.0807218356506136, "language_loss": 0.78956491, "learning_rate": 1.0167507943692476e-08, "loss": 0.81152177, "num_input_tokens_seen": 173953875, "step": 8057, "time_per_iteration": 2.437347412109375 }, { "auxiliary_loss_clip": 0.01152662, "auxiliary_loss_mlp": 0.01022737, "balance_loss_clip": 1.04859376, "balance_loss_mlp": 1.01495314, "epoch": 0.968917212769795, "flos": 19828332024960.0, "grad_norm": 2.4403470043959845, "language_loss": 0.71888971, "learning_rate": 1.008921189051093e-08, "loss": 0.74064374, "num_input_tokens_seen": 173971220, "step": 8058, "time_per_iteration": 2.4605627059936523 }, { "auxiliary_loss_clip": 0.01171139, "auxiliary_loss_mlp": 0.01032674, "balance_loss_clip": 1.05010891, "balance_loss_mlp": 1.0257895, "epoch": 0.9690374556604341, "flos": 21681752360160.0, "grad_norm": 2.4052835423705963, "language_loss": 0.77177775, "learning_rate": 1.0011217705213848e-08, "loss": 0.79381585, "num_input_tokens_seen": 173989095, "step": 8059, "time_per_iteration": 2.4493398666381836 }, { "auxiliary_loss_clip": 0.01148592, "auxiliary_loss_mlp": 0.01025809, "balance_loss_clip": 1.04501653, "balance_loss_mlp": 1.01912427, "epoch": 0.9691576985510731, "flos": 32635442505120.0, "grad_norm": 1.7827761145305305, "language_loss": 0.74761093, "learning_rate": 9.933525399632658e-09, "loss": 0.76935494, "num_input_tokens_seen": 174007330, "step": 8060, "time_per_iteration": 2.554089307785034 }, { "auxiliary_loss_clip": 0.01140367, "auxiliary_loss_mlp": 0.01028907, "balance_loss_clip": 1.04703355, "balance_loss_mlp": 1.02054453, "epoch": 0.9692779414417123, "flos": 35663178923040.0, "grad_norm": 1.9952548487526292, "language_loss": 0.64625496, "learning_rate": 9.856134985553488e-09, "loss": 0.66794771, "num_input_tokens_seen": 174027055, "step": 8061, "time_per_iteration": 2.6395132541656494 }, { "auxiliary_loss_clip": 0.01166529, "auxiliary_loss_mlp": 0.01021342, "balance_loss_clip": 1.04678738, "balance_loss_mlp": 1.01523256, "epoch": 0.9693981843323514, "flos": 28366956405120.0, "grad_norm": 1.5836313434468476, "language_loss": 0.73656738, "learning_rate": 9.77904647471628e-09, "loss": 0.7584461, "num_input_tokens_seen": 174050235, "step": 8062, "time_per_iteration": 2.506577253341675 }, { "auxiliary_loss_clip": 0.01102182, "auxiliary_loss_mlp": 0.01023326, "balance_loss_clip": 1.04016447, "balance_loss_mlp": 1.01646805, "epoch": 0.9695184272229904, "flos": 23622869135040.0, "grad_norm": 1.4966438838967298, "language_loss": 0.73962051, "learning_rate": 9.702259878815454e-09, "loss": 0.76087558, "num_input_tokens_seen": 174070560, "step": 8063, "time_per_iteration": 3.322434902191162 }, { "auxiliary_loss_clip": 0.01157287, "auxiliary_loss_mlp": 0.01028088, "balance_loss_clip": 1.04770231, "balance_loss_mlp": 1.02059317, "epoch": 0.9696386701136296, "flos": 23294681532480.0, "grad_norm": 1.9198362536600797, "language_loss": 0.74461663, "learning_rate": 9.625775209499254e-09, "loss": 0.76647043, "num_input_tokens_seen": 174090565, "step": 8064, "time_per_iteration": 2.4814233779907227 }, { "auxiliary_loss_clip": 0.01119161, "auxiliary_loss_mlp": 0.0102208, "balance_loss_clip": 1.0414381, "balance_loss_mlp": 1.01542199, "epoch": 0.9697589130042686, "flos": 15121879340640.0, "grad_norm": 2.2809699473775122, "language_loss": 0.73947841, "learning_rate": 9.549592478370172e-09, "loss": 0.76089084, "num_input_tokens_seen": 174108745, "step": 8065, "time_per_iteration": 2.5363759994506836 }, { "auxiliary_loss_clip": 0.01153631, "auxiliary_loss_mlp": 0.01024593, "balance_loss_clip": 1.04502082, "balance_loss_mlp": 1.01785481, "epoch": 0.9698791558949077, "flos": 18879536121120.0, "grad_norm": 1.6883643056229423, "language_loss": 0.79090327, "learning_rate": 9.473711696985632e-09, "loss": 0.81268549, "num_input_tokens_seen": 174128075, "step": 8066, "time_per_iteration": 2.4653894901275635 }, { "auxiliary_loss_clip": 0.01136609, "auxiliary_loss_mlp": 0.01026528, "balance_loss_clip": 1.04454887, "balance_loss_mlp": 1.01963186, "epoch": 0.9699993987855468, "flos": 17931458556480.0, "grad_norm": 2.7858056583587807, "language_loss": 0.75882995, "learning_rate": 9.398132876856201e-09, "loss": 0.78046131, "num_input_tokens_seen": 174147040, "step": 8067, "time_per_iteration": 2.4909865856170654 }, { "auxiliary_loss_clip": 0.01020157, "auxiliary_loss_mlp": 0.01003977, "balance_loss_clip": 1.00725508, "balance_loss_mlp": 1.0027132, "epoch": 0.9701196416761859, "flos": 67182197096640.0, "grad_norm": 1.9636288429086064, "language_loss": 0.60864818, "learning_rate": 9.322856029447379e-09, "loss": 0.62888944, "num_input_tokens_seen": 174208225, "step": 8068, "time_per_iteration": 3.052464008331299 }, { "auxiliary_loss_clip": 0.01165173, "auxiliary_loss_mlp": 0.01025973, "balance_loss_clip": 1.04704213, "balance_loss_mlp": 1.01913881, "epoch": 0.970239884566825, "flos": 24277807661760.0, "grad_norm": 1.9549154893858842, "language_loss": 0.80268836, "learning_rate": 9.247881166178695e-09, "loss": 0.82459974, "num_input_tokens_seen": 174226935, "step": 8069, "time_per_iteration": 3.149327039718628 }, { "auxiliary_loss_clip": 0.01133089, "auxiliary_loss_mlp": 0.01025049, "balance_loss_clip": 1.04391778, "balance_loss_mlp": 1.01807225, "epoch": 0.970360127457464, "flos": 25301693152320.0, "grad_norm": 2.0995129914747244, "language_loss": 0.76656115, "learning_rate": 9.173208298423274e-09, "loss": 0.78814256, "num_input_tokens_seen": 174248140, "step": 8070, "time_per_iteration": 3.356992244720459 }, { "auxiliary_loss_clip": 0.01107153, "auxiliary_loss_mlp": 0.00762065, "balance_loss_clip": 1.04302168, "balance_loss_mlp": 1.00051093, "epoch": 0.9704803703481032, "flos": 29572477937760.0, "grad_norm": 1.5918239129342997, "language_loss": 0.76441258, "learning_rate": 9.09883743750961e-09, "loss": 0.78310478, "num_input_tokens_seen": 174271030, "step": 8071, "time_per_iteration": 2.6549177169799805 }, { "auxiliary_loss_clip": 0.01137981, "auxiliary_loss_mlp": 0.01024179, "balance_loss_clip": 1.04488826, "balance_loss_mlp": 1.01724994, "epoch": 0.9706006132387422, "flos": 17380049460480.0, "grad_norm": 1.7132458246684228, "language_loss": 0.83948755, "learning_rate": 9.024768594719124e-09, "loss": 0.8611092, "num_input_tokens_seen": 174289410, "step": 8072, "time_per_iteration": 2.49808931350708 }, { "auxiliary_loss_clip": 0.01128144, "auxiliary_loss_mlp": 0.0102439, "balance_loss_clip": 1.04609275, "balance_loss_mlp": 1.0177207, "epoch": 0.9707208561293813, "flos": 18186424254240.0, "grad_norm": 2.052113047053184, "language_loss": 0.72338527, "learning_rate": 8.95100178128816e-09, "loss": 0.7449106, "num_input_tokens_seen": 174308550, "step": 8073, "time_per_iteration": 2.5262277126312256 }, { "auxiliary_loss_clip": 0.01140359, "auxiliary_loss_mlp": 0.01026876, "balance_loss_clip": 1.04446661, "balance_loss_mlp": 1.01971745, "epoch": 0.9708410990200205, "flos": 31248392681280.0, "grad_norm": 2.049300405663811, "language_loss": 0.70372236, "learning_rate": 8.877537008407321e-09, "loss": 0.72539473, "num_input_tokens_seen": 174328600, "step": 8074, "time_per_iteration": 2.6002182960510254 }, { "auxiliary_loss_clip": 0.01144542, "auxiliary_loss_mlp": 0.01021485, "balance_loss_clip": 1.04635191, "balance_loss_mlp": 1.01453245, "epoch": 0.9709613419106595, "flos": 30554454724320.0, "grad_norm": 1.6184397699722737, "language_loss": 0.68914086, "learning_rate": 8.804374287221028e-09, "loss": 0.71080112, "num_input_tokens_seen": 174349835, "step": 8075, "time_per_iteration": 2.5816404819488525 }, { "auxiliary_loss_clip": 0.01117001, "auxiliary_loss_mlp": 0.01022749, "balance_loss_clip": 1.03824365, "balance_loss_mlp": 1.0159688, "epoch": 0.9710815848012986, "flos": 23730169846560.0, "grad_norm": 1.6787973963840763, "language_loss": 0.84688091, "learning_rate": 8.731513628827958e-09, "loss": 0.86827838, "num_input_tokens_seen": 174369200, "step": 8076, "time_per_iteration": 2.5808966159820557 }, { "auxiliary_loss_clip": 0.01153863, "auxiliary_loss_mlp": 0.01024909, "balance_loss_clip": 1.04649353, "balance_loss_mlp": 1.01844192, "epoch": 0.9712018276919377, "flos": 23761878133920.0, "grad_norm": 1.9725208732123267, "language_loss": 0.82819068, "learning_rate": 8.658955044280825e-09, "loss": 0.84997845, "num_input_tokens_seen": 174388125, "step": 8077, "time_per_iteration": 2.502106189727783 }, { "auxiliary_loss_clip": 0.01151074, "auxiliary_loss_mlp": 0.01021962, "balance_loss_clip": 1.04580402, "balance_loss_mlp": 1.01447582, "epoch": 0.9713220705825768, "flos": 23330987190720.0, "grad_norm": 1.4625022337116835, "language_loss": 0.77639371, "learning_rate": 8.586698544587268e-09, "loss": 0.79812407, "num_input_tokens_seen": 174409735, "step": 8078, "time_per_iteration": 2.5408506393432617 }, { "auxiliary_loss_clip": 0.01129882, "auxiliary_loss_mlp": 0.01030549, "balance_loss_clip": 1.04233074, "balance_loss_mlp": 1.02357864, "epoch": 0.9714423134732159, "flos": 22200950331360.0, "grad_norm": 3.310291326026227, "language_loss": 0.74190414, "learning_rate": 8.514744140707853e-09, "loss": 0.76350844, "num_input_tokens_seen": 174428875, "step": 8079, "time_per_iteration": 2.509185314178467 }, { "auxiliary_loss_clip": 0.01166251, "auxiliary_loss_mlp": 0.01027141, "balance_loss_clip": 1.04768133, "balance_loss_mlp": 1.02083206, "epoch": 0.971562556363855, "flos": 20229921117120.0, "grad_norm": 1.770711692856649, "language_loss": 0.76477116, "learning_rate": 8.443091843558515e-09, "loss": 0.78670508, "num_input_tokens_seen": 174447960, "step": 8080, "time_per_iteration": 2.4663469791412354 }, { "auxiliary_loss_clip": 0.01133195, "auxiliary_loss_mlp": 0.01027412, "balance_loss_clip": 1.04334402, "balance_loss_mlp": 1.02039099, "epoch": 0.9716827992544941, "flos": 24970201189440.0, "grad_norm": 2.6937116980390385, "language_loss": 0.6462127, "learning_rate": 8.37174166400878e-09, "loss": 0.66781878, "num_input_tokens_seen": 174463535, "step": 8081, "time_per_iteration": 3.62100887298584 }, { "auxiliary_loss_clip": 0.01167536, "auxiliary_loss_mlp": 0.01026544, "balance_loss_clip": 1.04874301, "balance_loss_mlp": 1.01983511, "epoch": 0.9718030421451331, "flos": 24681479937600.0, "grad_norm": 2.004421964718734, "language_loss": 0.84879726, "learning_rate": 8.300693612881992e-09, "loss": 0.87073809, "num_input_tokens_seen": 174483600, "step": 8082, "time_per_iteration": 2.4922444820404053 }, { "auxiliary_loss_clip": 0.01153195, "auxiliary_loss_mlp": 0.00761839, "balance_loss_clip": 1.04838002, "balance_loss_mlp": 1.00054193, "epoch": 0.9719232850357723, "flos": 22090704429120.0, "grad_norm": 1.867665746868384, "language_loss": 0.81297505, "learning_rate": 8.22994770095664e-09, "loss": 0.83212543, "num_input_tokens_seen": 174502175, "step": 8083, "time_per_iteration": 2.481926202774048 }, { "auxiliary_loss_clip": 0.01143389, "auxiliary_loss_mlp": 0.0102986, "balance_loss_clip": 1.0501678, "balance_loss_mlp": 1.02270746, "epoch": 0.9720435279264114, "flos": 23656912024800.0, "grad_norm": 2.0718109826522157, "language_loss": 0.75356317, "learning_rate": 8.159503938964585e-09, "loss": 0.77529562, "num_input_tokens_seen": 174519495, "step": 8084, "time_per_iteration": 2.532074213027954 }, { "auxiliary_loss_clip": 0.01115036, "auxiliary_loss_mlp": 0.01025482, "balance_loss_clip": 1.04075551, "balance_loss_mlp": 1.01913667, "epoch": 0.9721637708170504, "flos": 28365914813280.0, "grad_norm": 2.010343071494923, "language_loss": 0.70370841, "learning_rate": 8.089362337592164e-09, "loss": 0.72511357, "num_input_tokens_seen": 174543120, "step": 8085, "time_per_iteration": 2.588228464126587 }, { "auxiliary_loss_clip": 0.01135839, "auxiliary_loss_mlp": 0.01022985, "balance_loss_clip": 1.04487324, "balance_loss_mlp": 1.01618981, "epoch": 0.9722840137076896, "flos": 29130811906560.0, "grad_norm": 1.635411230485086, "language_loss": 0.71952188, "learning_rate": 8.019522907479536e-09, "loss": 0.74111009, "num_input_tokens_seen": 174563480, "step": 8086, "time_per_iteration": 2.5492441654205322 }, { "auxiliary_loss_clip": 0.0115565, "auxiliary_loss_mlp": 0.01021544, "balance_loss_clip": 1.04677892, "balance_loss_mlp": 1.0150888, "epoch": 0.9724042565983286, "flos": 19243957548000.0, "grad_norm": 2.009823391367598, "language_loss": 0.77696049, "learning_rate": 7.949985659221558e-09, "loss": 0.79873246, "num_input_tokens_seen": 174580745, "step": 8087, "time_per_iteration": 2.472752332687378 }, { "auxiliary_loss_clip": 0.01141634, "auxiliary_loss_mlp": 0.01026573, "balance_loss_clip": 1.04451013, "balance_loss_mlp": 1.02029645, "epoch": 0.9725244994889677, "flos": 23039679917760.0, "grad_norm": 2.0067798490267994, "language_loss": 0.79092556, "learning_rate": 7.880750603366904e-09, "loss": 0.81260765, "num_input_tokens_seen": 174599615, "step": 8088, "time_per_iteration": 2.510291814804077 }, { "auxiliary_loss_clip": 0.0113565, "auxiliary_loss_mlp": 0.01027531, "balance_loss_clip": 1.04446685, "balance_loss_mlp": 1.01972854, "epoch": 0.9726447423796069, "flos": 23367472433760.0, "grad_norm": 3.9628077659332277, "language_loss": 0.79639769, "learning_rate": 7.811817750418282e-09, "loss": 0.81802946, "num_input_tokens_seen": 174618375, "step": 8089, "time_per_iteration": 3.311692476272583 }, { "auxiliary_loss_clip": 0.01124891, "auxiliary_loss_mlp": 