1{"step": 0, "loss": 4.941596031188965, "lr": 3.3333333333333333e-06, "grad_norm": 11.574689865112305, "tokens": 6745, "tok_per_s": 3448.674497642418, "t": 2.57}2{"step": 100, "loss": 1.024980902671814, "lr": 0.0003366666666666667, "grad_norm": 6.284777641296387, "tokens": 743327, "tok_per_s": 24635.56415165761, "t": 30.78}3{"step": 200, "loss": 0.3161088228225708, "lr": 0.00067, "grad_norm": 1.373868703842163, "tokens": 1487594, "tok_per_s": 30610.546350138597, "t": 49.21}4{"step": 300, "loss": 0.30001285672187805, "lr": 0.001, "grad_norm": 0.9516773223876953, "tokens": 2224234, "tok_per_s": 32789.997595509114, "t": 68.44}5{"step": 400, "loss": 0.3070284128189087, "lr": 0.0009998972383511385, "grad_norm": 0.535650372505188, "tokens": 2958062, "tok_per_s": 34020.53726643474, "t": 87.56}6{"step": 500, "loss": 0.26402828097343445, "lr": 0.0009995890003376942, "grad_norm": 0.28448614478111267, "tokens": 3686492, "tok_per_s": 34827.95914207498, "t": 106.46}7{"step": 600, "loss": 0.26615840196609497, "lr": 0.0009990754267376514, "grad_norm": 0.1614493727684021, "tokens": 4423182, "tok_per_s": 35174.945536219624, "t": 126.36}8{"step": 700, "loss": 0.26372361183166504, "lr": 0.0009983567521095432, "grad_norm": 0.1251145452260971, "tokens": 5161053, "tok_per_s": 35589.20662235398, "t": 145.63}9{"step": 800, "loss": 0.2783412039279938, "lr": 0.000997433304685324, "grad_norm": 0.15038050711154938, "tokens": 5899309, "tok_per_s": 35916.9021250508, "t": 164.86}10{"step": 900, "loss": 0.26982399821281433, "lr": 0.0009963055062204608, "grad_norm": 0.18118366599082947, "tokens": 6639495, "tok_per_s": 36137.222541925905, "t": 184.34}11{"step": 999, "val_loss_b0": 0.2594514171899984, "t": 203.77}12{"step": 1000, "loss": 0.23169617354869843, "lr": 0.000994973871801308, "grad_norm": 0.11366917192935944, "tokens": 7378163, "tok_per_s": 36281.5789031001, "t": 203.97}13{"step": 1100, "loss": 0.2553258240222931, "lr": 0.0009934390096098592, "grad_norm": 0.3918009102344513, "tokens": 8114548, "tok_per_s": 36353.84211880216, "t": 223.82}14{"step": 1200, "loss": 0.2367946058511734, "lr": 0.0009917016206459795, "grad_norm": 0.13034524023532867, "tokens": 8855020, "tok_per_s": 36497.31428960828, "t": 243.23}15{"step": 1300, "loss": 0.23620161414146423, "lr": 0.0009897624984072447, "grad_norm": 0.10524164885282516, "tokens": 9601501, "tok_per_s": 36627.59412114859, "t": 262.75}16{"step": 1400, "loss": 0.2609722912311554, "lr": 0.000987622528526538, "grad_norm": 0.1504184454679489, "tokens": 10343983, "tok_per_s": 36734.75257922134, "t": 282.2}17{"step": 1500, "loss": 0.24786852300167084, "lr": 0.0009852826883675633, "grad_norm": 0.09120374917984009, "tokens": 11090019, "tok_per_s": 36781.989637705716, "t": 302.12}18{"step": 1600, "loss": 0.27017742395401, "lr": 0.0009827440465784664, "grad_norm": 0.1314472109079361, "tokens": 11842461, "tok_per_s": 36910.85801666111, "t": 321.45}19{"step": 1700, "loss": 0.24035435914993286, "lr": 0.0009800077626037633, "grad_norm": 0.11899533122777939, "tokens": 12581243, "tok_per_s": 36989.74402483245, "t": 340.74}20{"step": 1800, "loss": 0.23397061228752136, "lr": 0.000977075086154801, "grad_norm": 0.08897548168897629, "tokens": 13321567, "tok_per_s": 37016.40677754357, "t": 360.49}21{"step": 1900, "loss": 0.2829379141330719, "lr": 0.00097394735663899, "grad_norm": 0.10278685390949249, "tokens": 14065036, "tok_per_s": 37033.02886921329, "t": 380.41}22{"step": 1999, "val_loss_b0": 