1{"step": 0, "loss": 0.4215627908706665, "lr": 1e-05, "grad_norm": 0.6118677854537964, "tokens": 157, "tok_per_s": 274.14230963615364, "t": 1.2}2{"step": 50, "loss": 0.38644373416900635, "lr": 0.00051, "grad_norm": 0.3403327763080597, "tokens": 8616, "tok_per_s": 327.50263718564526, "t": 26.94}3{"step": 100, "loss": 0.3993557095527649, "lr": 0.001, "grad_norm": 0.19693967700004578, "tokens": 16635, "tok_per_s": 317.1118277638921, "t": 53.09}4{"step": 150, "loss": 0.5319535136222839, "lr": 0.0009993400362330058, "grad_norm": 0.061642616987228394, "tokens": 25042, "tok_per_s": 316.8210057323757, "t": 79.67}5{"step": 200, "loss": 0.4607158601284027, "lr": 0.0009973620807194618, "grad_norm": 0.08856119215488434, "tokens": 33661, "tok_per_s": 320.36545008737136, "t": 105.71}6{"step": 250, "loss": 0.44208019971847534, "lr": 0.000994071935143687, "grad_norm": 0.13429038226604462, "tokens": 42410, "tok_per_s": 323.29051688479217, "t": 131.82}7{"step": 300, "loss": 0.40008068084716797, "lr": 0.000989479250069539, "grad_norm": 0.1254703253507614, "tokens": 50771, "tok_per_s": 322.9256924095149, "t": 157.85}8{"step": 350, "loss": 0.42055290937423706, "lr": 0.0009835974966336503, "grad_norm": 0.09505702555179596, "tokens": 59133, "tok_per_s": 322.3887426443442, "t": 184.05}9{"step": 400, "loss": 0.589392900466919, "lr": 0.0009764439270322612, "grad_norm": 0.0563763827085495, "tokens": 67299, "tok_per_s": 321.28861830513233, "t": 210.1}10{"step": 450, "loss": 0.4641851782798767, "lr": 0.0009680395239175562, "grad_norm": 0.014637819491326809, "tokens": 75575, "tok_per_s": 320.54594572175233, "t": 236.4}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.46040050192617205, "val_loss_b2": 0.45902826375947203, "val_loss_b4": 0.4576095216317739, "val_loss_b8": 0.456500826591691, "t": 269.32}12{"step": 500, "loss": 0.4064313769340515, "lr": 0.0009584089388519307, "grad_norm": 0.05891776457428932, "tokens": 84036, "tok_per_s": 312.1183919758909, "t": 269.87}13{"step": 550, "loss": 0.28864943981170654, "lr": 0.0009475804200007103, "grad_norm": 0.16637183725833893, "tokens": 92359, "tok_per_s": 312.30578299305637, "t": 296.37}14{"step": 600, "loss": 0.31248971819877625, "lr": 0.0009355857292754152, "grad_norm": 0.07882910221815109, "tokens": 100627, "tok_per_s": 312.7615907342303, "t": 322.37}15{"step": 650, "loss": 0.5155187845230103, "lr": 0.0009224600491706008, "grad_norm": 0.05953988432884216, "tokens": 108981, "tok_per_s": 313.1859851369123, "t": 348.61}16{"step": 700, "loss": 0.3829517066478729, "lr": 0.0009082418795675396, "grad_norm": 0.14407537877559662, "tokens": 117352, "tok_per_s": 313.6486047670118, "t": 374.79}17{"step": 750, "loss": 0.5017012357711792, "lr": 0.0008929729248074364, "grad_norm": 0.05127249285578728, "tokens": 125499, "tok_per_s": 313.8099252436701, "t": 400.55}18{"step": 800, "loss": 0.3804885149002075, "lr": 0.0008766979713654089, "grad_norm": 0.07205870747566223, "tokens": 133880, "tok_per_s": 314.27247882017707, "t": 426.63}19{"step": 850, "loss": 0.47106021642684937, "lr": 0.0008594647564840406, "grad_norm": 0.07446790486574173, "tokens": 142063, "tok_per_s": 314.534716276127, "t": 452.29}20{"step": 900, "loss": 0.34726813435554504, "lr": 0.0008413238281518224, "grad_norm": 0.03208323195576668, "tokens": 150436, "tok_per_s": 314.9248816256742, "t": 478.32}21{"step": 950, "loss": 0.4102481007575989, "lr": 0.0008223283968371944, "grad_norm": 0.07048038393259048, "tokens": 158527, "tok_per_s": 314.7631080601028, "t": 504.28}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45883833263353974, "val_loss_b2": 0.456586433549771, "val_loss_b4": 0.45600535870773445, "val_loss_b8": 0.4618213140300602, "t": 537.65}23{"step": 1000, "loss": 0.4026390612125397, "lr": 0.000802534179413072, "grad_norm": 0.16194237768650055, "tokens": 166917, "tok_per_s": 310.49529568524656, "t": 538.21}24{"step": 