0.01031106, "balance_loss_clip": 1.04597688, "balance_loss_mlp": 1.02391791, "epoch": 0.9727649852702459, "flos": 26541652661280.0, "grad_norm": 1.6341359213268571, "language_loss": 0.79886484, "learning_rate": 7.743187110833105e-09, "loss": 0.82042485, "num_input_tokens_seen": 174641135, "step": 8090, "time_per_iteration": 2.574924945831299 }, { "auxiliary_loss_clip": 0.01140431, "auxiliary_loss_mlp": 0.01025947, "balance_loss_clip": 1.04375875, "balance_loss_mlp": 1.01920009, "epoch": 0.972885228160885, "flos": 20522341815840.0, "grad_norm": 1.472351632709017, "language_loss": 0.80807632, "learning_rate": 7.674858695022602e-09, "loss": 0.82974011, "num_input_tokens_seen": 174659490, "step": 8091, "time_per_iteration": 2.5374796390533447 }, { "auxiliary_loss_clip": 0.01169952, "auxiliary_loss_mlp": 0.01028081, "balance_loss_clip": 1.04836857, "balance_loss_mlp": 1.02096725, "epoch": 0.9730054710515241, "flos": 17566067371680.0, "grad_norm": 3.0899495920684656, "language_loss": 0.76298189, "learning_rate": 7.606832513351591e-09, "loss": 0.78496218, "num_input_tokens_seen": 174677440, "step": 8092, "time_per_iteration": 2.4167776107788086 }, { "auxiliary_loss_clip": 0.01059547, "auxiliary_loss_mlp": 0.00752857, "balance_loss_clip": 1.00550056, "balance_loss_mlp": 1.00007999, "epoch": 0.9731257139421632, "flos": 68972021272320.0, "grad_norm": 0.8317729416378574, "language_loss": 0.63899457, "learning_rate": 7.539108576140264e-09, "loss": 0.65711862, "num_input_tokens_seen": 174741550, "step": 8093, "time_per_iteration": 3.125661611557007 }, { "auxiliary_loss_clip": 0.01109989, "auxiliary_loss_mlp": 0.01029104, "balance_loss_clip": 1.04240203, "balance_loss_mlp": 1.02274132, "epoch": 0.9732459568328022, "flos": 18478844952960.0, "grad_norm": 2.0086367478418943, "language_loss": 0.70442545, "learning_rate": 7.471686893661732e-09, "loss": 0.72581637, "num_input_tokens_seen": 174759845, "step": 8094, "time_per_iteration": 3.307655096054077 }, { "auxiliary_loss_clip": 0.01139683, "auxiliary_loss_mlp": 0.01021963, "balance_loss_clip": 1.04768836, "balance_loss_mlp": 1.01519465, "epoch": 0.9733661997234414, "flos": 20883889885920.0, "grad_norm": 1.7214038630533985, "language_loss": 0.64470685, "learning_rate": 7.4045674761442636e-09, "loss": 0.66632324, "num_input_tokens_seen": 174777175, "step": 8095, "time_per_iteration": 2.5421178340911865 }, { "auxiliary_loss_clip": 0.01165959, "auxiliary_loss_mlp": 0.0076193, "balance_loss_clip": 1.04710746, "balance_loss_mlp": 1.00052309, "epoch": 0.9734864426140805, "flos": 23766798757440.0, "grad_norm": 1.6931118621376868, "language_loss": 0.7467494, "learning_rate": 7.337750333769488e-09, "loss": 0.76602829, "num_input_tokens_seen": 174796980, "step": 8096, "time_per_iteration": 3.220421552658081 }, { "auxiliary_loss_clip": 0.01141622, "auxiliary_loss_mlp": 0.01024464, "balance_loss_clip": 1.04021192, "balance_loss_mlp": 1.01735938, "epoch": 0.9736066855047195, "flos": 35042426953920.0, "grad_norm": 1.8089611567260138, "language_loss": 0.7277354, "learning_rate": 7.2712354766737425e-09, "loss": 0.74939632, "num_input_tokens_seen": 174817310, "step": 8097, "time_per_iteration": 2.6362507343292236 }, { "auxiliary_loss_clip": 0.01119372, "auxiliary_loss_mlp": 0.01026001, "balance_loss_clip": 1.04618573, "balance_loss_mlp": 1.01918244, "epoch": 0.9737269283953586, "flos": 20410623318240.0, "grad_norm": 1.4639446811478796, "language_loss": 0.80694985, "learning_rate": 7.2050229149469565e-09, "loss": 0.82840365, "num_input_tokens_seen": 174837320, "step": 8098, "time_per_iteration": 2.56540846824646 }, { "auxiliary_loss_clip": 0.01129607, "auxiliary_loss_mlp": 0.01024932, "balance_loss_clip": 1.04257393, "balance_loss_mlp": 1.01808012, "epoch": 0.9738471712859977, "flos": 28911684946560.0, "grad_norm": 1.884197756681256, "language_loss": 0.63462996, "learning_rate": 7.139112658633984e-09, "loss": 0.65617537, "num_input_tokens_seen": 174857470, "step": 8099, "time_per_iteration": 2.6083271503448486 }, { "auxiliary_loss_clip": 0.01122919, "auxiliary_loss_mlp": 0.01025983, "balance_loss_clip": 1.04303181, "balance_loss_mlp": 1.01913726, "epoch": 0.9739674141766368, "flos": 27782330509440.0, "grad_norm": 2.3002893771798005, "language_loss": 0.70465988, "learning_rate": 7.073504717733048e-09, "loss": 0.72614884, "num_input_tokens_seen": 174877035, "step": 8100, "time_per_iteration": 2.635921001434326 }, { "auxiliary_loss_clip": 0.01015973, "auxiliary_loss_mlp": 0.00999982, "balance_loss_clip": 1.0093708, "balance_loss_mlp": 0.99887896, "epoch": 0.9740876570672758, "flos": 68863068380640.0, "grad_norm": 0.7348310250070809, "language_loss": 0.57245874, "learning_rate": 7.008199102196855e-09, "loss": 0.59261835, "num_input_tokens_seen": 174938460, "step": 8101, "time_per_iteration": 3.1438474655151367 }, { "auxiliary_loss_clip": 0.01036026, "auxiliary_loss_mlp": 0.01003096, "balance_loss_clip": 1.00821173, "balance_loss_mlp": 1.00186253, "epoch": 0.974207899957915, "flos": 58236631997280.0, "grad_norm": 0.7932683276727331, "language_loss": 0.58947933, "learning_rate": 6.9431958219321464e-09, "loss": 0.60987055, "num_input_tokens_seen": 174994625, "step": 8102, "time_per_iteration": 3.102973699569702 }, { "auxiliary_loss_clip": 0.01138451, "auxiliary_loss_mlp": 0.01021833, "balance_loss_clip": 1.04385614, "balance_loss_mlp": 1.01458192, "epoch": 0.9743281428485541, "flos": 22600061153280.0, "grad_norm": 1.4602511269490626, "language_loss": 0.77657306, "learning_rate": 6.878494886800146e-09, "loss": 0.79817593, "num_input_tokens_seen": 175015400, "step": 8103, "time_per_iteration": 2.5275840759277344 }, { "auxiliary_loss_clip": 0.01136188, "auxiliary_loss_mlp": 0.01023196, "balance_loss_clip": 1.04513228, "balance_loss_mlp": 1.01612997, "epoch": 0.9744483857391931, "flos": 20008818724320.0, "grad_norm": 2.0479541981228353, "language_loss": 0.76322937, "learning_rate": 6.814096306615669e-09, "loss": 0.78482324, "num_input_tokens_seen": 175033540, "step": 8104, "time_per_iteration": 2.4968926906585693 }, { "auxiliary_loss_clip": 0.01144842, "auxiliary_loss_mlp": 0.01023513, "balance_loss_clip": 1.04260373, "balance_loss_mlp": 1.01631248, "epoch": 0.9745686286298323, "flos": 17675271682080.0, "grad_norm": 2.254655433638686, "language_loss": 0.65102148, "learning_rate": 6.750000091148011e-09, "loss": 0.67270505, "num_input_tokens_seen": 175050835, "step": 8105, "time_per_iteration": 2.4809861183166504 }, { "auxiliary_loss_clip": 0.0116766, "auxiliary_loss_mlp": 0.01024777, "balance_loss_clip": 1.04826462, "balance_loss_mlp": 1.01785076, "epoch": 0.9746888715204713, "flos": 29460256602720.0, "grad_norm": 1.99627353584876, "language_loss": 0.72662532, "learning_rate": 6.686206250120729e-09, "loss": 0.74854964, "num_input_tokens_seen": 175072330, "step": 8106, "time_per_iteration": 2.5306053161621094 }, { "auxiliary_loss_clip": 0.01131417, "auxiliary_loss_mlp": 0.01025522, "balance_loss_clip": 1.04178572, "balance_loss_mlp": 1.01854789, "epoch": 0.9748091144111104, "flos": 18479311873440.0, "grad_norm": 1.707409263255645, "language_loss": 0.74765086, "learning_rate": 6.622714793210749e-09, "loss": 0.76922023, "num_input_tokens_seen": 175091250, "step": 8107, "time_per_iteration": 3.253624439239502 }, { "auxiliary_loss_clip": 0.01168874, "auxiliary_loss_mlp": 0.01021621, "balance_loss_clip": 1.04765701, "balance_loss_mlp": 1.01500463, "epoch": 0.9749293573017496, "flos": 20665158012480.0, "grad_norm": 1.7750817815222166, "language_loss": 0.7856251, "learning_rate": 6.559525730050364e-09, "loss": 0.80753005, "num_input_tokens_seen": 175111350, "step": 8108, "time_per_iteration": 2.4818003177642822 }, { "auxiliary_loss_clip": 0.01127967, "auxiliary_loss_mlp": 0.01022692, "balance_loss_clip": 1.0447855, "balance_loss_mlp": 1.01593924, "epoch": 0.9750496001923886, "flos": 18478593534240.0, "grad_norm": 1.9384044004372605, "language_loss": 0.76063615, "learning_rate": 6.496639070224574e-09, "loss": 0.78214276, "num_input_tokens_seen": 175129835, "step": 8109, "time_per_iteration": 2.5109214782714844 }, { "auxiliary_loss_clip": 0.01158638, "auxiliary_loss_mlp": 0.01029556, "balance_loss_clip": 1.04786801, "balance_loss_mlp": 1.02326131, "epoch": 0.9751698430830277, "flos": 19572971240640.0, "grad_norm": 2.2190812452736672, "language_loss": 0.83643913, "learning_rate": 6.4340548232739714e-09, "loss": 0.85832113, "num_input_tokens_seen": 175146035, "step": 8110, "time_per_iteration": 2.4643375873565674 }, { "auxiliary_loss_clip": 0.01129469, "auxiliary_loss_mlp": 0.01025715, "balance_loss_clip": 1.04203129, "balance_loss_mlp": 1.0191648, "epoch": 0.9752900859736668, "flos": 23550329652480.0, "grad_norm": 1.6288662921887884, "language_loss": 0.79094285, "learning_rate": 6.371772998692071e-09, "loss": 0.8124947, "num_input_tokens_seen": 175165290, "step": 8111, "time_per_iteration": 2.556912899017334 }, { "auxiliary_loss_clip": 0.01130808, "auxiliary_loss_mlp": 0.01023532, "balance_loss_clip": 1.04189491, "balance_loss_mlp": 1.01668024, "epoch": 0.9754103288643059, "flos": 20303214855840.0, "grad_norm": 3.3377327720797005, "language_loss": 0.64976454, "learning_rate": 6.309793605927094e-09, "loss": 0.67130792, "num_input_tokens_seen": 175183610, "step": 8112, "time_per_iteration": 2.5656752586364746 }, { "auxiliary_loss_clip": 0.0114017, "auxiliary_loss_mlp": 0.01023502, "balance_loss_clip": 1.04263079, "balance_loss_mlp": 1.01716638, "epoch": 0.975530571754945, "flos": 19350683588160.0, "grad_norm": 2.031102167630552, "language_loss": 0.80138993, "learning_rate": 6.248116654381297e-09, "loss": 0.82302666, "num_input_tokens_seen": 175202080, "step": 8113, "time_per_iteration": 2.496474504470825 }, { "auxiliary_loss_clip": 0.0114261, "auxiliary_loss_mlp": 0.01023791, "balance_loss_clip": 1.04327011, "balance_loss_mlp": 1.01728261, "epoch": 0.9756508146455841, "flos": 23583402784320.0, "grad_norm": 1.720443387341181, "language_loss": 0.7256301, "learning_rate": 6.186742153410751e-09, "loss": 0.74729407, "num_input_tokens_seen": 175221575, "step": 8114, "time_per_iteration": 2.5594534873962402 }, { "auxiliary_loss_clip": 0.01139394, "auxiliary_loss_mlp": 0.0103639, "balance_loss_clip": 1.0449456, "balance_loss_mlp": 1.02867699, "epoch": 0.9757710575362232, "flos": 22966924933440.0, "grad_norm": 2.4713648538499777, "language_loss": 0.8720243, "learning_rate": 6.125670112326453e-09, "loss": 0.89378214, "num_input_tokens_seen": 175240835, "step": 8115, "time_per_iteration": 3.312169313430786 }, { "auxiliary_loss_clip": 0.01152145, "auxiliary_loss_mlp": 0.01025644, "balance_loss_clip": 1.04330993, "balance_loss_mlp": 1.01868558, "epoch": 0.9758913004268622, "flos": 27966013818240.0, "grad_norm": 1.5803261960854418, "language_loss": 0.70182174, "learning_rate": 6.064900540392548e-09, "loss": 0.72359961, "num_input_tokens_seen": 175262930, "step": 8116, "time_per_iteration": 2.5511019229888916 }, { "auxiliary_loss_clip": 0.01132076, "auxiliary_loss_mlp": 0.01022985, "balance_loss_clip": 1.0440377, "balance_loss_mlp": 1.0165354, "epoch": 0.9760115433175014, "flos": 22200160158240.0, "grad_norm": 2.5188936851209753, "language_loss": 0.78544188, "learning_rate": 6.0044334468278835e-09, "loss": 0.80699247, "num_input_tokens_seen": 175282275, "step": 8117, "time_per_iteration": 2.5491902828216553 }, { "auxiliary_loss_clip": 0.0111433, "auxiliary_loss_mlp": 0.01023978, "balance_loss_clip": 1.04269278, "balance_loss_mlp": 1.01615524, "epoch": 0.9761317862081405, "flos": 26250740474880.0, "grad_norm": 2.0000946709257064, "language_loss": 0.71563661, "learning_rate": 5.944268840805345e-09, "loss": 0.73701966, "num_input_tokens_seen": 175303020, "step": 8118, "time_per_iteration": 2.627474308013916 }, { "auxiliary_loss_clip": 0.01120896, "auxiliary_loss_mlp": 0.01020212, "balance_loss_clip": 1.04331541, "balance_loss_mlp": 1.01367307, "epoch": 0.9762520290987795, "flos": 26575444132320.0, "grad_norm": 1.982584248801096, "language_loss": 0.63977277, "learning_rate": 5.88440673145163e-09, "loss": 0.66118383, "num_input_tokens_seen": 175324070, "step": 8119, "time_per_iteration": 2.5954654216766357 }, { "auxiliary_loss_clip": 0.01152096, "auxiliary_loss_mlp": 0.0102306, "balance_loss_clip": 1.04804468, "balance_loss_mlp": 1.01620317, "epoch": 0.9763722719894187, "flos": 18005039630880.0, "grad_norm": 2.650073240839937, "language_loss": 0.82669139, "learning_rate": 5.824847127848142e-09, "loss": 0.84844297, "num_input_tokens_seen": 175342595, "step": 8120, "time_per_iteration": 3.250901937484741 }, { "auxiliary_loss_clip": 0.01117015, "auxiliary_loss_mlp": 0.01032189, "balance_loss_clip": 1.04583955, "balance_loss_mlp": 1.02501225, "epoch": 0.9764925148800577, "flos": 22455664610400.0, "grad_norm": 2.2584391458076047, "language_loss": 0.7907055, "learning_rate": 5.765590039029433e-09, "loss": 0.81219757, "num_input_tokens_seen": 175361915, "step": 8121, "time_per_iteration": 2.5830020904541016 }, { "auxiliary_loss_clip": 0.01166837, "auxiliary_loss_mlp": 0.01020297, "balance_loss_clip": 1.04878402, "balance_loss_mlp": 1.01369298, "epoch": 0.9766127577706968, "flos": 36757089708960.0, "grad_norm": 1.64795198782884, "language_loss": 0.7098009, "learning_rate": 5.706635473985422e-09, "loss": 