0.2532890506143083, "t": 399.96}23{"step": 2000, "loss": 0.25057756900787354, "lr": 0.000970626002548074, "grad_norm": 0.12103113532066345, "tokens": 14797545, "tok_per_s": 37036.8869671652, "t": 400.15}24{"step": 2100, "loss": 0.23892416059970856, "lr": 0.0009671125408057098, "grad_norm": 0.08961229771375656, "tokens": 15540833, "tok_per_s": 36457.20327366902, "t": 426.89}25{"step": 2200, "loss": 0.24898186326026917, "lr": 0.0009634085760746613, "grad_norm": 0.14835017919540405, "tokens": 16275558, "tok_per_s": 35731.45537336063, "t": 456.11}26{"step": 2300, "loss": 0.23595838248729706, "lr": 0.00095951580002392, "grad_norm": 0.16141901910305023, "tokens": 17022061, "tok_per_s": 35139.94385487974, "t": 485.02}27{"step": 2400, "loss": 0.24106518924236298, "lr": 0.0009554359905560886, "grad_norm": 0.0851842537522316, "tokens": 17763312, "tok_per_s": 34671.051664215454, "t": 512.95}28{"step": 2500, "loss": 0.20480993390083313, "lr": 0.0009511710109953803, "grad_norm": 0.0695372149348259, "tokens": 18498245, "tok_per_s": 34175.30851428889, "t": 541.89}29{"step": 2600, "loss": 0.2770192623138428, "lr": 0.000946722809236604, "grad_norm": 0.1085411012172699, "tokens": 19238526, "tok_per_s": 33737.99756176212, "t": 570.84}30{"step": 2700, "loss": 0.22194169461727142, "lr": 0.0009420934168555253, "grad_norm": 0.07198011875152588, "tokens": 19973518, "tok_per_s": 33280.918869557994, "t": 600.76}31{"step": 2800, "loss": 0.24996443092823029, "lr": 0.0009372849481810073, "grad_norm": 0.11180724948644638, "tokens": 20706938, "tok_per_s": 32946.24092319232, "t": 629.12}32{"step": 2900, "loss": 0.25116056203842163, "lr": 0.0009322995993293593, "grad_norm": 0.08267257362604141, "tokens": 21455185, "tok_per_s": 32592.520323627323, "t": 658.9}33{"step": 2999, "val_loss_b0": 0.2490523274094323, "t": 688.03}34{"step": 3000, "loss": 0.2259361892938614, "lr": 0.0009271396472013282, "grad_norm": 0.08093008399009705, "tokens": 22195465, "tok_per_s": 32275.792960844774, "t": 688.29}35{"step": 3100, "loss": 0.23215486109256744, "lr": 0.0009218074484421977, "grad_norm": 0.09304545074701309, "tokens": 22927890, "tok_per_s": 32022.084999155453, "t": 716.61}36{"step": 3200, "loss": 0.26163896918296814, "lr": 0.0009163054383654632, "grad_norm": 0.10492544621229172, "tokens": 23665276, "tok_per_s": 31863.551082722563, "t": 743.32}37{"step": 3300, "loss": 0.25118446350097656, "lr": 0.0009106361298405805, "grad_norm": 0.09422911703586578, "tokens": 24408498, "tok_per_s": 32003.86730721813, "t": 763.28}38{"step": 3400, "loss": 0.239703968167305, "lr": 0.0009048021121452915, "grad_norm": 0.1178416982293129, "tokens": 25152353, "tok_per_s": 32149.98823824634, "t": 782.95}39{"step": 3500, "loss": 0.25807860493659973, "lr": 0.0008988060497830534, "grad_norm": 0.09707698225975037, "tokens": 25888139, "tok_per_s": 32197.485808180354, "t": 804.66}40{"step": 3600, "loss": 0.23271173238754272, "lr": 0.0008926506812661105, "grad_norm": 0.08569994568824768, "tokens": 26630664, "tok_per_s": 31958.49396409735, "t": 833.9}41{"step": 3700, "loss": 0.2531582713127136, "lr": 0.0008863388178647659, "grad_norm": 0.0903114601969719, "tokens": 27369665, "tok_per_s": 31740.88070223157, "t": 862.9}42{"step": 3800, "loss": 0.2413078397512436, "lr": 0.0008798733423234219, "grad_norm": 0.08409066498279572, "tokens": 28115168, "tok_per_s": 31492.4174954626, "t": 893.37}43{"step": 3900, "loss": 0.22927093505859375, "lr": 0.0008732572075439775, "grad_norm": 