1{"step": 0, "loss": 0.7847523093223572, "lr": 1e-05, "grad_norm": 3.7255656719207764, "tokens": 139, "tok_per_s": 94.60415957077906, "t": 2.43}2{"step": 50, "loss": 0.3623970150947571, "lr": 0.00051, "grad_norm": 1.2555731534957886, "tokens": 8357, "tok_per_s": 827.8294011683108, "t": 11.06}3{"step": 100, "loss": 0.357541024684906, "lr": 0.001, "grad_norm": 0.8759458065032959, "tokens": 16659, "tok_per_s": 967.4040146743553, "t": 18.18}4{"step": 150, "loss": 0.49112144112586975, "lr": 0.0009993400362330058, "grad_norm": 0.8309666514396667, "tokens": 25169, "tok_per_s": 1027.7012596594475, "t": 25.45}5{"step": 200, "loss": 0.5060253143310547, "lr": 0.0009973620807194618, "grad_norm": 0.7428900003433228, "tokens": 33572, "tok_per_s": 1059.1043830994042, "t": 32.66}6{"step": 250, "loss": 0.43746238946914673, "lr": 0.000994071935143687, "grad_norm": 0.5766422152519226, "tokens": 41743, "tok_per_s": 1072.9922676860851, "t": 39.87}7{"step": 300, "loss": 0.4347538948059082, "lr": 0.000989479250069539, "grad_norm": 0.7003477811813354, "tokens": 49922, "tok_per_s": 1080.5467935110441, "t": 47.17}8{"step": 350, "loss": 0.3830159902572632, "lr": 0.0009835974966336503, "grad_norm": 0.39303648471832275, "tokens": 58442, "tok_per_s": 1092.1420338510165, "t": 54.47}9{"step": 400, "loss": 0.3463587760925293, "lr": 0.0009764439270322612, "grad_norm": 0.4299878776073456, "tokens": 66780, "tok_per_s": 1100.9052684702249, "t": 61.63}10{"step": 450, "loss": 0.5036681890487671, "lr": 0.0009680395239175562, "grad_norm": 0.7424328923225403, "tokens": 74983, "tok_per_s": 1102.7277288014332, "t": 68.96}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45268325091230865, "val_loss_b2": 0.453475615612799, "val_loss_b4": 0.46196556749339174, "val_loss_b8": 0.48360392164866756, "t": 79.79}12{"step": 500, "loss": 0.332805335521698, "lr": 0.0009584089388519307, "grad_norm": 0.37873801589012146, "tokens": 83177, "tok_per_s": 1052.9023276503387, "t": 79.96}13{"step": 550, "loss": 0.43671518564224243, "lr": 0.0009475804200007103, "grad_norm": 0.45310527086257935, "tokens": 91423, "tok_per_s": 1058.4272646170575, "t": 87.33}14{"step": 600, "loss": 0.3821231722831726, "lr": 0.0009355857292754152, "grad_norm": 0.3358429968357086, "tokens": 99986, "tok_per_s": 1069.9380189125668, "t": 94.43}15{"step": 650, "loss": 0.3512137532234192, "lr": 0.0009224600491706008, "grad_norm": 0.2908254861831665, "tokens": 108206, "tok_per_s": 1072.8091572733244, "t": 101.82}16{"step": 700, "loss": 0.40387898683547974, "lr": 0.0009082418795675396, "grad_norm": 0.2781226634979248, "tokens": 116415, "tok_per_s": 1077.0374251854719, "t": 109.05}17{"step": 750, "loss": 0.34831446409225464, "lr": 0.0008929729248074364, "grad_norm": 0.24993573129177094, "tokens": 124709, "tok_per_s": 1078.5536136561143, "t": 116.58}18{"step": 800, "loss": 0.4234451651573181, "lr": 0.0008766979713654089, "grad_norm": 0.254151850938797, "tokens": 132874, "tok_per_s": 1081.8017104535993, "t": 123.79}19{"step": 850, "loss": 0.38831058144569397, "lr": 0.0008594647564840406, "grad_norm": 0.378019779920578, "tokens": 141506, "tok_per_s": 1088.5168195190774, "t": 130.96}20{"step": 900, "loss": 0.2919617295265198, "lr": 0.0008413238281518224, "grad_norm": 0.31646159291267395, "tokens": 149769, "tok_per_s": 1091.1755346945276, "t": 138.21}21{"step": 950, "loss": 0.37705036997795105, "lr": 0.0008223283968371944, "grad_norm": 0.29431140422821045, "tokens": 158176, "tok_per_s": 1094.3783954110677, "t": 145.5}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4515376892390658, "val_loss_b2": 0.44921430077324837, "val_loss_b4": 0.451628811433434, "val_loss_b8": 0.4614453604798296, "t": 156.82}23{"step": 1000, "loss": 0.4173162579536438, "lr": 0.000802534179413072, "grad_norm": 0.23804211616516113, "tokens": 166573, "tok_per_s": 1067.559261326244, "t": 157.0}24{"step": 1050, "loss": 0.35681411623954773, "lr": 0.0007819992357296598, "grad_norm": 0.2763034403324127, "tokens": 174773, "tok_per_s": 1062.63365962079, "t": 165.44}25{"step": 1100, "loss": 0.43333232402801514, "lr": 0.0007607837983149056, "grad_norm": 0.3275335133075714, "tokens": 183200, "tok_per_s": 1059.8596464663945, "t": 173.82}26{"step": 1150, "loss": 0.46724170446395874, "lr": 0.0007389500957021191, "grad_norm": 0.29268404841423035, "tokens": 191474, "tok_per_s": 