0.73167223, "num_input_tokens_seen": 175385785, "step": 8122, "time_per_iteration": 3.3610692024230957 }, { "auxiliary_loss_clip": 0.0115205, "auxiliary_loss_mlp": 0.01018826, "balance_loss_clip": 1.04558396, "balance_loss_mlp": 1.01154852, "epoch": 0.976733000661336, "flos": 22309975056960.0, "grad_norm": 4.17963230928295, "language_loss": 0.85161704, "learning_rate": 5.6479834416591764e-09, "loss": 0.87332582, "num_input_tokens_seen": 175405145, "step": 8123, "time_per_iteration": 2.4938158988952637 }, { "auxiliary_loss_clip": 0.01151019, "auxiliary_loss_mlp": 0.00762052, "balance_loss_clip": 1.04555273, "balance_loss_mlp": 1.00047421, "epoch": 0.976853243551975, "flos": 25810942125600.0, "grad_norm": 1.775737244597795, "language_loss": 0.68645918, "learning_rate": 5.589633950947803e-09, "loss": 0.70558989, "num_input_tokens_seen": 175422645, "step": 8124, "time_per_iteration": 2.5211503505706787 }, { "auxiliary_loss_clip": 0.01136263, "auxiliary_loss_mlp": 0.01029443, "balance_loss_clip": 1.04331636, "balance_loss_mlp": 1.0214684, "epoch": 0.9769734864426141, "flos": 21397448894400.0, "grad_norm": 2.0827758943844303, "language_loss": 0.6987294, "learning_rate": 5.5315870107035535e-09, "loss": 0.72038645, "num_input_tokens_seen": 175440695, "step": 8125, "time_per_iteration": 2.503098964691162 }, { "auxiliary_loss_clip": 0.01137563, "auxiliary_loss_mlp": 0.01026473, "balance_loss_clip": 1.04781997, "balance_loss_mlp": 1.01975811, "epoch": 0.9770937293332532, "flos": 13990980474240.0, "grad_norm": 1.7676925263994534, "language_loss": 0.78700602, "learning_rate": 5.473842629731607e-09, "loss": 0.80864638, "num_input_tokens_seen": 175459195, "step": 8126, "time_per_iteration": 2.498810052871704 }, { "auxiliary_loss_clip": 0.01147836, "auxiliary_loss_mlp": 0.00762869, "balance_loss_clip": 1.04383302, "balance_loss_mlp": 1.0005362, "epoch": 0.9772139722238923, "flos": 17931997310880.0, "grad_norm": 1.9087433370848843, "language_loss": 0.78389537, "learning_rate": 5.416400816792066e-09, "loss": 0.80300248, "num_input_tokens_seen": 175476710, "step": 8127, "time_per_iteration": 2.488624334335327 }, { "auxiliary_loss_clip": 0.01165651, "auxiliary_loss_mlp": 0.0102361, "balance_loss_clip": 1.04674375, "balance_loss_mlp": 1.01662445, "epoch": 0.9773342151145313, "flos": 20446174720320.0, "grad_norm": 2.9393904960036004, "language_loss": 0.78653562, "learning_rate": 5.359261580598407e-09, "loss": 0.80842829, "num_input_tokens_seen": 175492550, "step": 8128, "time_per_iteration": 2.4387764930725098 }, { "auxiliary_loss_clip": 0.01157504, "auxiliary_loss_mlp": 0.0102224, "balance_loss_clip": 1.04744887, "balance_loss_mlp": 1.01437855, "epoch": 0.9774544580051704, "flos": 11837309543040.0, "grad_norm": 2.4579320810138783, "language_loss": 0.78159487, "learning_rate": 5.302424929819027e-09, "loss": 0.80339229, "num_input_tokens_seen": 175506560, "step": 8129, "time_per_iteration": 2.444946527481079 }, { "auxiliary_loss_clip": 0.01155416, "auxiliary_loss_mlp": 0.01025282, "balance_loss_clip": 1.04312587, "balance_loss_mlp": 1.01845467, "epoch": 0.9775747008958096, "flos": 13479935652960.0, "grad_norm": 2.8136857134034234, "language_loss": 0.7271657, "learning_rate": 5.24589087307592e-09, "loss": 0.74897265, "num_input_tokens_seen": 175524180, "step": 8130, "time_per_iteration": 2.484118700027466 }, { "auxiliary_loss_clip": 0.01169978, "auxiliary_loss_mlp": 0.01026964, "balance_loss_clip": 1.04811084, "balance_loss_mlp": 1.02004397, "epoch": 0.9776949437864486, "flos": 59532321851520.0, "grad_norm": 1.4647722406026376, "language_loss": 0.64947772, "learning_rate": 5.189659418944891e-09, "loss": 0.6714471, "num_input_tokens_seen": 175554355, "step": 8131, "time_per_iteration": 2.8449432849884033 }, { "auxiliary_loss_clip": 0.01168211, "auxiliary_loss_mlp": 0.01022008, "balance_loss_clip": 1.04857302, "balance_loss_mlp": 1.01495934, "epoch": 0.9778151866770877, "flos": 21178609270080.0, "grad_norm": 2.2547787939278097, "language_loss": 0.78642792, "learning_rate": 5.133730575956674e-09, "loss": 0.80833006, "num_input_tokens_seen": 175574025, "step": 8132, "time_per_iteration": 2.4850704669952393 }, { "auxiliary_loss_clip": 0.0114182, "auxiliary_loss_mlp": 0.01025996, "balance_loss_clip": 1.04534674, "balance_loss_mlp": 1.01856029, "epoch": 0.9779354295677268, "flos": 20886799159680.0, "grad_norm": 3.2930790185337675, "language_loss": 0.7204237, "learning_rate": 5.0781043525953696e-09, "loss": 0.74210185, "num_input_tokens_seen": 175592090, "step": 8133, "time_per_iteration": 3.2783751487731934 }, { "auxiliary_loss_clip": 0.01133579, "auxiliary_loss_mlp": 0.01023332, "balance_loss_clip": 1.04415965, "balance_loss_mlp": 1.01661754, "epoch": 0.9780556724583659, "flos": 23440622504640.0, "grad_norm": 1.8116161683120273, "language_loss": 0.73799735, "learning_rate": 5.0227807572995605e-09, "loss": 0.75956643, "num_input_tokens_seen": 175614065, "step": 8134, "time_per_iteration": 2.5970654487609863 }, { "auxiliary_loss_clip": 0.01139337, "auxiliary_loss_mlp": 0.01022271, "balance_loss_clip": 1.04203582, "balance_loss_mlp": 1.01542509, "epoch": 0.9781759153490049, "flos": 20923248485760.0, "grad_norm": 2.156582110293406, "language_loss": 0.67488825, "learning_rate": 4.967759798461646e-09, "loss": 0.69650435, "num_input_tokens_seen": 175632410, "step": 8135, "time_per_iteration": 2.512518882751465 }, { "auxiliary_loss_clip": 0.01164754, "auxiliary_loss_mlp": 0.01025321, "balance_loss_clip": 1.047225, "balance_loss_mlp": 1.01941729, "epoch": 0.9782961582396441, "flos": 28293195745920.0, "grad_norm": 1.9238805913476862, "language_loss": 0.74999428, "learning_rate": 4.913041484428282e-09, "loss": 0.77189505, "num_input_tokens_seen": 175652885, "step": 8136, "time_per_iteration": 2.535006284713745 }, { "auxiliary_loss_clip": 0.01155326, "auxiliary_loss_mlp": 0.01025577, "balance_loss_clip": 1.04608214, "balance_loss_mlp": 1.0190618, "epoch": 0.9784164011302832, "flos": 25552959403200.0, "grad_norm": 1.750518035816681, "language_loss": 0.74182308, "learning_rate": 4.858625823500384e-09, "loss": 0.76363212, "num_input_tokens_seen": 175670585, "step": 8137, "time_per_iteration": 2.5085678100585938 }, { "auxiliary_loss_clip": 0.01157022, "auxiliary_loss_mlp": 0.01023072, "balance_loss_clip": 1.04645085, "balance_loss_mlp": 1.0158602, "epoch": 0.9785366440209222, "flos": 29965949796960.0, "grad_norm": 1.9287208345353017, "language_loss": 0.73590326, "learning_rate": 4.80451282393246e-09, "loss": 0.75770426, "num_input_tokens_seen": 175690570, "step": 8138, "time_per_iteration": 2.573394298553467 }, { "auxiliary_loss_clip": 0.01138725, "auxiliary_loss_mlp": 0.0102571, "balance_loss_clip": 1.04587996, "balance_loss_mlp": 1.01894438, "epoch": 0.9786568869115614, "flos": 32343596477760.0, "grad_norm": 2.4012440494263214, "language_loss": 0.67513341, "learning_rate": 4.750702493933722e-09, "loss": 0.69677776, "num_input_tokens_seen": 175710455, "step": 8139, "time_per_iteration": 2.595562219619751 }, { "auxiliary_loss_clip": 0.01141054, "auxiliary_loss_mlp": 0.00761559, "balance_loss_clip": 1.04728436, "balance_loss_mlp": 1.00050807, "epoch": 0.9787771298022004, "flos": 23331418194240.0, "grad_norm": 1.8913966624494711, "language_loss": 0.85067689, "learning_rate": 4.697194841666974e-09, "loss": 0.86970294, "num_input_tokens_seen": 175729380, "step": 8140, "time_per_iteration": 2.5762908458709717 }, { "auxiliary_loss_clip": 0.01155845, "auxiliary_loss_mlp": 0.01025377, "balance_loss_clip": 1.04568779, "balance_loss_mlp": 1.01842439, "epoch": 0.9788973726928395, "flos": 21468551698560.0, "grad_norm": 1.943610548969255, "language_loss": 0.81992567, "learning_rate": 4.6439898752492764e-09, "loss": 0.84173787, "num_input_tokens_seen": 175749520, "step": 8141, "time_per_iteration": 3.30473256111145 }, { "auxiliary_loss_clip": 0.01051148, "auxiliary_loss_mlp": 0.00752782, "balance_loss_clip": 1.00622022, "balance_loss_mlp": 1.00013351, "epoch": 0.9790176155834787, "flos": 68897470440000.0, "grad_norm": 0.7476640653272564, "language_loss": 0.63669378, "learning_rate": 4.591087602751731e-09, "loss": 0.65473312, "num_input_tokens_seen": 175811380, "step": 8142, "time_per_iteration": 3.189567804336548 }, { "auxiliary_loss_clip": 0.01152045, "auxiliary_loss_mlp": 0.01024737, "balance_loss_clip": 1.04624248, "balance_loss_mlp": 1.01856208, "epoch": 0.9791378584741177, "flos": 21430881195840.0, "grad_norm": 1.6891686181746068, "language_loss": 0.718714, "learning_rate": 4.538488032199916e-09, "loss": 0.74048179, "num_input_tokens_seen": 175829480, "step": 8143, "time_per_iteration": 2.5007150173187256 }, { "auxiliary_loss_clip": 0.01156722, "auxiliary_loss_mlp": 0.0102356, "balance_loss_clip": 1.04382455, "balance_loss_mlp": 1.01676476, "epoch": 0.9792581013647568, "flos": 20153035682400.0, "grad_norm": 2.080762317021573, "language_loss": 0.68610001, "learning_rate": 4.486191171572784e-09, "loss": 0.70790285, "num_input_tokens_seen": 175846750, "step": 8144, "time_per_iteration": 2.4792423248291016 }, { "auxiliary_loss_clip": 0.01156284, "auxiliary_loss_mlp": 0.01020937, "balance_loss_clip": 1.04708004, "balance_loss_mlp": 1.01494646, "epoch": 0.9793783442553959, "flos": 23728194413760.0, "grad_norm": 1.5849203233421045, "language_loss": 0.77566582, "learning_rate": 4.434197028803766e-09, "loss": 0.79743803, "num_input_tokens_seen": 175865975, "step": 8145, "time_per_iteration": 2.5150504112243652 }, { "auxiliary_loss_clip": 0.01133148, "auxiliary_loss_mlp": 0.01032793, "balance_loss_clip": 1.04438412, "balance_loss_mlp": 1.02604616, "epoch": 0.979498587146035, "flos": 23038746076800.0, "grad_norm": 2.086885286987494, "language_loss": 0.8209976, "learning_rate": 4.3825056117805514e-09, "loss": 0.84265697, "num_input_tokens_seen": 175881860, "step": 8146, "time_per_iteration": 2.57879376411438 }, { "auxiliary_loss_clip": 0.01166684, "auxiliary_loss_mlp": 0.01025243, "balance_loss_clip": 1.04583442, "balance_loss_mlp": 1.01825702, "epoch": 0.979618830036674, "flos": 14318844824160.0, "grad_norm": 2.6702056366731726, "language_loss": 0.79162896, "learning_rate": 4.331116928344425e-09, "loss": 0.81354827, "num_input_tokens_seen": 175898175, "step": 8147, "time_per_iteration": 3.1794254779815674 }, { "auxiliary_loss_clip": 0.01143998, "auxiliary_loss_mlp": 0.00762258, "balance_loss_clip": 1.04313505, "balance_loss_mlp": 1.00050557, "epoch": 0.9797390729273132, "flos": 16727517370080.0, "grad_norm": 2.0511251301114606, "language_loss": 0.62469482, "learning_rate": 4.28003098629115e-09, "loss": 0.64375734, "num_input_tokens_seen": 175914310, "step": 8148, "time_per_iteration": 3.2751691341400146 }, { "auxiliary_loss_clip": 0.01122278, "auxiliary_loss_mlp": 0.01020374, "balance_loss_clip": 1.03873456, "balance_loss_mlp": 1.01358795, "epoch": 0.9798593158179523, "flos": 24532665608640.0, "grad_norm": 1.7617086003917715, "language_loss": 0.78331667, "learning_rate": 4.229247793370305e-09, "loss": 0.80474317, "num_input_tokens_seen": 175933435, "step": 8149, "time_per_iteration": 2.5787065029144287 }, { "auxiliary_loss_clip": 0.01169138, "auxiliary_loss_mlp": 0.01033378, "balance_loss_clip": 1.04785097, "balance_loss_mlp": 1.02646041, "epoch": 0.9799795587085913, "flos": 27308812523040.0, "grad_norm": 1.7517504407497275, "language_loss": 0.70307887, "learning_rate": 4.178767357285951e-09, "loss": 0.72510397, "num_input_tokens_seen": 175955065, "step": 8150, "time_per_iteration": 2.509169101715088 }, { "auxiliary_loss_clip": 0.01156734, "auxiliary_loss_mlp": 0.00761673, "balance_loss_clip": 1.04846036, "balance_loss_mlp": 1.00047278, "epoch": 0.9800998015992305, "flos": 26286579212640.0, "grad_norm": 4.650903798432152, "language_loss": 0.7141701, "learning_rate": 4.128589685695516e-09, "loss": 0.73335421, "num_input_tokens_seen": 175975490, "step": 8151, "time_per_iteration": 2.5506255626678467 }, { "auxiliary_loss_clip": 0.01168873, "auxiliary_loss_mlp": 0.01026597, "balance_loss_clip": 1.04850423, "balance_loss_mlp": 1.0198822, "epoch": 0.9802200444898695, "flos": 16723638338400.0, "grad_norm": 3.2859575228781233, "language_loss": 0.84542012, "learning_rate": 4.078714786211135e-09, "loss": 0.86737478, "num_input_tokens_seen": 175991340, "step": 8152, "time_per_iteration": 2.4102327823638916 }, { "auxiliary_loss_clip": 0.01150192, "auxiliary_loss_mlp": 0.01020721, "balance_loss_clip": 1.04563355, "balance_loss_mlp": 1.0143522, "epoch": 0.9803402873805086, "flos": 24900463229760.0, "grad_norm": 1.7208244854326802, "language_loss": 0.76460588, "learning_rate": 4.029142666398977e-09, "loss": 0.78631496, "num_input_tokens_seen": 176011505, "step": 8153, "time_per_iteration": 2.529961585998535 }, { "auxiliary_loss_clip": 0.01165623, "auxiliary_loss_mlp": 0.01026554, "balance_loss_clip": 1.04793513, "balance_loss_mlp": 1.02020621, "epoch": 0.9804605302711478, "flos": 22564940754720.0, "grad_norm": 1.9559481599855704, "language_loss": 0.80161142, "learning_rate": 3.979873333778805e-09, "loss": 0.82353318, "num_input_tokens_seen": 176029680, "step": 8154, "time_per_iteration": 2.5140862464904785 }, { "auxiliary_loss_clip": 0.01146256, "auxiliary_loss_mlp": 0.01030093, "balance_loss_clip": 1.04725492, "balance_loss_mlp": 