1050, "loss": 0.44254764914512634, "lr": 0.0007819992357296598, "grad_norm": 0.08429207652807236, "tokens": 175091, "tok_per_s": 311.1380772381598, "t": 563.38}25{"step": 1100, "loss": 0.3624999225139618, "lr": 0.0007607837983149056, "grad_norm": 0.10605435818433762, "tokens": 183159, "tok_per_s": 310.8249659762624, "t": 589.9}26{"step": 1150, "loss": 0.29011058807373047, "lr": 0.0007389500957021191, "grad_norm": 0.09928447008132935, "tokens": 191581, "tok_per_s": 311.22750091096566, "t": 616.21}27{"step": 1200, "loss": 0.5198277235031128, "lr": 0.0007165621699029615, "grad_norm": 0.11012042313814163, "tokens": 200185, "tok_per_s": 311.81995916986955, "t": 642.62}28{"step": 1250, "loss": 0.6001372933387756, "lr": 0.000693685688561191, "grad_norm": 0.18121665716171265, "tokens": 208401, "tok_per_s": 311.7817445619283, "t": 669.05}29{"step": 1300, "loss": 0.43161654472351074, "lr": 0.0006703877523381495, "grad_norm": 0.07582087069749832, "tokens": 216624, "tok_per_s": 311.74261393170747, "t": 695.51}30{"step": 1350, "loss": 0.5370020866394043, "lr": 0.0006467366980949609, "grad_norm": 0.18134166300296783, "tokens": 225093, "tok_per_s": 312.2574471943352, "t": 721.49}31{"step": 1400, "loss": 0.4010993540287018, "lr": 0.0006228018984487442, "grad_norm": 0.2446669042110443, "tokens": 233304, "tok_per_s": 312.4905588133485, "t": 747.23}32{"step": 1450, "loss": 0.4130832254886627, "lr": 0.0005986535582907738, "grad_norm": 0.18402060866355896, "tokens": 241497, "tok_per_s": 312.4623786188919, "t": 773.52}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4580521064182994, "val_loss_b2": 0.4554033951550179, "val_loss_b4": 0.4519530324997016, "val_loss_b8": 0.4495427200746043, "t": 805.33}34{"step": 1500, "loss": 0.45379316806793213, "lr": 0.000574362508863438, "grad_norm": 0.08189226686954498, "tokens": 249582, "tok_per_s": 309.95225064401683, "t": 805.86}35{"step": 1550, "loss": 0.32397177815437317, "lr": 0.00055, "grad_norm": 0.1170770674943924, "tokens": 257750, "tok_per_s": 310.05787878528304, "t": 831.93}36{"step": 1600, "loss": 0.44211632013320923, "lr": 0.0005256374911365621, "grad_norm": 0.27388671040534973, "tokens": 266068, "tok_per_s": 310.3107840778067, "t": 858.06}37{"step": 1650, "loss": 0.4663999378681183, "lr": 0.0005013464417092262, "grad_norm": 0.17251084744930267, "tokens": 274463, "tok_per_s": 310.3616142405872, "t": 884.97}38{"step": 1700, "loss": 0.30981576442718506, "lr": 0.00047719810155125587, "grad_norm": 0.11516217887401581, "tokens": 282527, "tok_per_s": 310.37046812596947, "t": 910.92}39{"step": 1750, "loss": 0.4652133882045746, "lr": 0.0004532633019050392, "grad_norm": 0.09670969843864441, "tokens": 290793, "tok_per_s": 310.42901698090066, "t": 937.39}40{"step": 1800, "loss": 0.4121433198451996, "lr": 0.0004296122476618507, "grad_norm": 0.10236097127199173, "tokens": 299230, "tok_per_s": 310.65488122621434, "t": 963.87}41{"step": 1850, "loss": 0.3820798695087433, "lr": 0.00040631431143880916, "grad_norm": 0.13996821641921997, "tokens": 307566, "tok_per_s": 310.8384685252244, "t": 990.1}42{"step": 1900, "loss": 0.39447590708732605, "lr": 0.00038343783009703845, "grad_norm": 0.15041464567184448, "tokens": 315780, "tok_per_s": 310.7747834618672, "t": 1016.74}43{"step": 1950, "loss": 0.4192514419555664, "lr": 0.000361049904297881, "grad_norm": 0.2027776539325714, "tokens": 324229, "tok_per_s": 310.9321268563887, "t": 1043.4}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.453916520541263, "val_loss_b2": 0.45076847499435674, "val_loss_b4": 0.4492485428044454, "val_loss_b8": 0.45280949031854273, "t": 1076.8}45{"step": 2000, "loss": 0.46930620074272156, "lr": 0.0003392162016850945, "grad_norm": 0.2262699007987976, "tokens": 332550, "tok_per_s": 308.86284622301207, "t": 1077.32}46{"step": 2050, "loss": 0.5228667259216309, "lr": 0.0003180007642703402, "grad_norm": 