0.08642493188381195, "tokens": 28852349, "tok_per_s": 31122.18916815221, "t": 927.68}44{"step": 3999, "val_loss_b0": 0.24673577576941852, "t": 966.29}45{"step": 4000, "loss": 0.24888259172439575, "lr": 0.0008664934352371838, "grad_norm": 0.08086570352315903, "tokens": 29598638, "tok_per_s": 30637.600806213628, "t": 966.7}46{"step": 4100, "loss": 0.23915867507457733, "lr": 0.0008595851145425734, "grad_norm": 0.08897045254707336, "tokens": 30340887, "tok_per_s": 30066.126563382648, "t": 1009.75}47{"step": 4200, "loss": 0.2620323598384857, "lr": 0.0008525354006175926, "grad_norm": 0.09605882316827774, "tokens": 31075053, "tok_per_s": 29704.82556897383, "t": 1046.74}48{"step": 4300, "loss": 0.23395636677742004, "lr": 0.0008453475131965832, "grad_norm": 0.09766869246959686, "tokens": 31805156, "tok_per_s": 29540.262818070223, "t": 1077.28}49{"step": 4400, "loss": 0.21586599946022034, "lr": 0.0008380247351202708, "grad_norm": 0.08858782798051834, "tokens": 32546876, "tok_per_s": 29485.45496382066, "t": 1104.44}50{"step": 4500, "loss": 0.23905375599861145, "lr": 0.0008305704108364302, "grad_norm": 0.10618754476308823, "tokens": 33283782, "tok_per_s": 29487.24124656912, "t": 1129.36}51{"step": 4600, "loss": 0.24594713747501373, "lr": 0.0008229879448724149, "grad_norm": 0.09999126195907593, "tokens": 34028571, "tok_per_s": 29513.085509099164, "t": 1153.61}52{"step": 4700, "loss": 0.23214955627918243, "lr": 0.000815280800280247, "grad_norm": 0.08282716572284698, "tokens": 34766082, "tok_per_s": 29545.36412883303, "t": 1177.31}53{"step": 4800, "loss": 0.23158258199691772, "lr": 0.0008074524970549763, "grad_norm": 0.0798744335770607, "tokens": 35503604, "tok_per_s": 29570.258821686195, "t": 1201.27}54{"step": 4900, "loss": 0.22221843898296356, "lr": 0.0007995066105270354, "grad_norm": 0.08601359277963638, "tokens": 36246522, "tok_per_s": 29586.25869193613, "t": 1225.73}55{"step": 4999, "val_loss_b0": 0.2457874837291204, "t": 1251.74}56{"step": 4999, "checkpoint_sha256": "71927bfc6ee34ea898424f7c3c655f14e7c3257b12075f126003d51f5a1886f1", "t": 1252.0}57{"step": 5000, "loss": 0.22774429619312286, "lr": 0.0007914467697293183, "grad_norm": 0.0813710018992424, "tokens": 36994918, "tok_per_s": 29557.386772961494, "t": 1252.25}58{"step": 5100, "loss": 0.23766636848449707, "lr": 0.0007832766557397363, "grad_norm": 0.11291572451591492, "tokens": 37736785, "tok_per_s": 29557.424459246053, "t": 1277.34}59{"step": 5200, "loss": 0.24797874689102173, "lr": 0.0007750000000000001, "grad_norm": 0.09950819611549377, "tokens": 38482669, "tok_per_s": 29550.341683950977, "t": 1302.89}60{"step": 5300, "loss": 0.2394346445798874, "lr": 0.0007666205826114025, "grad_norm": 0.09214076399803162, "tokens": 39217877, "tok_per_s": 29552.652553165975, "t": 1327.68}61{"step": 5400, "loss": 0.22690193355083466, "lr": 0.000758142230608376, "grad_norm": 0.08511560410261154, "tokens": 39961932, "tok_per_s": 29588.87784254362, "t": 1351.19}62{"step": 5500, "loss": 0.22264301776885986, "lr": 0.000749568816210616, "grad_norm": 0.07027542591094971, "tokens": 40689343, "tok_per_s": 29511.284566847276, "t": 1379.39}63{"step": 5600, "loss": 0.2372397482395172, "lr": 0.0007409042550545667, "grad_norm": 0.0994800329208374, "tokens": 41419954, "tok_per_s": 29347.896095082524, "t": 1411.95}64{"step": 5700, "loss": 0.2575412094593048, "lr": 0.0007321525044050772, "grad_norm": 0.15150783956050873, "tokens": 42153008, "tok_per_s": 29307.818599319744, "t": 