1.0232172, "epoch": 0.9805807731617868, "flos": 38905373096160.0, "grad_norm": 2.012517107721438, "language_loss": 0.74126494, "learning_rate": 3.930906795824862e-09, "loss": 0.76302844, "num_input_tokens_seen": 176050355, "step": 8155, "time_per_iteration": 2.741706609725952 }, { "auxiliary_loss_clip": 0.01150049, "auxiliary_loss_mlp": 0.01022235, "balance_loss_clip": 1.04474854, "balance_loss_mlp": 1.01579142, "epoch": 0.9807010160524259, "flos": 17822002827360.0, "grad_norm": 1.984740879721692, "language_loss": 0.76924425, "learning_rate": 3.882243059965207e-09, "loss": 0.79096711, "num_input_tokens_seen": 176068070, "step": 8156, "time_per_iteration": 2.4837942123413086 }, { "auxiliary_loss_clip": 0.01145546, "auxiliary_loss_mlp": 0.01019724, "balance_loss_clip": 1.04321897, "balance_loss_mlp": 1.01214206, "epoch": 0.980821258943065, "flos": 13552977972960.0, "grad_norm": 2.407192617449048, "language_loss": 0.65526295, "learning_rate": 3.833882133582156e-09, "loss": 0.67691571, "num_input_tokens_seen": 176083730, "step": 8157, "time_per_iteration": 2.4872019290924072 }, { "auxiliary_loss_clip": 0.01155421, "auxiliary_loss_mlp": 0.01022427, "balance_loss_clip": 1.04552484, "balance_loss_mlp": 1.01564658, "epoch": 0.9809415018337041, "flos": 21689797759200.0, "grad_norm": 1.7878494195838806, "language_loss": 0.78371942, "learning_rate": 3.785824024012285e-09, "loss": 0.80549794, "num_input_tokens_seen": 176102730, "step": 8158, "time_per_iteration": 2.564223289489746 }, { "auxiliary_loss_clip": 0.01132812, "auxiliary_loss_mlp": 0.01027194, "balance_loss_clip": 1.0460788, "balance_loss_mlp": 1.02075386, "epoch": 0.9810617447243432, "flos": 23294861117280.0, "grad_norm": 1.4481379987942327, "language_loss": 0.78484309, "learning_rate": 3.738068738545541e-09, "loss": 0.8064431, "num_input_tokens_seen": 176121815, "step": 8159, "time_per_iteration": 3.2759816646575928 }, { "auxiliary_loss_clip": 0.01157888, "auxiliary_loss_mlp": 0.01032195, "balance_loss_clip": 1.04653955, "balance_loss_mlp": 1.0253675, "epoch": 0.9811819876149822, "flos": 18332041973760.0, "grad_norm": 2.5102678464336536, "language_loss": 0.78627706, "learning_rate": 3.6906162844265733e-09, "loss": 0.80817783, "num_input_tokens_seen": 176138900, "step": 8160, "time_per_iteration": 2.461444139480591 }, { "auxiliary_loss_clip": 0.01135593, "auxiliary_loss_mlp": 0.01025616, "balance_loss_clip": 1.04403424, "balance_loss_mlp": 1.01834393, "epoch": 0.9813022305056214, "flos": 22601964752160.0, "grad_norm": 1.7360747467719213, "language_loss": 0.70749927, "learning_rate": 3.643466668853845e-09, "loss": 0.72911137, "num_input_tokens_seen": 176156925, "step": 8161, "time_per_iteration": 2.567422866821289 }, { "auxiliary_loss_clip": 0.01141038, "auxiliary_loss_mlp": 0.01024729, "balance_loss_clip": 1.04365253, "balance_loss_mlp": 1.01792789, "epoch": 0.9814224733962604, "flos": 25413339816000.0, "grad_norm": 2.11944727625488, "language_loss": 0.75723195, "learning_rate": 3.59661989898008e-09, "loss": 0.7788896, "num_input_tokens_seen": 176177980, "step": 8162, "time_per_iteration": 2.5707991123199463 }, { "auxiliary_loss_clip": 0.01117512, "auxiliary_loss_mlp": 0.01022995, "balance_loss_clip": 1.04302561, "balance_loss_mlp": 1.01673055, "epoch": 0.9815427162868995, "flos": 25007189269920.0, "grad_norm": 3.3678027835526545, "language_loss": 0.76559132, "learning_rate": 3.5500759819115934e-09, "loss": 0.78699636, "num_input_tokens_seen": 176198345, "step": 8163, "time_per_iteration": 2.6196181774139404 }, { "auxiliary_loss_clip": 0.0117217, "auxiliary_loss_mlp": 0.01026593, "balance_loss_clip": 1.05057645, "balance_loss_mlp": 1.01929164, "epoch": 0.9816629591775387, "flos": 20662607908320.0, "grad_norm": 2.5468553616048744, "language_loss": 0.81028771, "learning_rate": 3.5038349247094034e-09, "loss": 0.83227533, "num_input_tokens_seen": 176215605, "step": 8164, "time_per_iteration": 2.4552934169769287 }, { "auxiliary_loss_clip": 0.01140145, "auxiliary_loss_mlp": 0.01026425, "balance_loss_clip": 1.04375291, "balance_loss_mlp": 1.01928473, "epoch": 0.9817832020681777, "flos": 17712223845600.0, "grad_norm": 2.1007282447958278, "language_loss": 0.7710346, "learning_rate": 3.4578967343878994e-09, "loss": 0.79270023, "num_input_tokens_seen": 176231810, "step": 8165, "time_per_iteration": 2.513758420944214 }, { "auxiliary_loss_clip": 0.01138753, "auxiliary_loss_mlp": 0.01026296, "balance_loss_clip": 1.04511213, "balance_loss_mlp": 1.01998103, "epoch": 0.9819034449588168, "flos": 22530035857920.0, "grad_norm": 1.8323379620938527, "language_loss": 0.81018847, "learning_rate": 3.4122614179161733e-09, "loss": 0.83183897, "num_input_tokens_seen": 176251770, "step": 8166, "time_per_iteration": 2.5487418174743652 }, { "auxiliary_loss_clip": 0.01114001, "auxiliary_loss_mlp": 0.01024693, "balance_loss_clip": 1.0405432, "balance_loss_mlp": 1.01780605, "epoch": 0.9820236878494559, "flos": 20011225160640.0, "grad_norm": 1.7006283515392162, "language_loss": 0.7811591, "learning_rate": 3.36692898221691e-09, "loss": 0.80254608, "num_input_tokens_seen": 176270135, "step": 8167, "time_per_iteration": 3.3056881427764893 }, { "auxiliary_loss_clip": 0.01154702, "auxiliary_loss_mlp": 0.01024661, "balance_loss_clip": 1.04621625, "balance_loss_mlp": 1.01807439, "epoch": 0.982143930740095, "flos": 18807319891200.0, "grad_norm": 2.2127840020729104, "language_loss": 0.7332747, "learning_rate": 3.3218994341668305e-09, "loss": 0.75506836, "num_input_tokens_seen": 176289065, "step": 8168, "time_per_iteration": 2.475557804107666 }, { "auxiliary_loss_clip": 0.01166227, "auxiliary_loss_mlp": 0.01031623, "balance_loss_clip": 1.04883456, "balance_loss_mlp": 1.02473569, "epoch": 0.982264173630734, "flos": 26578030153440.0, "grad_norm": 1.5059072697813762, "language_loss": 0.75558639, "learning_rate": 3.2771727805971373e-09, "loss": 0.77756488, "num_input_tokens_seen": 176310450, "step": 8169, "time_per_iteration": 2.4891340732574463 }, { "auxiliary_loss_clip": 0.01103509, "auxiliary_loss_mlp": 0.01022166, "balance_loss_clip": 1.03743267, "balance_loss_mlp": 1.01536489, "epoch": 0.9823844165213732, "flos": 22014465499680.0, "grad_norm": 1.7750134438091472, "language_loss": 0.77143919, "learning_rate": 3.232749028292847e-09, "loss": 0.79269588, "num_input_tokens_seen": 176327415, "step": 8170, "time_per_iteration": 2.5747954845428467 }, { "auxiliary_loss_clip": 0.01167932, "auxiliary_loss_mlp": 0.01027996, "balance_loss_clip": 1.04639339, "balance_loss_mlp": 1.02113247, "epoch": 0.9825046594120123, "flos": 21908170463040.0, "grad_norm": 1.7359871886391058, "language_loss": 0.88613319, "learning_rate": 3.188628183992792e-09, "loss": 0.9080925, "num_input_tokens_seen": 176347680, "step": 8171, "time_per_iteration": 2.49027681350708 }, { "auxiliary_loss_clip": 0.01051218, "auxiliary_loss_mlp": 0.01000445, "balance_loss_clip": 1.00515676, "balance_loss_mlp": 0.99915755, "epoch": 0.9826249023026513, "flos": 59494621820160.0, "grad_norm": 0.732867518650167, "language_loss": 0.6261121, "learning_rate": 3.1448102543902844e-09, "loss": 0.64662874, "num_input_tokens_seen": 176411595, "step": 8172, "time_per_iteration": 3.807446241378784 }, { "auxiliary_loss_clip": 0.01135492, "auxiliary_loss_mlp": 0.01023125, "balance_loss_clip": 1.04627585, "balance_loss_mlp": 1.01635385, "epoch": 0.9827451451932905, "flos": 16071034414080.0, "grad_norm": 1.948562111413412, "language_loss": 0.675354, "learning_rate": 3.1012952461324515e-09, "loss": 0.69694018, "num_input_tokens_seen": 176430570, "step": 8173, "time_per_iteration": 2.524872303009033 }, { "auxiliary_loss_clip": 0.01153031, "auxiliary_loss_mlp": 0.01026874, "balance_loss_clip": 1.04889631, "balance_loss_mlp": 1.02000499, "epoch": 0.9828653880839295, "flos": 20262778747200.0, "grad_norm": 2.472256423140893, "language_loss": 0.7359705, "learning_rate": 3.0580831658204575e-09, "loss": 0.75776958, "num_input_tokens_seen": 176448150, "step": 8174, "time_per_iteration": 3.2525315284729004 }, { "auxiliary_loss_clip": 0.0115227, "auxiliary_loss_mlp": 0.01029974, "balance_loss_clip": 1.04693484, "balance_loss_mlp": 1.02338731, "epoch": 0.9829856309745686, "flos": 21616144850880.0, "grad_norm": 1.6745849830657624, "language_loss": 0.77881861, "learning_rate": 3.015174020009281e-09, "loss": 0.800641, "num_input_tokens_seen": 176467475, "step": 8175, "time_per_iteration": 2.519629716873169 }, { "auxiliary_loss_clip": 0.01128137, "auxiliary_loss_mlp": 0.01023649, "balance_loss_clip": 1.04153776, "balance_loss_mlp": 1.01715207, "epoch": 0.9831058738652078, "flos": 23764212736320.0, "grad_norm": 1.794705975255735, "language_loss": 0.75207925, "learning_rate": 2.9725678152086043e-09, "loss": 0.77359706, "num_input_tokens_seen": 176486045, "step": 8176, "time_per_iteration": 2.5911061763763428 }, { "auxiliary_loss_clip": 0.01128679, "auxiliary_loss_mlp": 0.01024801, "balance_loss_clip": 1.04367948, "balance_loss_mlp": 1.01747537, "epoch": 0.9832261167558468, "flos": 11320913094720.0, "grad_norm": 3.5733863444368654, "language_loss": 0.82142508, "learning_rate": 2.930264557881257e-09, "loss": 0.84295988, "num_input_tokens_seen": 176501230, "step": 8177, "time_per_iteration": 2.532912015914917 }, { "auxiliary_loss_clip": 0.01059476, "auxiliary_loss_mlp": 0.01000121, "balance_loss_clip": 1.00540149, "balance_loss_mlp": 0.99883932, "epoch": 0.9833463596464859, "flos": 60000315014400.0, "grad_norm": 0.8472991873631369, "language_loss": 0.58238375, "learning_rate": 2.8882642544452163e-09, "loss": 0.60297966, "num_input_tokens_seen": 176565955, "step": 8178, "time_per_iteration": 3.0744965076446533 }, { "auxiliary_loss_clip": 0.01127696, "auxiliary_loss_mlp": 0.0102254, "balance_loss_clip": 1.04128027, "balance_loss_mlp": 1.01563764, "epoch": 0.983466602537125, "flos": 13626702715200.0, "grad_norm": 2.18414171232076, "language_loss": 0.74618638, "learning_rate": 2.8465669112716083e-09, "loss": 0.76768875, "num_input_tokens_seen": 176583480, "step": 8179, "time_per_iteration": 2.5299479961395264 }, { "auxiliary_loss_clip": 0.01155546, "auxiliary_loss_mlp": 0.00762006, "balance_loss_clip": 1.04578614, "balance_loss_mlp": 1.00055563, "epoch": 0.9835868454277641, "flos": 22926847994400.0, "grad_norm": 1.9298999077624057, "language_loss": 0.76133859, "learning_rate": 2.8051725346858177e-09, "loss": 0.78051412, "num_input_tokens_seen": 176603740, "step": 8180, "time_per_iteration": 2.4936025142669678 }, { "auxiliary_loss_clip": 0.01167952, "auxiliary_loss_mlp": 0.01023345, "balance_loss_clip": 1.04623306, "balance_loss_mlp": 1.01638937, "epoch": 0.9837070883184031, "flos": 27673413534720.0, "grad_norm": 2.2123395133986703, "language_loss": 0.70971912, "learning_rate": 2.7640811309674883e-09, "loss": 0.73163211, "num_input_tokens_seen": 176623240, "step": 8181, "time_per_iteration": 2.5417959690093994 }, { "auxiliary_loss_clip": 0.01116932, "auxiliary_loss_mlp": 0.01025656, "balance_loss_clip": 1.04357159, "balance_loss_mlp": 1.01897097, "epoch": 0.9838273312090423, "flos": 29241955732800.0, "grad_norm": 1.6741997477700397, "language_loss": 0.80778444, "learning_rate": 2.7232927063498557e-09, "loss": 0.82921034, "num_input_tokens_seen": 176643615, "step": 8182, "time_per_iteration": 2.5992088317871094 }, { "auxiliary_loss_clip": 0.01152842, "auxiliary_loss_mlp": 0.01022997, "balance_loss_clip": 1.04534495, "balance_loss_mlp": 1.01595771, "epoch": 0.9839475740996814, "flos": 40110212206560.0, "grad_norm": 2.3439744906694897, "language_loss": 0.68885493, "learning_rate": 2.682807267020859e-09, "loss": 0.71061325, "num_input_tokens_seen": 176666375, "step": 8183, "time_per_iteration": 2.6939585208892822 }, { "auxiliary_loss_clip": 0.0115329, "auxiliary_loss_mlp": 0.01021712, "balance_loss_clip": 1.0467999, "balance_loss_mlp": 1.01478016, "epoch": 0.9840678169903204, "flos": 24169393524480.0, "grad_norm": 2.0154004780313706, "language_loss": 0.62274379, "learning_rate": 2.642624819121808e-09, "loss": 0.64449382, "num_input_tokens_seen": 176686525, "step": 8184, "time_per_iteration": 2.5132217407226562 }, { "auxiliary_loss_clip": 0.01138738, "auxiliary_loss_mlp": 0.01021104, "balance_loss_clip": 1.04690194, "balance_loss_mlp": 1.01465797, "epoch": 0.9841880598809596, "flos": 14684487427680.0, "grad_norm": 1.9989357494342372, "language_loss": 0.61584949, "learning_rate": 2.6027453687487154e-09, "loss": 0.63744795, "num_input_tokens_seen": 176703615, "step": 8185, "time_per_iteration": 3.268890619277954 }, { "auxiliary_loss_clip": 0.01141028, "auxiliary_loss_mlp": 0.01021907, "balance_loss_clip": 1.04536104, "balance_loss_mlp": 1.01471233, "epoch": 0.9843083027715986, "flos": 22344772202880.0, "grad_norm": 2.332541020483216, "language_loss": 0.53750169, "learning_rate": 2.5631689219509643e-09, "loss": 0.55913103, "num_input_tokens_seen": 176722295, "step": 8186, "time_per_iteration": 2.5096075534820557 }, { "auxiliary_loss_clip": 0.01134847, "auxiliary_loss_mlp": 0.01019512, "balance_loss_clip": 1.04573774, "balance_loss_mlp": 1.01311934, "epoch": 0.9844285456622377, "flos": 21800115495360.0, "grad_norm": 1.676949809380621, "language_loss": 0.83511442, "learning_rate": 2.523895484732197e-09, "loss": 0.85665804, "num_input_tokens_seen": 176741750, "step": 