0.19293752312660217, "tokens": 340880, "tok_per_s": 309.06047567273947, "t": 1103.59}47{"step": 2100, "loss": 0.30423012375831604, "lr": 0.000297465820586928, "grad_norm": 0.30613431334495544, "tokens": 349421, "tok_per_s": 309.61979463337735, "t": 1129.18}48{"step": 2150, "loss": 0.4111465513706207, "lr": 0.0002776716031628058, "grad_norm": 0.14085763692855835, "tokens": 357735, "tok_per_s": 309.68674935333956, "t": 1155.79}49{"step": 2200, "loss": 0.4642895460128784, "lr": 0.00025867617184817756, "grad_norm": 0.15492048859596252, "tokens": 366168, "tok_per_s": 309.92509753795355, "t": 1182.11}50{"step": 2250, "loss": 0.3819395899772644, "lr": 0.00024053524351595949, "grad_norm": 0.14563636481761932, "tokens": 374489, "tok_per_s": 310.33092997489496, "t": 1207.37}51{"step": 2300, "loss": 0.31462031602859497, "lr": 0.0002233020286345913, "grad_norm": 0.11556868255138397, "tokens": 382832, "tok_per_s": 310.54728358504707, "t": 1233.4}52{"step": 2350, "loss": 0.5440794229507446, "lr": 0.00020702707519256364, "grad_norm": 0.12039979547262192, "tokens": 391097, "tok_per_s": 310.68165923886966, "t": 1259.47}53{"step": 2400, "loss": 0.28712886571884155, "lr": 0.00019175812043246033, "grad_norm": 0.2307298630475998, "tokens": 399648, "tok_per_s": 310.9897847571771, "t": 1285.72}54{"step": 2450, "loss": 0.3632950186729431, "lr": 0.00017753995082939931, "grad_norm": 0.14065538346767426, "tokens": 407914, "tok_per_s": 311.0332301329446, "t": 1312.11}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45291194379946587, "val_loss_b2": 0.45033118547389756, "val_loss_b4": 0.44955164108915596, "val_loss_b8": 0.45482027101775235, "t": 1344.73}56{"step": 2500, "loss": 0.2594369053840637, "lr": 0.00016441427072458494, "grad_norm": 0.17658741772174835, "tokens": 416524, "tok_per_s": 309.75830629796286, "t": 1345.31}57{"step": 2550, "loss": 0.3230692744255066, "lr": 0.0001524195799992897, "grad_norm": 0.13594764471054077, "tokens": 424563, "tok_per_s": 309.7782754249756, "t": 1371.18}58{"step": 2600, "loss": 0.281651109457016, "lr": 0.0001415910611480694, "grad_norm": 0.16234524548053741, "tokens": 432991, "tok_per_s": 310.0067706058924, "t": 1397.35}59{"step": 2650, "loss": 0.39591020345687866, "lr": 0.00013196047608244387, "grad_norm": 0.26276707649230957, "tokens": 441165, "tok_per_s": 310.1064690420483, "t": 1423.26}60{"step": 2700, "loss": 0.34772899746894836, "lr": 0.00012355607296773896, "grad_norm": 0.07940234243869781, "tokens": 449578, "tok_per_s": 310.3069061623993, "t": 1449.45}61{"step": 2750, "loss": 0.38338199257850647, "lr": 0.00011640250336634969, "grad_norm": 0.22599227726459503, "tokens": 457759, "tok_per_s": 310.4080809044847, "t": 1475.34}62{"step": 2800, "loss": 0.37128984928131104, "lr": 0.00011052074993046101, "grad_norm": 0.14386668801307678, "tokens": 465905, "tok_per_s": 310.4815140242521, "t": 1501.23}63{"step": 2850, "loss": 0.41163355112075806, "lr": 0.00010592806485631325, "grad_norm": 0.22036252915859222, "tokens": 474059, "tok_per_s": 310.42871731437873, "t": 1527.74}64{"step": 2900, "loss": 0.540281355381012, "lr": 0.00010263791928053818, "grad_norm": 0.12238268554210663, "tokens": 482466, "tok_per_s": 310.5899054303517, "t": 1554.02}65{"step": 2950, "loss": 0.3741905391216278, "lr": 0.0001006599637669943, "grad_norm": 0.1669616550207138, "tokens": 490971, "tok_per_s": 310.85948621345744, "t": 1580.04}66{"step": 2999, "loss": 0.37146925926208496, "lr": 0.0001000002640500325, "grad_norm": 0.2679428160190582, "tokens": 499017, "tok_per_s": 310.94627085620965, "t": 1605.47}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45138869419539723, "val_loss_b2": 0.4488783733429963, "val_loss_b4": 0.4492385162985225, "val_loss_b8": 0.45711572608552936, "t": 1612.75}68{"step": 2999, "checkpoint_sha256": "a9dbe32c9093e4c4b620426c9bbe8f9381c278a76a7c3a5083091e98627de091", "t": 1612.85}69