1438.9}65{"step": 5800, "loss": 0.2326875925064087, "lr": 0.0007233175613480459, "grad_norm": 0.09410254657268524, "tokens": 42891344, "tok_per_s": 29242.80011876678, "t": 1467.34}66{"step": 5900, "loss": 0.22659200429916382, "lr": 0.0007144034609648779, "grad_norm": 0.07550016790628433, "tokens": 43636103, "tok_per_s": 29178.531651642897, "t": 1496.1}67{"step": 5999, "val_loss_b0": 0.24712026692656713, "t": 1525.28}68{"step": 6000, "loss": 0.23550210893154144, "lr": 0.0007054142744895884, "grad_norm": 0.09134833514690399, "tokens": 44374883, "tok_per_s": 29099.076135675226, "t": 1525.57}69{"step": 6100, "loss": 0.24578362703323364, "lr": 0.0006963541074493962, "grad_norm": 0.09056496620178223, "tokens": 45111988, "tok_per_s": 29000.670551559306, "t": 1556.16}70{"step": 6200, "loss": 0.20576344430446625, "lr": 0.0006872270977896524, "grad_norm": 0.08477317541837692, "tokens": 45852797, "tok_per_s": 28935.58037167095, "t": 1585.26}71{"step": 6300, "loss": 0.22447314858436584, "lr": 0.0006780374139839646, "grad_norm": 0.09896229207515717, "tokens": 46588536, "tok_per_s": 28870.616629293163, "t": 1614.31}72{"step": 6400, "loss": 0.21796773374080658, "lr": 0.0006687892531303794, "grad_norm": 0.09591249376535416, "tokens": 47327551, "tok_per_s": 28820.224747576834, "t": 1642.78}73{"step": 6500, "loss": 0.22234489023685455, "lr": 0.000659486839034489, "grad_norm": 0.08444668352603912, "tokens": 48061930, "tok_per_s": 28780.156796154653, "t": 1670.58}74{"step": 6600, "loss": 0.22978655993938446, "lr": 0.0006501344202803415, "grad_norm": 0.09093228727579117, "tokens": 48799166, "tok_per_s": 28825.732926860233, "t": 1693.52}75{"step": 6700, "loss": 0.22826802730560303, "lr": 0.000640736268290035, "grad_norm": 0.0852239578962326, "tokens": 49542736, "tok_per_s": 28817.446488615966, "t": 1719.8}76{"step": 6800, "loss": 0.24896718561649323, "lr": 0.0006312966753728795, "grad_norm": 0.11489948630332947, "tokens": 50279703, "tok_per_s": 28817.993277229, "t": 1745.34}77{"step": 6900, "loss": 0.24561363458633423, "lr": 0.0006218199527650207, "grad_norm": 0.10542216151952744, "tokens": 51014086, "tok_per_s": 28883.15061642516, "t": 1766.83}78{"step": 6999, "val_loss_b0": 0.2523172920760197, "t": 1798.64}79{"step": 7000, "loss": 0.22365900874137878, "lr": 0.0006123104286604183, "grad_norm": 0.10404888540506363, "tokens": 51756321, "tok_per_s": 28779.93868386071, "t": 1798.96}80{"step": 7100, "loss": 0.21811366081237793, "lr": 0.0006027724462340809, "grad_norm": 0.09947623312473297, "tokens": 52501419, "tok_per_s": 28741.686862435763, "t": 1827.27}81{"step": 7200, "loss": 0.18835552036762238, "lr": 0.0005932103616584568, "grad_norm": 0.09692477434873581, "tokens": 53242348, "tok_per_s": 28672.112593973903, "t": 1857.55}82{"step": 7300, "loss": 0.222006157040596, "lr": 0.000583628542113891, "grad_norm": 0.08617004752159119, "tokens": 53980673, "tok_per_s": 28646.261278749655, "t": 1885.0}83{"step": 7400, "loss": 0.23116207122802734, "lr": 0.0005740313637940521, "grad_norm": 0.10457632690668106, "tokens": 54718992, "tok_per_s": 28581.463639404403, "t": 1915.1}84{"step": 7500, "loss": 0.21480606496334076, "lr": 0.0005644232099072449, "grad_norm": 0.09749004244804382, "tokens": 55464703, "tok_per_s": 28561.72064571933, "t": 1942.54}85{"step": 7600, "loss": 0.22075559198856354, "lr": 0.0005548084686745181, "grad_norm": 0.09191872179508209, "tokens": 56202787, "tok_per_s": 28584.31991083341, "t": 