8187, "time_per_iteration": 2.525838851928711 }, { "auxiliary_loss_clip": 0.0115788, "auxiliary_loss_mlp": 0.01026134, "balance_loss_clip": 1.04534197, "balance_loss_mlp": 1.01884699, "epoch": 0.9845487885528769, "flos": 18035383073760.0, "grad_norm": 1.9376289568206837, "language_loss": 0.74514222, "learning_rate": 2.4849250630505357e-09, "loss": 0.76698232, "num_input_tokens_seen": 176759995, "step": 8188, "time_per_iteration": 2.490772247314453 }, { "auxiliary_loss_clip": 0.01071151, "auxiliary_loss_mlp": 0.01021515, "balance_loss_clip": 1.03773713, "balance_loss_mlp": 1.01457667, "epoch": 0.9846690314435159, "flos": 25228614915360.0, "grad_norm": 1.7602661130672237, "language_loss": 0.73262167, "learning_rate": 2.4462576628172528e-09, "loss": 0.75354838, "num_input_tokens_seen": 176778625, "step": 8189, "time_per_iteration": 2.6734936237335205 }, { "auxiliary_loss_clip": 0.0114928, "auxiliary_loss_mlp": 0.01028189, "balance_loss_clip": 1.04450393, "balance_loss_mlp": 1.02110219, "epoch": 0.984789274334155, "flos": 18552174608640.0, "grad_norm": 1.8910772617912894, "language_loss": 0.74337518, "learning_rate": 2.407893289898766e-09, "loss": 0.76514983, "num_input_tokens_seen": 176797655, "step": 8190, "time_per_iteration": 2.4771957397460938 }, { "auxiliary_loss_clip": 0.01114881, "auxiliary_loss_mlp": 0.01025816, "balance_loss_clip": 1.04004383, "balance_loss_mlp": 1.018911, "epoch": 0.984909517224794, "flos": 27345441433920.0, "grad_norm": 3.1623812253131187, "language_loss": 0.8402307, "learning_rate": 2.3698319501144202e-09, "loss": 0.86163771, "num_input_tokens_seen": 176818640, "step": 8191, "time_per_iteration": 2.6177239418029785 }, { "auxiliary_loss_clip": 0.01154758, "auxiliary_loss_mlp": 0.01023616, "balance_loss_clip": 1.04535723, "balance_loss_mlp": 1.01662135, "epoch": 0.9850297601154332, "flos": 18734062069440.0, "grad_norm": 1.5357253473629717, "language_loss": 0.73152637, "learning_rate": 2.3320736492382644e-09, "loss": 0.75331008, "num_input_tokens_seen": 176837475, "step": 8192, "time_per_iteration": 2.472404956817627 }, { "auxiliary_loss_clip": 0.01165286, "auxiliary_loss_mlp": 0.0102469, "balance_loss_clip": 1.04739892, "balance_loss_mlp": 1.01796103, "epoch": 0.9851500030060723, "flos": 22308251042880.0, "grad_norm": 1.7496975483650288, "language_loss": 0.67928803, "learning_rate": 2.29461839299816e-09, "loss": 0.70118785, "num_input_tokens_seen": 176857190, "step": 8193, "time_per_iteration": 3.2769277095794678 }, { "auxiliary_loss_clip": 0.01124267, "auxiliary_loss_mlp": 0.01026131, "balance_loss_clip": 1.04289436, "balance_loss_mlp": 1.01932132, "epoch": 0.9852702458967113, "flos": 26353695234240.0, "grad_norm": 1.5470274580398147, "language_loss": 0.79771793, "learning_rate": 2.257466187076229e-09, "loss": 0.81922197, "num_input_tokens_seen": 176876395, "step": 8194, "time_per_iteration": 2.6043128967285156 }, { "auxiliary_loss_clip": 0.01158419, "auxiliary_loss_mlp": 0.00761929, "balance_loss_clip": 1.04679954, "balance_loss_mlp": 1.00051868, "epoch": 0.9853904887873505, "flos": 20883602550240.0, "grad_norm": 1.9208072735489707, "language_loss": 0.71320283, "learning_rate": 2.2206170371081854e-09, "loss": 0.73240632, "num_input_tokens_seen": 176894980, "step": 8195, "time_per_iteration": 2.513458013534546 }, { "auxiliary_loss_clip": 0.01141567, "auxiliary_loss_mlp": 0.01036074, "balance_loss_clip": 1.04504871, "balance_loss_mlp": 1.02887392, "epoch": 0.9855107316779895, "flos": 25263447978240.0, "grad_norm": 2.1374790932513608, "language_loss": 0.84904248, "learning_rate": 2.1840709486842247e-09, "loss": 0.87081885, "num_input_tokens_seen": 176914600, "step": 8196, "time_per_iteration": 2.564866065979004 }, { "auxiliary_loss_clip": 0.01132761, "auxiliary_loss_mlp": 0.01029622, "balance_loss_clip": 1.04351616, "balance_loss_mlp": 1.02291012, "epoch": 0.9856309745686286, "flos": 19062106004160.0, "grad_norm": 2.146475736587979, "language_loss": 0.7923826, "learning_rate": 2.1478279273481335e-09, "loss": 0.81400645, "num_input_tokens_seen": 176933085, "step": 8197, "time_per_iteration": 2.517979621887207 }, { "auxiliary_loss_clip": 0.01155247, "auxiliary_loss_mlp": 0.01023712, "balance_loss_clip": 1.04906988, "balance_loss_mlp": 1.01695299, "epoch": 0.9857512174592677, "flos": 34130762798400.0, "grad_norm": 3.8333646204098386, "language_loss": 0.80437881, "learning_rate": 2.1118879785981815e-09, "loss": 0.82616842, "num_input_tokens_seen": 176953225, "step": 8198, "time_per_iteration": 3.31778621673584 }, { "auxiliary_loss_clip": 0.01137759, "auxiliary_loss_mlp": 0.01025653, "balance_loss_clip": 1.04438353, "balance_loss_mlp": 1.01923621, "epoch": 0.9858714603499068, "flos": 25994697268320.0, "grad_norm": 2.2363566901929066, "language_loss": 0.79318976, "learning_rate": 2.0762511078862288e-09, "loss": 0.81482381, "num_input_tokens_seen": 176973570, "step": 8199, "time_per_iteration": 2.5688905715942383 }, { "auxiliary_loss_clip": 0.01142491, "auxiliary_loss_mlp": 0.01022333, "balance_loss_clip": 1.04377794, "balance_loss_mlp": 1.01588416, "epoch": 0.9859917032405459, "flos": 23696234707680.0, "grad_norm": 2.243857098412969, "language_loss": 0.65180361, "learning_rate": 2.0409173206186183e-09, "loss": 0.6734519, "num_input_tokens_seen": 176992810, "step": 8200, "time_per_iteration": 3.288823366165161 }, { "auxiliary_loss_clip": 0.01125237, "auxiliary_loss_mlp": 0.01022965, "balance_loss_clip": 1.04760003, "balance_loss_mlp": 1.01619959, "epoch": 0.986111946131185, "flos": 19938290591520.0, "grad_norm": 1.9214031461881866, "language_loss": 0.87335753, "learning_rate": 2.0058866221550617e-09, "loss": 0.89483958, "num_input_tokens_seen": 177011050, "step": 8201, "time_per_iteration": 2.528109073638916 }, { "auxiliary_loss_clip": 0.01166207, "auxiliary_loss_mlp": 0.01024689, "balance_loss_clip": 1.04535913, "balance_loss_mlp": 1.01809645, "epoch": 0.9862321890218241, "flos": 19828834862400.0, "grad_norm": 1.988692046297203, "language_loss": 0.75311625, "learning_rate": 1.971159017809976e-09, "loss": 0.77502525, "num_input_tokens_seen": 177029340, "step": 8202, "time_per_iteration": 2.439517021179199 }, { "auxiliary_loss_clip": 0.01152664, "auxiliary_loss_mlp": 0.01025499, "balance_loss_clip": 1.04582965, "balance_loss_mlp": 1.01841164, "epoch": 0.9863524319124631, "flos": 21652055422560.0, "grad_norm": 2.221098757996728, "language_loss": 0.77790856, "learning_rate": 1.93673451285159e-09, "loss": 0.79969019, "num_input_tokens_seen": 177048390, "step": 8203, "time_per_iteration": 2.4892029762268066 }, { "auxiliary_loss_clip": 0.01042314, "auxiliary_loss_mlp": 0.01000238, "balance_loss_clip": 1.00563991, "balance_loss_mlp": 0.9988969, "epoch": 0.9864726748031023, "flos": 52769987341440.0, "grad_norm": 0.7355321804061518, "language_loss": 0.56556183, "learning_rate": 1.9026131125019495e-09, "loss": 0.58598733, "num_input_tokens_seen": 177105760, "step": 8204, "time_per_iteration": 3.04736328125 }, { "auxiliary_loss_clip": 0.01147751, "auxiliary_loss_mlp": 0.01026677, "balance_loss_clip": 1.04617143, "balance_loss_mlp": 1.02010834, "epoch": 0.9865929176937414, "flos": 23364634993920.0, "grad_norm": 1.644690735471355, "language_loss": 0.86814964, "learning_rate": 1.8687948219371363e-09, "loss": 0.88989401, "num_input_tokens_seen": 177124985, "step": 8205, "time_per_iteration": 2.4900734424591064 }, { "auxiliary_loss_clip": 0.01170165, "auxiliary_loss_mlp": 0.0103029, "balance_loss_clip": 1.04684043, "balance_loss_mlp": 1.02229702, "epoch": 0.9867131605843804, "flos": 21616683605280.0, "grad_norm": 2.1345013088208966, "language_loss": 0.88438284, "learning_rate": 1.835279646287491e-09, "loss": 0.90638739, "num_input_tokens_seen": 177142995, "step": 8206, "time_per_iteration": 2.43799090385437 }, { "auxiliary_loss_clip": 0.01161459, "auxiliary_loss_mlp": 0.01029495, "balance_loss_clip": 1.04784536, "balance_loss_mlp": 1.02205598, "epoch": 0.9868334034750196, "flos": 22271406630240.0, "grad_norm": 2.1013862989193326, "language_loss": 0.7661981, "learning_rate": 1.8020675906371685e-09, "loss": 0.78810763, "num_input_tokens_seen": 177162390, "step": 8207, "time_per_iteration": 2.506932020187378 }, { "auxiliary_loss_clip": 0.01108368, "auxiliary_loss_mlp": 0.01017284, "balance_loss_clip": 1.04033875, "balance_loss_mlp": 1.01079893, "epoch": 0.9869536463656586, "flos": 25809577281120.0, "grad_norm": 2.0879301440089804, "language_loss": 0.74858063, "learning_rate": 1.7691586600243612e-09, "loss": 0.76983714, "num_input_tokens_seen": 177181290, "step": 8208, "time_per_iteration": 2.6137404441833496 }, { "auxiliary_loss_clip": 0.01139905, "auxiliary_loss_mlp": 0.01024894, "balance_loss_clip": 1.04880834, "balance_loss_mlp": 1.01812255, "epoch": 0.9870738892562977, "flos": 16398503677440.0, "grad_norm": 2.495815068912202, "language_loss": 0.87171483, "learning_rate": 1.7365528594415202e-09, "loss": 0.89336276, "num_input_tokens_seen": 177195360, "step": 8209, "time_per_iteration": 2.512491464614868 }, { "auxiliary_loss_clip": 0.01158196, "auxiliary_loss_mlp": 0.00762077, "balance_loss_clip": 1.04618466, "balance_loss_mlp": 1.00053418, "epoch": 0.9871941321469369, "flos": 35481363296160.0, "grad_norm": 1.626297036560485, "language_loss": 0.67659497, "learning_rate": 1.7042501938346888e-09, "loss": 0.69579768, "num_input_tokens_seen": 177218090, "step": 8210, "time_per_iteration": 2.6155097484588623 }, { "auxiliary_loss_clip": 0.01126434, "auxiliary_loss_mlp": 0.01023255, "balance_loss_clip": 1.03922236, "balance_loss_mlp": 1.01718473, "epoch": 0.9873143750375759, "flos": 21434221473120.0, "grad_norm": 2.320504310762811, "language_loss": 0.7655226, "learning_rate": 1.6722506681043913e-09, "loss": 0.78701949, "num_input_tokens_seen": 177237050, "step": 8211, "time_per_iteration": 3.2438392639160156 }, { "auxiliary_loss_clip": 0.01145511, "auxiliary_loss_mlp": 0.01027725, "balance_loss_clip": 1.04665565, "balance_loss_mlp": 1.02109361, "epoch": 0.987434617928215, "flos": 16326502949280.0, "grad_norm": 2.1088064081475206, "language_loss": 0.69105285, "learning_rate": 1.640554287104745e-09, "loss": 0.71278512, "num_input_tokens_seen": 177255325, "step": 8212, "time_per_iteration": 2.532593250274658 }, { "auxiliary_loss_clip": 0.01125256, "auxiliary_loss_mlp": 0.01027375, "balance_loss_clip": 1.03864086, "balance_loss_mlp": 1.0203563, "epoch": 0.9875548608188541, "flos": 17851987100640.0, "grad_norm": 2.0378570293717497, "language_loss": 0.7974323, "learning_rate": 1.609161055644348e-09, "loss": 0.81895852, "num_input_tokens_seen": 177271250, "step": 8213, "time_per_iteration": 2.5259108543395996 }, { "auxiliary_loss_clip": 0.0115899, "auxiliary_loss_mlp": 0.01017803, "balance_loss_clip": 1.04603875, "balance_loss_mlp": 1.01062083, "epoch": 0.9876751037094932, "flos": 26132880177120.0, "grad_norm": 2.305764595117502, "language_loss": 0.68296355, "learning_rate": 1.5780709784849467e-09, "loss": 0.70473146, "num_input_tokens_seen": 177288270, "step": 8214, "time_per_iteration": 2.5404186248779297 }, { "auxiliary_loss_clip": 0.01101149, "auxiliary_loss_mlp": 0.01026505, "balance_loss_clip": 1.04415369, "balance_loss_mlp": 1.01951313, "epoch": 0.9877953466001322, "flos": 15991347456480.0, "grad_norm": 2.1976639544669765, "language_loss": 0.82485569, "learning_rate": 1.5472840603436565e-09, "loss": 0.84613216, "num_input_tokens_seen": 177305500, "step": 8215, "time_per_iteration": 2.584747552871704 }, { "auxiliary_loss_clip": 0.01139832, "auxiliary_loss_mlp": 0.01020808, "balance_loss_clip": 1.04522395, "balance_loss_mlp": 1.01398039, "epoch": 0.9879155894907714, "flos": 18806781136800.0, "grad_norm": 2.2453983545493994, "language_loss": 0.77957523, "learning_rate": 1.5168003058900757e-09, "loss": 0.80118155, "num_input_tokens_seen": 177323500, "step": 8216, "time_per_iteration": 2.510396957397461 }, { "auxiliary_loss_clip": 0.01124044, "auxiliary_loss_mlp": 0.01026054, "balance_loss_clip": 1.04229462, "balance_loss_mlp": 1.01947355, "epoch": 0.9880358323814105, "flos": 22382047619040.0, "grad_norm": 2.0444527904799763, "language_loss": 0.91786236, "learning_rate": 1.4866197197491715e-09, "loss": 0.9393633, "num_input_tokens_seen": 177342860, "step": 8217, "time_per_iteration": 2.5500545501708984 }, { "auxiliary_loss_clip": 0.01158864, "auxiliary_loss_mlp": 0.00762642, "balance_loss_clip": 1.04804397, "balance_loss_mlp": 1.0005362, "epoch": 0.9881560752720495, "flos": 15668834733600.0, "grad_norm": 4.15897797731218, "language_loss": 0.78741705, "learning_rate": 1.4567423064988371e-09, "loss": 0.80663216, "num_input_tokens_seen": 177360210, "step": 8218, "time_per_iteration": 2.4714465141296387 }, { "auxiliary_loss_clip": 0.01169722, "auxiliary_loss_mlp": 0.01025323, "balance_loss_clip": 1.04751265, "balance_loss_mlp": 1.01809263, "epoch": 0.9882763181626887, "flos": 21500116318080.0, "grad_norm": 1.9268312556784026, "language_loss": 0.7807197, "learning_rate": 1.4271680706718913e-09, "loss": 0.80267012, "num_input_tokens_seen": 177377885, "step": 8219, "time_per_iteration": 3.2223777770996094 }, { "auxiliary_loss_clip": 0.01159401, "auxiliary_loss_mlp": 0.01033857, "balance_loss_clip": 1.04940796, "balance_loss_mlp": 1.02612591, "epoch": 0.9883965610533277, "flos": 28034602435200.0, "grad_norm": 1.6546315321190521, "language_loss": 0.8243928, "learning_rate": 1.3978970167543013e-09, "loss": 0.8463254, "num_input_tokens_seen": 177398065, "step": 8220, "time_per_iteration": 2.5477375984191895 }, { "auxiliary_loss_clip": 0.01131777, "auxiliary_loss_mlp": 0.01024084, "balance_loss_clip": 1.04434478, "balance_loss_mlp": 1.01717234, "epoch": 0.9885168039439668, "flos": 14098604438400.0, "grad_norm": 2.1436034930262493, "language_loss": 0.77265793, "learning_rate": 1.3689291491867372e-09, "loss": 0.79421651, "num_input_tokens_seen": 177416380, "step": 8221, "time_per_iteration": 2.4782609939575195 }, { "auxiliary_loss_clip": 0.01168663, "auxiliary_loss_mlp": 0.0102443, "balance_loss_clip": 1.04755807, "balance_loss_mlp": 1.01752734, "epoch": 0.988637046834606, "flos": 26432017347360.0, "grad_norm": 2.010743494218711, "language_loss": 0.73389947, "learning_rate": 1.3402644723636836e-09, "loss": 0.75583041, "num_input_tokens_seen": 177438410, "step": 8222, "time_per_iteration": 2.5033645629882812 }, { "auxiliary_loss_clip": 0.01138374, "auxiliary_loss_mlp": 0.01030134, "balance_loss_clip": 1.04804206, "balance_loss_mlp": 1.02291286, "epoch": 0.988757289725245, "flos": 25229117752800.0, "grad_norm": 2.2724834244306145, "language_loss": 0.83125299, "learning_rate": 1.311902990633218e-09, "loss": 0.85293806, "num_input_tokens_seen": 177457375, "step": 8223, "time_per_iteration": 2.5279111862182617 }, { "auxiliary_loss_clip": 0.0113143, "auxiliary_loss_mlp": 0.01023231, "balance_loss_clip": 1.04046571, "balance_loss_mlp": 1.01648116, "epoch": 0.9888775326158841, "flos": 26359046861280.0, "grad_norm": 1.7064104514831298, "language_loss": 0.71190357, "learning_rate": 1.2838447082978987e-09, "loss": 0.73345017, "num_input_tokens_seen": 177478530, "step": 8224, "time_per_iteration": 3.2552146911621094 }, { "auxiliary_loss_clip": 0.01150439, "auxiliary_loss_mlp": 0.01024746, "balance_loss_clip": 1.04398799, "balance_loss_mlp": 1.01765573, "epoch": 0.9889977755065231, "flos": 24316124669760.0, "grad_norm": 2.639749828666929, "language_loss": 0.83123219, "learning_rate": 1.2560896296143208e-09, "loss": 0.85298401, "num_input_tokens_seen": 177496995, "step": 8225, "time_per_iteration": 2.4881789684295654 }, { "auxiliary_loss_clip": 0.01169118, "auxiliary_loss_mlp": 0.01023732, "balance_loss_clip": 1.04851949, "balance_loss_mlp": 1.01660347, "epoch": 0.9891180183971623, "flos": 18951069928800.0, "grad_norm": 2.1841051293722167, "language_loss": 0.82007861, "learning_rate": 1.2286377587926722e-09, "loss": 0.84200716, "num_input_tokens_seen": 177513785, "step": 8226, "time_per_iteration": 3.218830108642578 }, { "auxiliary_loss_clip": 0.01166572, "auxiliary_loss_mlp": 0.01019445, "balance_loss_clip": 1.04617929, "balance_loss_mlp": 1.01322269, "epoch": 0.9892382612878013, "flos": 26176584729120.0, "grad_norm": 3.5584824586817834, "language_loss": 0.74846745, "learning_rate": 1.2014890999973992e-09, "loss": 0.77032757, "num_input_tokens_seen": 177530705, "step": 8227, "time_per_iteration": 2.4990639686584473 }, { "auxiliary_loss_clip": 0.01165771, "auxiliary_loss_mlp": 0.0102099, "balance_loss_clip": 1.04646635, "balance_loss_mlp": 1.01438618, "epoch": 0.9893585041784404, "flos": 25449609557280.0, "grad_norm": 1.6345638120223127, "language_loss": 0.78737271, "learning_rate": 1.1746436573472073e-09, "loss": 0.80924028, "num_input_tokens_seen": 177552440, "step": 8228, "time_per_iteration": 2.5012047290802 }, { "auxiliary_loss_clip": 0.01149494, "auxiliary_loss_mlp": 0.01027431, "balance_loss_clip": 1.04703593, "balance_loss_mlp": 1.02073431, "epoch": 0.9894787470690796, "flos": 20189305423680.0, "grad_norm": 1.9557827869877835, "language_loss": 0.69532216, "learning_rate": 1.1481014349141726e-09, "loss": 0.71709138, "num_input_tokens_seen": 177569660, "step": 8229, "time_per_iteration": 2.508206367492676 }, { "auxiliary_loss_clip": 0.01142612, "auxiliary_loss_mlp": 0.01028872, "balance_loss_clip": 1.04648662, "balance_loss_mlp": 1.0217458, "epoch": 0.9895989899597186, "flos": 24644312272320.0, "grad_norm": 2.1633741586632107, "language_loss": 0.84605849, "learning_rate": 1.121862436724852e-09, "loss": 0.86777341, "num_input_tokens_seen": 177588500, "step": 8230, "time_per_iteration": 2.523656129837036 }, { "auxiliary_loss_clip": 0.011556, "auxiliary_loss_mlp": 0.01029242, "balance_loss_clip": 1.04805362, "balance_loss_mlp": 1.02204823, "epoch": 0.9897192328503577, "flos": 21799037986560.0, "grad_norm": 1.607046858459395, "language_loss": 0.70313454, "learning_rate": 1.0959266667598388e-09, "loss": 0.72498298, "num_input_tokens_seen": 177607315, "step": 8231, "time_per_iteration": 2.497019052505493 }, { "auxiliary_loss_clip": 0.01129888, "auxiliary_loss_mlp": 0.01031751, "balance_loss_clip": 1.04652834, "balance_loss_mlp": 1.02373457, "epoch": 0.9898394757409968, "flos": 21325232664480.0, "grad_norm": 2.2246089878618536, "language_loss": 0.74869871, "learning_rate": 1.0702941289533196e-09, "loss": 0.77031511, "num_input_tokens_seen": 177625990, "step": 8232, "time_per_iteration": 2.537625312805176 }, { "auxiliary_loss_clip": 0.01122161, "auxiliary_loss_mlp": 0.01027212, "balance_loss_clip": 1.04119539, "balance_loss_mlp": 1.02083468, "epoch": 0.9899597186316359, "flos": 18545027133600.0, "grad_norm": 2.057453047006282, "language_loss": 0.89215481, "learning_rate": 1.0449648271939615e-09, "loss": 0.91364849, "num_input_tokens_seen": 177642335, "step": 8233, "time_per_iteration": 2.5485641956329346 }, { "auxiliary_loss_clip": 0.01118068, "auxiliary_loss_mlp": 0.00761587, "balance_loss_clip": 1.04692769, "balance_loss_mlp": 1.00053668, "epoch": 0.990079961522275, "flos": 23766726923520.0, "grad_norm": 1.563831833831272, "language_loss": 0.72544032, "learning_rate": 1.0199387653240243e-09, "loss": 0.74423683, "num_input_tokens_seen": 177662025, "step": 8234, "time_per_iteration": 2.60733699798584 }, { "auxiliary_loss_clip": 0.0113197, "auxiliary_loss_mlp": 0.01024396, "balance_loss_clip": 1.04436791, "balance_loss_mlp": 1.01809573, "epoch": 0.9902002044129141, "flos": 16399186099680.0, "grad_norm": 1.740778149131406, "language_loss": 0.70762277, "learning_rate": 9.952159471400267e-10, "loss": 0.72918636, "num_input_tokens_seen": 177679065, "step": 8235, "time_per_iteration": 2.523419141769409 }, { "auxiliary_loss_clip": 0.01152515, "auxiliary_loss_mlp": 0.0076142, "balance_loss_clip": 1.04607165, "balance_loss_mlp": 1.00050497, "epoch": 0.9903204473035532, "flos": 22559660961600.0, "grad_norm": 1.9141395166389654, "language_loss": 0.84665895, "learning_rate": 9.707963763923022e-10, "loss": 0.86579823, "num_input_tokens_seen": 177698115, "step": 8236, "time_per_iteration": 2.4940855503082275 }, { "auxiliary_loss_clip": 0.01139184, "auxiliary_loss_mlp": 0.01027825, "balance_loss_clip": 1.043787, "balance_loss_mlp": 1.0209527, "epoch": 0.9904406901941922, "flos": 16144004900160.0, "grad_norm": 1.8172362746002464, "language_loss": 0.78747869, "learning_rate": 9.466800567854427e-10, "loss": 0.80914879, "num_input_tokens_seen": 177716715, "step": 8237, "time_per_iteration": 3.2102625370025635 }, { "auxiliary_loss_clip": 0.01122952, "auxiliary_loss_mlp": 0.01028564, "balance_loss_clip": 1.04148269, "balance_loss_mlp": 1.02103353, "epoch": 0.9905609330848314, "flos": 26651503476960.0, "grad_norm": 1.9766932176903875, "language_loss": 0.67964685, "learning_rate": 9.228669919778553e-10, "loss": 0.70116204, "num_input_tokens_seen": 177735640, "step": 8238, "time_per_iteration": 2.5826892852783203 }, { "auxiliary_loss_clip": 0.01130755, "auxiliary_loss_mlp": 0.01022926, "balance_loss_clip": 1.04291666, "balance_loss_mlp": 1.01603603, "epoch": 0.9906811759754705, "flos": 23111824313760.0, "grad_norm": 2.902545510145575, "language_loss": 0.79686546, "learning_rate": 8.993571855817617e-10, "loss": 0.81840229, "num_input_tokens_seen": 177754470, "step": 8239, "time_per_iteration": 2.547873020172119 }, { "auxiliary_loss_clip": 0.01153387, "auxiliary_loss_mlp": 0.01022092, "balance_loss_clip": 1.046345, "balance_loss_mlp": 1.01549983, "epoch": 0.9908014188661095, "flos": 22090596678240.0, "grad_norm": 2.0101166422482755, "language_loss": 0.74929082, "learning_rate": 8.761506411638642e-10, "loss": 0.77104568, "num_input_tokens_seen": 177773935, "step": 8240, "time_per_iteration": 2.470427989959717 }, { "auxiliary_loss_clip": 0.01139431, "auxiliary_loss_mlp": 0.01028608, "balance_loss_clip": 1.04622555, "balance_loss_mlp": 1.0216136, "epoch": 0.9909216617567487, "flos": 19242951873120.0, "grad_norm": 1.976338419188817, "language_loss": 0.73585999, "learning_rate": 8.53247362244236e-10, "loss": 0.75754035, "num_input_tokens_seen": 177792745, "step": 8241, "time_per_iteration": 2.5065579414367676 }, { "auxiliary_loss_clip": 0.01140247, "auxiliary_loss_mlp": 0.01025232, "balance_loss_clip": 1.04580283, "balance_loss_mlp": 1.01830649, "epoch": 0.9910419046473877, "flos": 23621216954880.0, "grad_norm": 1.5932774231705058, "language_loss": 0.68331885, "learning_rate": 8.306473522976532e-10, "loss": 0.7049737, "num_input_tokens_seen": 177812150, "step": 8242, "time_per_iteration": 2.521763324737549 }, { "auxiliary_loss_clip": 0.01167242, "auxiliary_loss_mlp": 0.01024757, "balance_loss_clip": 1.04727387, "balance_loss_mlp": 1.01820946, "epoch": 0.9911621475380268, "flos": 22711384564320.0, "grad_norm": 1.7872577913163787, "language_loss": 0.71601307, "learning_rate": 8.083506147522623e-10, "loss": 0.73793304, "num_input_tokens_seen": 177831545, "step": 8243, "time_per_iteration": 2.461918354034424 }, { "auxiliary_loss_clip": 0.01146883, "auxiliary_loss_mlp": 0.01026255, "balance_loss_clip": 1.04426396, "balance_loss_mlp": 1.01934135, "epoch": 0.991282390428666, "flos": 13516959650400.0, "grad_norm": 2.0935396792139653, "language_loss": 0.85274321, "learning_rate": 7.863571529906909e-10, "loss": 0.87447459, "num_input_tokens_seen": 177847130, "step": 8244, "time_per_iteration": 2.4168219566345215 }, { "auxiliary_loss_clip": 0.01051106, "auxiliary_loss_mlp": 0.01000776, "balance_loss_clip": 1.00546598, "balance_loss_mlp": 0.99947637, "epoch": 0.991402633319305, "flos": 61830503464800.0, "grad_norm": 0.7275445684660024, "language_loss": 0.5964998, "learning_rate": 7.646669703489372e-10, "loss": 0.61701858, "num_input_tokens_seen": 177911440, "step": 8245, "time_per_iteration": 3.9250314235687256 }, { "auxiliary_loss_clip": 0.01061469, "auxiliary_loss_mlp": 0.01030504, "balance_loss_clip": 1.0354352, "balance_loss_mlp": 1.02348864, "epoch": 0.9915228762099441, "flos": 18770152225920.0, "grad_norm": 2.2740089348252064, "language_loss": 0.57413828, "learning_rate": 7.432800701177023e-10, "loss": 0.59505802, "num_input_tokens_seen": 177929440, "step": 8246, "time_per_iteration": 2.848615884780884 }, { "auxiliary_loss_clip": 0.01041948, "auxiliary_loss_mlp": 0.01000408, "balance_loss_clip": 1.00596607, "balance_loss_mlp": 0.99912035, "epoch": 0.9916431191005832, "flos": 65936670458880.0, "grad_norm": 0.7967498504409751, "language_loss": 0.57852203, "learning_rate": 7.221964555415017e-10, "loss": 0.59894562, "num_input_tokens_seen": 177989100, "step": 8247, "time_per_iteration": 3.3447089195251465 }, { "auxiliary_loss_clip": 0.01136467, "auxiliary_loss_mlp": 0.01023152, "balance_loss_clip": 1.04361284, "balance_loss_mlp": 1.01700997, "epoch": 0.9917633619912223, "flos": 16581576397920.0, "grad_norm": 1.8109750631596906, "language_loss": 0.75000864, "learning_rate": 7.01416129818222e-10, "loss": 0.77160478, "num_input_tokens_seen": 178006720, "step": 8248, "time_per_iteration": 2.520055055618286 }, { "auxiliary_loss_clip": 0.01134183, "auxiliary_loss_mlp": 0.01021335, "balance_loss_clip": 1.04455519, "balance_loss_mlp": 1.01498461, "epoch": 0.9918836048818613, "flos": 25411112964480.0, "grad_norm": 3.3626516810587628, "language_loss": 0.58332616, "learning_rate": 6.809390961006745e-10, "loss": 0.60488135, "num_input_tokens_seen": 178026850, "step": 8249, "time_per_iteration": 3.317108392715454 }, { "auxiliary_loss_clip": 0.0113888, "auxiliary_loss_mlp": 0.01032511, "balance_loss_clip": 1.04501343, "balance_loss_mlp": 1.02568889, "epoch": 0.9920038477725005, "flos": 25046871122400.0, "grad_norm": 1.8759136448091278, "language_loss": 0.68411863, "learning_rate": 6.607653574948191e-10, "loss": 0.70583248, "num_input_tokens_seen": 178047630, "step": 8250, "time_per_iteration": 2.559769868850708 }, { "auxiliary_loss_clip": 0.01141648, "auxiliary_loss_mlp": 0.01023804, "balance_loss_clip": 1.04100919, "balance_loss_mlp": 1.01762891, "epoch": 0.9921240906631396, "flos": 21829776516000.0, "grad_norm": 1.7709731842192902, "language_loss": 0.81521559, "learning_rate": 6.408949170613187e-10, "loss": 0.83687007, "num_input_tokens_seen": 178066895, "step": 8251, "time_per_iteration": 2.5355279445648193 }, { "auxiliary_loss_clip": 0.01136078, "auxiliary_loss_mlp": 0.01030649, "balance_loss_clip": 1.04288626, "balance_loss_mlp": 1.02317452, "epoch": 0.9922443335537786, "flos": 24864229405440.0, "grad_norm": 16.466883346134733, "language_loss": 0.81760478, "learning_rate": 6.213277778144288e-10, "loss": 0.83927208, "num_input_tokens_seen": 178088540, "step": 8252, "time_per_iteration": 3.325249195098877 }, { "auxiliary_loss_clip": 