1966.83}86{"step": 7700, "loss": 0.233114093542099, "lr": 0.0005451915313254822, "grad_norm": 0.11089321970939636, "tokens": 56944732, "tok_per_s": 28587.51970203095, "t": 1992.55}87{"step": 7800, "loss": 0.20631960034370422, "lr": 0.0005355767900927552, "grad_norm": 0.10566965490579605, "tokens": 57687990, "tok_per_s": 28600.29852093074, "t": 2017.65}88{"step": 7900, "loss": 0.21017761528491974, "lr": 0.0005259686362059481, "grad_norm": 0.10289314389228821, "tokens": 58429571, "tok_per_s": 28637.497046198143, "t": 2040.93}89{"step": 7999, "val_loss_b0": 0.26403789397260957, "t": 2065.78}90{"step": 8000, "loss": 0.21417582035064697, "lr": 0.0005163714578861091, "grad_norm": 0.11148758232593536, "tokens": 59162425, "tok_per_s": 28641.94408279909, "t": 2066.2}91{"step": 8100, "loss": 0.1658760905265808, "lr": 0.0005067896383415433, "grad_norm": 0.08317413926124573, "tokens": 59903257, "tok_per_s": 28592.063785446953, "t": 2095.71}92{"step": 8200, "loss": 0.20030342042446136, "lr": 0.0004972275537659192, "grad_norm": 0.13063369691371918, "tokens": 60647639, "tok_per_s": 28560.088128883992, "t": 2124.12}93{"step": 8300, "loss": 0.20933914184570312, "lr": 0.0004876895713395816, "grad_norm": 0.11644300073385239, "tokens": 61390376, "tok_per_s": 28528.063007501823, "t": 2152.54}94{"step": 8400, "loss": 0.1932450383901596, "lr": 0.0004781800472349794, "grad_norm": 0.11099286377429962, "tokens": 62132384, "tok_per_s": 28485.408573198627, "t": 2181.81}95{"step": 8500, "loss": 0.20041947066783905, "lr": 0.0004687033246271206, "grad_norm": 0.11818403750658035, "tokens": 62871250, "tok_per_s": 28490.70425245547, "t": 2207.34}96{"step": 8600, "loss": 0.2051982879638672, "lr": 0.00045926373170996514, "grad_norm": 0.12751081585884094, "tokens": 63599816, "tok_per_s": 28484.227708673072, "t": 2233.42}97{"step": 8700, "loss": 0.2038002759218216, "lr": 0.0004498655797196586, "grad_norm": 0.11766581237316132, "tokens": 64339358, "tok_per_s": 28452.33818232927, "t": 2261.91}98{"step": 8800, "loss": 0.19860033690929413, "lr": 0.00044051316096551117, "grad_norm": 0.1309284120798111, "tokens": 65080898, "tok_per_s": 28413.644137607473, "t": 2291.09}99{"step": 8900, "loss": 0.20576757192611694, "lr": 0.0004312107468696206, "grad_norm": 0.12305575609207153, "tokens": 65823979, "tok_per_s": 28355.499522381953, "t": 2321.99}100{"step": 8999, "val_loss_b0": 0.28220341540577676, "t": 2351.33}101{"step": 9000, "loss": 0.19135911762714386, "lr": 0.00042196258601603546, "grad_norm": 0.14417314529418945, "tokens": 66572928, "tok_per_s": 28316.65131685448, "t": 2351.63}102{"step": 9100, "loss": 0.2039412409067154, "lr": 0.0004127729022103478, "grad_norm": 0.14502999186515808, "tokens": 67310597, "tok_per_s": 28278.401227138056, "t": 2380.89}103{"step": 9200, "loss": 0.19043566286563873, "lr": 0.0004036458925506039, "grad_norm": 0.1531902253627777, "tokens": 68048686, "tok_per_s": 28272.657269177755, "t": 2407.48}104{"step": 9300, "loss": 0.1848042607307434, "lr": 0.0003945857255104116, "grad_norm": 0.13151361048221588, "tokens": 68785548, "tok_per_s": 28289.174138224687, "t": 2432.13}105{"step": 9400, "loss": 0.16617880761623383, "lr": 0.00038559653903512224, "grad_norm": 0.13529454171657562, "tokens": 69518667, "tok_per_s": 28315.869980071766, "t": 2455.72}106{"step": 9500, "loss": 0.1821175515651703, "lr": 0.00037668243865195415, "grad_norm": 0.15534792840480804, "tokens": 70256594, "tok_per_s": 28380.86706972068, "t": 2476.1}107