0.01095685, "auxiliary_loss_mlp": 0.01024023, "balance_loss_clip": 1.03952479, "balance_loss_mlp": 1.01707363, "epoch": 0.9923645764444178, "flos": 21613091909280.0, "grad_norm": 2.5712728532782854, "language_loss": 0.66916841, "learning_rate": 6.020639427224416e-10, "loss": 0.69036549, "num_input_tokens_seen": 178106185, "step": 8253, "time_per_iteration": 2.5975377559661865 }, { "auxiliary_loss_clip": 0.01141, "auxiliary_loss_mlp": 0.01026632, "balance_loss_clip": 1.0461812, "balance_loss_mlp": 1.01994133, "epoch": 0.9924848193350568, "flos": 25001801725920.0, "grad_norm": 2.002227937191797, "language_loss": 0.72419322, "learning_rate": 5.831034147076864e-10, "loss": 0.74586952, "num_input_tokens_seen": 178123435, "step": 8254, "time_per_iteration": 2.5341615676879883 }, { "auxiliary_loss_clip": 0.01046979, "auxiliary_loss_mlp": 0.01001733, "balance_loss_clip": 1.00547266, "balance_loss_mlp": 1.00044584, "epoch": 0.9926050622256959, "flos": 68912555563200.0, "grad_norm": 0.6844404689827635, "language_loss": 0.55734372, "learning_rate": 5.644461966463065e-10, "loss": 0.57783091, "num_input_tokens_seen": 178191045, "step": 8255, "time_per_iteration": 3.1580703258514404 }, { "auxiliary_loss_clip": 0.01139576, "auxiliary_loss_mlp": 0.01023281, "balance_loss_clip": 1.0459559, "balance_loss_mlp": 1.01599181, "epoch": 0.9927253051163349, "flos": 20923679489280.0, "grad_norm": 3.2618719516293426, "language_loss": 0.75906122, "learning_rate": 5.460922913687049e-10, "loss": 0.78068978, "num_input_tokens_seen": 178210135, "step": 8256, "time_per_iteration": 2.5079634189605713 }, { "auxiliary_loss_clip": 0.01107583, "auxiliary_loss_mlp": 0.00762456, "balance_loss_clip": 1.0389607, "balance_loss_mlp": 1.00054407, "epoch": 0.9928455480069741, "flos": 22308215125920.0, "grad_norm": 1.8641787130833625, "language_loss": 0.75149977, "learning_rate": 5.280417016593208e-10, "loss": 0.77020013, "num_input_tokens_seen": 178229925, "step": 8257, "time_per_iteration": 2.6107654571533203 }, { "auxiliary_loss_clip": 0.01152823, "auxiliary_loss_mlp": 0.00761154, "balance_loss_clip": 1.0489676, "balance_loss_mlp": 1.00055408, "epoch": 0.9929657908976132, "flos": 17383892575200.0, "grad_norm": 1.6676336473322044, "language_loss": 0.75025696, "learning_rate": 5.102944302559642e-10, "loss": 0.76939678, "num_input_tokens_seen": 178247420, "step": 8258, "time_per_iteration": 2.463709592819214 }, { "auxiliary_loss_clip": 0.01107572, "auxiliary_loss_mlp": 0.01027199, "balance_loss_clip": 1.04401374, "balance_loss_mlp": 1.01993334, "epoch": 0.9930860337882522, "flos": 22674683819520.0, "grad_norm": 2.0082708218577916, "language_loss": 0.79454452, "learning_rate": 4.9285047985137e-10, "loss": 0.81589222, "num_input_tokens_seen": 178266840, "step": 8259, "time_per_iteration": 2.6414079666137695 }, { "auxiliary_loss_clip": 0.01160146, "auxiliary_loss_mlp": 0.01029268, "balance_loss_clip": 1.04878724, "balance_loss_mlp": 1.02237201, "epoch": 0.9932062766788914, "flos": 28147793528160.0, "grad_norm": 1.7708544161911501, "language_loss": 0.74494952, "learning_rate": 4.757098530916436e-10, "loss": 0.76684362, "num_input_tokens_seen": 178287285, "step": 8260, "time_per_iteration": 2.537822723388672 }, { "auxiliary_loss_clip": 0.01155352, "auxiliary_loss_mlp": 0.01023429, "balance_loss_clip": 1.04795766, "balance_loss_mlp": 1.01610637, "epoch": 0.9933265195695304, "flos": 20156663295360.0, "grad_norm": 2.9279327541873412, "language_loss": 0.78047597, "learning_rate": 4.5887255257670563e-10, "loss": 0.80226374, "num_input_tokens_seen": 178304325, "step": 8261, "time_per_iteration": 2.4826462268829346 }, { "auxiliary_loss_clip": 0.01168725, "auxiliary_loss_mlp": 0.01029431, "balance_loss_clip": 1.04776883, "balance_loss_mlp": 1.02229023, "epoch": 0.9934467624601695, "flos": 21362041160160.0, "grad_norm": 2.025693014991829, "language_loss": 0.76884413, "learning_rate": 4.4233858086117906e-10, "loss": 0.79082566, "num_input_tokens_seen": 178322850, "step": 8262, "time_per_iteration": 2.439681053161621 }, { "auxiliary_loss_clip": 0.01110051, "auxiliary_loss_mlp": 0.01025158, "balance_loss_clip": 1.04560542, "balance_loss_mlp": 1.01749611, "epoch": 0.9935670053508087, "flos": 19756043961120.0, "grad_norm": 2.1092569332346778, "language_loss": 0.67694741, "learning_rate": 4.261079404528356e-10, "loss": 0.69829947, "num_input_tokens_seen": 178342330, "step": 8263, "time_per_iteration": 3.3315186500549316 }, { "auxiliary_loss_clip": 0.01149267, "auxiliary_loss_mlp": 0.01031402, "balance_loss_clip": 1.04411197, "balance_loss_mlp": 1.02456808, "epoch": 0.9936872482414477, "flos": 21978842263680.0, "grad_norm": 2.064714555834907, "language_loss": 0.69127905, "learning_rate": 4.1018063381437205e-10, "loss": 0.71308577, "num_input_tokens_seen": 178362715, "step": 8264, "time_per_iteration": 2.508101224899292 }, { "auxiliary_loss_clip": 0.01048583, "auxiliary_loss_mlp": 0.00999087, "balance_loss_clip": 1.00944901, "balance_loss_mlp": 0.99800265, "epoch": 0.9938074911320868, "flos": 69810679024800.0, "grad_norm": 0.8672873984661329, "language_loss": 0.61182308, "learning_rate": 3.9455666336141167e-10, "loss": 0.63229978, "num_input_tokens_seen": 178426495, "step": 8265, "time_per_iteration": 3.1475627422332764 }, { "auxiliary_loss_clip": 0.01168817, "auxiliary_loss_mlp": 0.01026203, "balance_loss_clip": 1.04841137, "balance_loss_mlp": 1.01885986, "epoch": 0.9939277340227259, "flos": 15084172920960.0, "grad_norm": 2.8172571077368023, "language_loss": 0.83747721, "learning_rate": 3.7923603146450267e-10, "loss": 0.85942745, "num_input_tokens_seen": 178442555, "step": 8266, "time_per_iteration": 2.4328489303588867 }, { "auxiliary_loss_clip": 0.01128186, "auxiliary_loss_mlp": 0.01026792, "balance_loss_clip": 1.04268694, "balance_loss_mlp": 1.0201993, "epoch": 0.994047976913365, "flos": 17712367513440.0, "grad_norm": 1.8428659493480402, "language_loss": 0.8123014, "learning_rate": 3.642187404473418e-10, "loss": 0.83385122, "num_input_tokens_seen": 178460715, "step": 8267, "time_per_iteration": 2.536980152130127 }, { "auxiliary_loss_clip": 0.0115343, "auxiliary_loss_mlp": 0.01024178, "balance_loss_clip": 1.04480553, "balance_loss_mlp": 1.01754141, "epoch": 0.994168219804004, "flos": 19171561733280.0, "grad_norm": 2.4655091201877144, "language_loss": 0.8607496, "learning_rate": 3.495047925885508e-10, "loss": 0.88252568, "num_input_tokens_seen": 178479050, "step": 8268, "time_per_iteration": 2.471449851989746 }, { "auxiliary_loss_clip": 0.01136339, "auxiliary_loss_mlp": 0.01024478, "balance_loss_clip": 1.0430963, "balance_loss_mlp": 1.01712012, "epoch": 0.9942884626946432, "flos": 17851592014080.0, "grad_norm": 2.2852221169186278, "language_loss": 0.82637274, "learning_rate": 3.350941901199e-10, "loss": 0.84798086, "num_input_tokens_seen": 178495970, "step": 8269, "time_per_iteration": 2.48555326461792 }, { "auxiliary_loss_clip": 0.01143551, "auxiliary_loss_mlp": 0.01022448, "balance_loss_clip": 1.04470944, "balance_loss_mlp": 1.0157603, "epoch": 0.9944087055852823, "flos": 18796580720160.0, "grad_norm": 2.3270731290166227, "language_loss": 0.83562565, "learning_rate": 3.2098693522764066e-10, "loss": 0.85728562, "num_input_tokens_seen": 178509170, "step": 8270, "time_per_iteration": 3.246222734451294 }, { "auxiliary_loss_clip": 0.01143824, "auxiliary_loss_mlp": 0.00762171, "balance_loss_clip": 1.04388881, "balance_loss_mlp": 1.00052691, "epoch": 0.9945289484759213, "flos": 20996973228000.0, "grad_norm": 1.8930431457693566, "language_loss": 0.8130368, "learning_rate": 3.071830300516165e-10, "loss": 0.83209682, "num_input_tokens_seen": 178527000, "step": 8271, "time_per_iteration": 2.5290515422821045 }, { "auxiliary_loss_clip": 0.01159942, "auxiliary_loss_mlp": 0.01030201, "balance_loss_clip": 1.04610419, "balance_loss_mlp": 1.02270234, "epoch": 0.9946491913665605, "flos": 14756955076320.0, "grad_norm": 2.100312902275003, "language_loss": 0.70829517, "learning_rate": 2.9368247668615234e-10, "loss": 0.7301966, "num_input_tokens_seen": 178545590, "step": 8272, "time_per_iteration": 2.4766056537628174 }, { "auxiliary_loss_clip": 0.01172688, "auxiliary_loss_mlp": 0.01024758, "balance_loss_clip": 1.04870367, "balance_loss_mlp": 1.01766825, "epoch": 0.9947694342571995, "flos": 12669933246240.0, "grad_norm": 2.675673191601291, "language_loss": 0.61321867, "learning_rate": 2.804852771789434e-10, "loss": 0.63519311, "num_input_tokens_seen": 178558890, "step": 8273, "time_per_iteration": 2.404090404510498 }, { "auxiliary_loss_clip": 0.0116519, "auxiliary_loss_mlp": 0.01023072, "balance_loss_clip": 1.04696584, "balance_loss_mlp": 1.01693606, "epoch": 0.9948896771478386, "flos": 18843445964640.0, "grad_norm": 1.8587302946482187, "language_loss": 0.55896842, "learning_rate": 2.675914335321661e-10, "loss": 0.58085108, "num_input_tokens_seen": 178577645, "step": 8274, "time_per_iteration": 2.488347291946411 }, { "auxiliary_loss_clip": 0.01159055, "auxiliary_loss_mlp": 0.01029112, "balance_loss_clip": 1.04634368, "balance_loss_mlp": 1.02142334, "epoch": 0.9950099200384778, "flos": 24900211811040.0, "grad_norm": 2.9685688635929237, "language_loss": 0.79748118, "learning_rate": 2.550009477018111e-10, "loss": 0.81936282, "num_input_tokens_seen": 178596415, "step": 8275, "time_per_iteration": 3.24175763130188 }, { "auxiliary_loss_clip": 0.01139577, "auxiliary_loss_mlp": 0.00762123, "balance_loss_clip": 1.0464983, "balance_loss_mlp": 1.00052166, "epoch": 0.9951301629291168, "flos": 23733617874720.0, "grad_norm": 1.844438390126138, "language_loss": 0.63028592, "learning_rate": 2.4271382159790634e-10, "loss": 0.64930296, "num_input_tokens_seen": 178613845, "step": 8276, "time_per_iteration": 2.5235161781311035 }, { "auxiliary_loss_clip": 0.01108337, "auxiliary_loss_mlp": 0.01025378, "balance_loss_clip": 1.04415536, "balance_loss_mlp": 1.01864576, "epoch": 0.9952504058197559, "flos": 22236896820000.0, "grad_norm": 1.8182847596412108, "language_loss": 0.85923529, "learning_rate": 2.3073005708429406e-10, "loss": 0.88057244, "num_input_tokens_seen": 178633490, "step": 8277, "time_per_iteration": 3.391592502593994 }, { "auxiliary_loss_clip": 0.01124309, "auxiliary_loss_mlp": 0.01029254, "balance_loss_clip": 1.04502976, "balance_loss_mlp": 1.0228194, "epoch": 0.995370648710395, "flos": 21211035896640.0, "grad_norm": 1.6852362176508913, "language_loss": 0.71959591, "learning_rate": 2.190496559788535e-10, "loss": 0.74113154, "num_input_tokens_seen": 178651775, "step": 8278, "time_per_iteration": 2.5221452713012695 }, { "auxiliary_loss_clip": 0.01139919, "auxiliary_loss_mlp": 0.01020946, "balance_loss_clip": 1.04646611, "balance_loss_mlp": 1.01413, "epoch": 0.9954908916010341, "flos": 14866734058080.0, "grad_norm": 2.400997401765866, "language_loss": 0.76486516, "learning_rate": 2.0767262005372265e-10, "loss": 0.78647381, "num_input_tokens_seen": 178669290, "step": 8279, "time_per_iteration": 2.5051989555358887 }, { "auxiliary_loss_clip": 0.01128684, "auxiliary_loss_mlp": 0.01024314, "balance_loss_clip": 1.04221606, "balance_loss_mlp": 1.01724744, "epoch": 0.9956111344916732, "flos": 19208262478080.0, "grad_norm": 2.6186437334362336, "language_loss": 0.75081545, "learning_rate": 1.965989510346322e-10, "loss": 0.77234542, "num_input_tokens_seen": 178688410, "step": 8280, "time_per_iteration": 2.5131545066833496 }, { "auxiliary_loss_clip": 0.01104914, "auxiliary_loss_mlp": 0.01024906, "balance_loss_clip": 1.04029548, "balance_loss_mlp": 1.01786041, "epoch": 0.9957313773823123, "flos": 20047063898400.0, "grad_norm": 2.225015183623731, "language_loss": 0.71052659, "learning_rate": 1.8582865060134955e-10, "loss": 0.73182482, "num_input_tokens_seen": 178706600, "step": 8281, "time_per_iteration": 2.5501155853271484 }, { "auxiliary_loss_clip": 0.01059396, "auxiliary_loss_mlp": 0.01000433, "balance_loss_clip": 1.00529134, "balance_loss_mlp": 0.99912804, "epoch": 0.9958516202729514, "flos": 57483264248160.0, "grad_norm": 0.7791169752938295, "language_loss": 0.55735689, "learning_rate": 1.7536172038790098e-10, "loss": 0.57795519, "num_input_tokens_seen": 178766910, "step": 8282, "time_per_iteration": 3.1246843338012695 }, { "auxiliary_loss_clip": 0.01139019, "auxiliary_loss_mlp": 0.01028486, "balance_loss_clip": 1.04508734, "balance_loss_mlp": 1.0219053, "epoch": 0.9959718631635904, "flos": 27782905180800.0, "grad_norm": 2.185364534754423, "language_loss": 0.6926347, "learning_rate": 1.651981619819054e-10, "loss": 0.71430969, "num_input_tokens_seen": 178784060, "step": 8283, "time_per_iteration": 2.5711593627929688 }, { "auxiliary_loss_clip": 0.01114199, "auxiliary_loss_mlp": 0.01030836, "balance_loss_clip": 1.0424825, "balance_loss_mlp": 1.02413321, "epoch": 0.9960921060542296, "flos": 24024099057600.0, "grad_norm": 2.52428159945151, "language_loss": 0.70989883, "learning_rate": 1.5533797692546257e-10, "loss": 0.73134917, "num_input_tokens_seen": 178802795, "step": 8284, "time_per_iteration": 2.6123902797698975 }, { "auxiliary_loss_clip": 0.01148679, "auxiliary_loss_mlp": 0.01031534, "balance_loss_clip": 1.04393935, "balance_loss_mlp": 1.02437294, "epoch": 0.9962123489448687, "flos": 18697397241600.0, "grad_norm": 1.9682477690481437, "language_loss": 0.84038818, "learning_rate": 1.4578116671404296e-10, "loss": 0.86219037, "num_input_tokens_seen": 178821075, "step": 8285, "time_per_iteration": 2.4787373542785645 }, { "auxiliary_loss_clip": 0.01150251, "auxiliary_loss_mlp": 0.01025045, "balance_loss_clip": 1.04795909, "balance_loss_mlp": 1.01852417, "epoch": 0.9963325918355077, "flos": 20010758240160.0, "grad_norm": 2.9718079488129456, "language_loss": 0.71270454, "learning_rate": 1.3652773279759777e-10, "loss": 0.73445749, "num_input_tokens_seen": 178837725, "step": 8286, "time_per_iteration": 2.4636850357055664 }, { "auxiliary_loss_clip": 0.01152053, "auxiliary_loss_mlp": 0.0102637, "balance_loss_clip": 1.04445791, "balance_loss_mlp": 1.01911068, "epoch": 0.9964528347261468, "flos": 33108493571040.0, "grad_norm": 2.325775989599601, "language_loss": 0.63300085, "learning_rate": 1.2757767657989305e-10, "loss": 0.65478516, "num_input_tokens_seen": 178861515, "step": 8287, "time_per_iteration": 2.6070237159729004 }, { "auxiliary_loss_clip": 0.0115052, "auxiliary_loss_mlp": 0.01022452, "balance_loss_clip": 1.04568934, "balance_loss_mlp": 1.01589847, "epoch": 0.9965730776167859, "flos": 23109346043520.0, "grad_norm": 3.653142105798712, "language_loss": 0.86769402, "learning_rate": 1.1893099941850948e-10, "loss": 0.88942373, "num_input_tokens_seen": 178880410, "step": 8288, "time_per_iteration": 2.4808995723724365 }, { "auxiliary_loss_clip": 0.01142461, "auxiliary_loss_mlp": 0.01025291, "balance_loss_clip": 1.04256141, "balance_loss_mlp": 1.01785576, "epoch": 0.996693320507425, "flos": 22965847424640.0, "grad_norm": 2.06870541826252, "language_loss": 0.77167106, "learning_rate": 1.105877026252866e-10, "loss": 0.79334861, "num_input_tokens_seen": 178898740, "step": 8289, "time_per_iteration": 3.285635232925415 }, { "auxiliary_loss_clip": 0.0116772, "auxiliary_loss_mlp": 0.01026568, "balance_loss_clip": 1.04665899, "balance_loss_mlp": 1.01910555, "epoch": 0.996813563398064, "flos": 13222743103680.0, "grad_norm": 2.1322736206259982, "language_loss": 0.72203922, "learning_rate": 1.0254778746565663e-10, "loss": 0.74398214, "num_input_tokens_seen": 178914015, "step": 8290, "time_per_iteration": 2.4002206325531006 }, { "auxiliary_loss_clip": 0.01123812, "auxiliary_loss_mlp": 0.0102162, "balance_loss_clip": 1.04345489, "balance_loss_mlp": 1.01520967, "epoch": 0.9969338062887032, "flos": 14647858516800.0, "grad_norm": 2.8604769811176456, "language_loss": 0.73382175, "learning_rate": 9.481125515953259e-11, "loss": 0.75527608, "num_input_tokens_seen": 178932075, "step": 8291, "time_per_iteration": 2.5126118659973145 }, { "auxiliary_loss_clip": 0.01111872, "auxiliary_loss_mlp": 0.01023933, "balance_loss_clip": 1.03888679, "balance_loss_mlp": 1.01683998, "epoch": 0.9970540491793423, "flos": 25735744788000.0, "grad_norm": 2.1571420868141518, "language_loss": 0.79606164, "learning_rate": 8.737810688064228e-11, "loss": 0.81741965, "num_input_tokens_seen": 178951910, "step": 8292, "time_per_iteration": 2.600177049636841 }, { "auxiliary_loss_clip": 0.01121925, "auxiliary_loss_mlp": 0.01025515, "balance_loss_clip": 1.0441879, "balance_loss_mlp": 1.01777279, "epoch": 0.9971742920699813, "flos": 21470239795680.0, "grad_norm": 2.2493089571772944, "language_loss": 0.79018372, "learning_rate": 8.024834375608414e-11, "loss": 0.81165814, "num_input_tokens_seen": 178970500, "step": 8293, "time_per_iteration": 2.5337600708007812 }, { "auxiliary_loss_clip": 0.01059236, "auxiliary_loss_mlp": 0.01001054, "balance_loss_clip": 1.00521946, "balance_loss_mlp": 0.99977255, "epoch": 0.9972945349606205, "flos": 72211234337760.0, "grad_norm": 0.8201747904526188, "language_loss": 0.62847757, "learning_rate": 7.342196686788149e-11, "loss": 0.64908051, "num_input_tokens_seen": 179023665, "step": 8294, "time_per_iteration": 2.959709405899048 }, { "auxiliary_loss_clip": 0.01140636, "auxiliary_loss_mlp": 0.01026249, "balance_loss_clip": 1.0506382, "balance_loss_mlp": 1.01889634, "epoch": 0.9974147778512595, "flos": 19678296519360.0, "grad_norm": 1.9863805652390156, "language_loss": 0.68806487, "learning_rate": 6.689897725142834e-11, "loss": 0.70973372, "num_input_tokens_seen": 179043140, "step": 8295, "time_per_iteration": 2.49088454246521 }, { "auxiliary_loss_clip": 0.01140678, "auxiliary_loss_mlp": 0.01022675, "balance_loss_clip": 1.04347444, "balance_loss_mlp": 1.01598442, "epoch": 0.9975350207418986, "flos": 15960824428800.0, "grad_norm": 2.0478491762152276, "language_loss": 0.88692081, "learning_rate": 6.067937589615545e-11, "loss": 0.90855432, "num_input_tokens_seen": 179061215, "step": 8296, "time_per_iteration": 3.2624266147613525 }, { "auxiliary_loss_clip": 0.01041441, "auxiliary_loss_mlp": 0.01001567, "balance_loss_clip": 1.00543952, "balance_loss_mlp": 1.00030386, "epoch": 0.9976552636325378, "flos": 59961890255520.0, "grad_norm": 0.7495866288854534, "language_loss": 0.57821685, "learning_rate": 5.476316374575241e-11, "loss": 0.59864694, "num_input_tokens_seen": 179124700, "step": 8297, "time_per_iteration": 3.1068170070648193 }, { "auxiliary_loss_clip": 0.01172802, "auxiliary_loss_mlp": 0.0102792, "balance_loss_clip": 1.05022383, "balance_loss_mlp": 1.02060974, "epoch": 0.9977755065231768, "flos": 22487875735200.0, "grad_norm": 1.9998302091263787, "language_loss": 0.72363555, "learning_rate": 4.9150341697723476e-11, "loss": 0.74564272, "num_input_tokens_seen": 179144590, "step": 8298, "time_per_iteration": 2.47367787361145 }, { "auxiliary_loss_clip": 0.01138474, "auxiliary_loss_mlp": 0.01031398, "balance_loss_clip": 1.04685843, "balance_loss_mlp": 1.0247668, "epoch": 0.9978957494138159, "flos": 26030284587360.0, "grad_norm": 1.469300675506298, "language_loss": 0.6657306, "learning_rate": 4.384091060338768e-11, "loss": 0.68742931, "num_input_tokens_seen": 179165060, "step": 8299, "time_per_iteration": 2.552598237991333 }, { "auxiliary_loss_clip": 0.01152203, "auxiliary_loss_mlp": 0.01026749, "balance_loss_clip": 1.04679847, "balance_loss_mlp": 1.01992726, "epoch": 0.998015992304455, "flos": 22637839406880.0, "grad_norm": 2.1063958788642503, "language_loss": 0.73583126, "learning_rate": 3.883487126810081e-11, "loss": 0.75762081, "num_input_tokens_seen": 179184320, "step": 8300, "time_per_iteration": 2.4927279949188232 }, { "auxiliary_loss_clip": 0.01144301, "auxiliary_loss_mlp": 0.01023174, "balance_loss_clip": 1.04367733, "balance_loss_mlp": 1.01645052, "epoch": 0.9981362351950941, "flos": 18223448251680.0, "grad_norm": 1.7067517279613247, "language_loss": 0.79302204, "learning_rate": 3.41322244516995e-11, "loss": 0.81469679, "num_input_tokens_seen": 179202265, "step": 8301, "time_per_iteration": 3.208986282348633 }, { "auxiliary_loss_clip": 0.01100535, "auxiliary_loss_mlp": 0.01020289, "balance_loss_clip": 1.0419488, "balance_loss_mlp": 1.01363158, "epoch": 0.9982564780857331, "flos": 33474100257600.0, "grad_norm": 1.813370220481955, "language_loss": 0.62961125, "learning_rate": 2.9732970866946925e-11, "loss": 0.65081948, "num_input_tokens_seen": 179222145, "step": 8302, "time_per_iteration": 2.6766116619110107 }, { "auxiliary_loss_clip": 0.01111489, "auxiliary_loss_mlp": 0.01026738, "balance_loss_clip": 1.03804994, "balance_loss_mlp": 1.01885843, "epoch": 0.9983767209763723, "flos": 15523468432800.0, "grad_norm": 2.098890231752989, "language_loss": 0.77708864, "learning_rate": 2.563711118175327e-11, "loss": 0.79847091, "num_input_tokens_seen": 179239030, "step": 8303, "time_per_iteration": 3.3017594814300537 }, { "auxiliary_loss_clip": 0.01122117, "auxiliary_loss_mlp": 0.01030572, "balance_loss_clip": 1.04539287, "balance_loss_mlp": 1.02373815, "epoch": 0.9984969638670114, "flos": 19974380748000.0, "grad_norm": 3.8024231752943254, "language_loss": 0.83390558, "learning_rate": 2.184464601717728e-11, "loss": 0.85543251, "num_input_tokens_seen": 179257345, "step": 8304, "time_per_iteration": 2.534515380859375 }, { "auxiliary_loss_clip": 0.01158153, "auxiliary_loss_mlp": 0.01026662, "balance_loss_clip": 1.0494113, "balance_loss_mlp": 1.01968181, "epoch": 0.9986172067576504, "flos": 20375754338400.0, "grad_norm": 2.615696053291089, "language_loss": 0.77468836, "learning_rate": 1.8355575948758585e-11, "loss": 0.79653645, "num_input_tokens_seen": 179275330, "step": 8305, "time_per_iteration": 2.5064234733581543 }, { "auxiliary_loss_clip": 0.01138367, "auxiliary_loss_mlp": 0.01030719, "balance_loss_clip": 1.0425663, "balance_loss_mlp": 1.02382922, "epoch": 0.9987374496482896, "flos": 23727907078080.0, "grad_norm": 2.0561989778366865, "language_loss": 0.73266363, "learning_rate": 1.5169901505407424e-11, "loss": 0.75435454, "num_input_tokens_seen": 179292395, "step": 8306, "time_per_iteration": 2.5287954807281494 }, { "auxiliary_loss_clip": 0.01141941, "auxiliary_loss_mlp": 0.01024303, "balance_loss_clip": 1.04900384, "balance_loss_mlp": 1.01810694, "epoch": 0.9988576925389286, "flos": 25044033682560.0, "grad_norm": 1.6520088819012877, "language_loss": 0.73645604, "learning_rate": 1.228762317073695e-11, "loss": 0.75811839, "num_input_tokens_seen": 179311225, "step": 8307, "time_per_iteration": 2.5580503940582275 }, { "auxiliary_loss_clip": 0.01139734, "auxiliary_loss_mlp": 0.01026639, "balance_loss_clip": 1.04579389, "balance_loss_mlp": 1.01994586, "epoch": 0.9989779354295677, "flos": 31285632180480.0, "grad_norm": 1.8258318821047423, "language_loss": 0.78798681, "learning_rate": 9.70874138195299e-12, "loss": 0.80965054, "num_input_tokens_seen": 179333135, "step": 8308, "time_per_iteration": 2.619689702987671 }, { "auxiliary_loss_clip": 0.01168171, "auxiliary_loss_mlp": 0.01028229, "balance_loss_clip": 1.04673696, "balance_loss_mlp": 1.02118981, "epoch": 0.9990981783202069, "flos": 19573402244160.0, "grad_norm": 1.5693192286862772, "language_loss": 0.74651575, "learning_rate": 7.433256530076093e-12, "loss": 0.7684797, "num_input_tokens_seen": 179353090, "step": 8309, "time_per_iteration": 2.5684211254119873 }, { "auxiliary_loss_clip": 0.01116418, "auxiliary_loss_mlp": 0.01021979, "balance_loss_clip": 1.04006445, "balance_loss_mlp": 1.01577139, "epoch": 0.9992184212108459, "flos": 17199670512000.0, "grad_norm": 2.2936013748762645, "language_loss": 0.75445902, "learning_rate": 5.46116896038562e-12, "loss": 0.77584302, "num_input_tokens_seen": 179367500, "step": 8310, "time_per_iteration": 2.5219619274139404 }, { "auxiliary_loss_clip": 0.01138023, "auxiliary_loss_mlp": 0.01025688, "balance_loss_clip": 1.04581547, "balance_loss_mlp": 1.01891124, "epoch": 0.999338664101485, "flos": 46497859427520.0, "grad_norm": 1.9889826037744924, "language_loss": 0.62290227, "learning_rate": 3.792478972197699e-12, "loss": 0.64453936, "num_input_tokens_seen": 179388085, "step": 8311, "time_per_iteration": 2.742492437362671 }, { "auxiliary_loss_clip": 0.01166667, "auxiliary_loss_mlp": 0.01023813, "balance_loss_clip": 1.04656005, "balance_loss_mlp": 1.01721191, "epoch": 0.9994589069921241, "flos": 15158256832800.0, "grad_norm": 2.763449399346301, "language_loss": 0.69928932, "learning_rate": 2.4271868181990895e-12, "loss": 0.72119409, "num_input_tokens_seen": 179405250, "step": 8312, "time_per_iteration": 2.427100658416748 }, { "auxiliary_loss_clip": 0.0115502, "auxiliary_loss_mlp": 0.01026336, "balance_loss_clip": 1.04536271, "balance_loss_mlp": 1.01934409, "epoch": 0.9995791498827632, "flos": 12531463001760.0, "grad_norm": 2.0306482706549693, "language_loss": 0.81106454, "learning_rate": 1.3652927060014973e-12, "loss": 0.83287811, "num_input_tokens_seen": 179420845, "step": 8313, "time_per_iteration": 2.4409046173095703 }, { "auxiliary_loss_clip": 0.01130761, "auxiliary_loss_mlp": 0.01029337, "balance_loss_clip": 1.04553306, "balance_loss_mlp": 1.02219629, "epoch": 0.9996993927734023, "flos": 19245178724640.0, "grad_norm": 1.9953947068412459, "language_loss": 0.63625073, "learning_rate": 6.067967965872612e-13, "loss": 0.6578517, "num_input_tokens_seen": 179440455, "step": 8314, "time_per_iteration": 2.517552614212036 }, { "auxiliary_loss_clip": 0.01125683, "auxiliary_loss_mlp": 0.01025324, "balance_loss_clip": 1.04565096, "balance_loss_mlp": 1.01793289, "epoch": 0.9998196356640414, "flos": 62952811789440.0, "grad_norm": 1.5248943277333322, "language_loss": 0.76961887, "learning_rate": 1.5169920497548615e-13, "loss": 0.79112887, "num_input_tokens_seen": 179465075, "step": 8315, "time_per_iteration": 3.602729082107544 }, { "auxiliary_loss_clip": 0.01105112, "auxiliary_loss_mlp": 0.01015243, "balance_loss_clip": 1.02575731, "balance_loss_mlp": 1.01085186, "epoch": 0.9999398785546805, "flos": 50922385553760.0, "grad_norm": 1.2005556504037709, "language_loss": 0.55061448, "learning_rate": 0.0, "loss": 0.57181805, "num_input_tokens_seen": 179513955, "step": 8316, "time_per_iteration": 3.0654306411743164 }, { "epoch": 0.9999398785546805, "num_input_tokens_seen": 179513955, "step": 8316, "total_flos": 6.996750297464832e+17, "train_loss": 0.7890177294791118, "train_runtime": 23940.5794, "train_samples_per_second": 13.895, "train_steps_per_second": 0.347 } ], "logging_steps": 1.0, "max_steps": 8316, "num_input_tokens_seen": 179513955, "num_train_epochs": 1, "save_steps": 1664, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 6.996750297464832e+17, "train_batch_size": 5, "trial_name": null, "trial_params": null }