1056.7860479710525, "t": 182.15}27{"step": 1200, "loss": 0.40516453981399536, "lr": 0.0007165621699029615, "grad_norm": 0.28969886898994446, "tokens": 200061, "tok_per_s": 1054.813450991745, "t": 190.62}28{"step": 1250, "loss": 0.3800254166126251, "lr": 0.000693685688561191, "grad_norm": 0.2518923878669739, "tokens": 208455, "tok_per_s": 1052.6155266464318, "t": 199.0}29{"step": 1300, "loss": 0.3713598847389221, "lr": 0.0006703877523381495, "grad_norm": 0.33583664894104004, "tokens": 216814, "tok_per_s": 1049.5833413078772, "t": 207.53}30{"step": 1350, "loss": 0.33553555607795715, "lr": 0.0006467366980949609, "grad_norm": 0.32408949732780457, "tokens": 225136, "tok_per_s": 1047.366301598125, "t": 215.91}31{"step": 1400, "loss": 0.33104875683784485, "lr": 0.0006228018984487442, "grad_norm": 0.2826470136642456, "tokens": 233471, "tok_per_s": 1045.3141833031968, "t": 224.31}32{"step": 1450, "loss": 0.2977364957332611, "lr": 0.0005986535582907738, "grad_norm": 0.18422867357730865, "tokens": 241795, "tok_per_s": 1043.3564829251413, "t": 232.72}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4491505244496187, "val_loss_b2": 0.44731528186751096, "val_loss_b4": 0.4514878077011146, "val_loss_b8": 0.4606788751819323, "t": 245.19}34{"step": 1500, "loss": 0.3126184940338135, "lr": 0.000574362508863438, "grad_norm": 0.20622104406356812, "tokens": 250085, "tok_per_s": 1023.1702053254295, "t": 245.4}35{"step": 1550, "loss": 0.39052850008010864, "lr": 0.00055, "grad_norm": 0.34642305970191956, "tokens": 258187, "tok_per_s": 1021.3099732801359, "t": 253.76}36{"step": 1600, "loss": 0.3600459694862366, "lr": 0.0005256374911365621, "grad_norm": 0.32358941435813904, "tokens": 266247, "tok_per_s": 1019.1567798202367, "t": 262.2}37{"step": 1650, "loss": 0.29107916355133057, "lr": 0.0005013464417092262, "grad_norm": 0.27334755659103394, "tokens": 274738, "tok_per_s": 1018.5675953116505, "t": 270.69}38{"step": 1700, "loss": 0.35359323024749756, "lr": 0.00047719810155125587, "grad_norm": 0.23127613961696625, "tokens": 283093, "tok_per_s": 1017.7233440055288, "t": 279.13}39{"step": 1750, "loss": 0.38842839002609253, "lr": 0.0004532633019050392, "grad_norm": 0.27287590503692627, "tokens": 291261, "tok_per_s": 1016.741677264887, "t": 287.42}40{"step": 1800, "loss": 0.3819948136806488, "lr": 0.0004296122476618507, "grad_norm": 0.29248499870300293, "tokens": 299555, "tok_per_s": 1016.2673617244237, "t": 295.72}41{"step": 1850, "loss": 0.3956453204154968, "lr": 0.00040631431143880916, "grad_norm": 0.21573400497436523, "tokens": 307521, "tok_per_s": 1015.1582356296609, "t": 303.89}42{"step": 1900, "loss": 0.3833693563938141, "lr": 0.00038343783009703845, "grad_norm": 0.22151067852973938, "tokens": 315892, "tok_per_s": 1015.2224754088986, "t": 312.12}43{"step": 1950, "loss": 0.34972941875457764, "lr": 0.000361049904297881, "grad_norm": 0.25364282727241516, "tokens": 324298, "tok_per_s": 1014.3624129723005, "t": 320.68}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.44756288467339544, "val_loss_b2": 0.44440963410461637, "val_loss_b4": 0.4467914011632356, "val_loss_b8": 0.4546514372921977, "t": 332.99}45{"step": 2000, "loss": 0.28332147002220154, "lr": 0.0003392162016850945, "grad_norm": 0.25190848112106323, "tokens": 332486, "tok_per_s": 1000.9029458193031, "t": 333.15}46{"step": 2050, "loss": 0.42470934987068176, "lr": 0.0003180007642703402, "grad_norm": 0.31494641304016113, "tokens": 340738, "tok_per_s": 1000.3965234570783, "t": 341.57}47{"step": 2100, "loss": 0.3602575957775116, "lr": 0.000297465820586928, "grad_norm": 0.283297061920166, "tokens": 349148, "tok_per_s": 1000.387899829726, "t": 349.97}48{"step": 2150, "loss": 0.33586204051971436, "lr": 0.0002776716031628058, "grad_norm": 0.27943792939186096, "tokens": 357576, "tok_per_s": 1000.6299860145011, "t": 358.31}49{"step": 2200, "loss": 0.32031095027923584, "lr": 0.00025867617184817756, "grad_norm": 0.36055389046669006, "tokens": 366097, "tok_per_s": 1000.6449172699647, "t": 366.82}50{"step": 2250, "loss": 0.26610079407691956, "lr": 0.00024053524351595949, "grad_norm": 0.3297097086906433, "tokens": 374285, "tok_per_s": 999.4729368511379, "t": 375.44}51{"step": 2300, "loss": 0.36331915855407715, "lr": 0.0002233020286345913, "grad_norm": 0.25571250915527344, "tokens": 382864, "tok_per_s": 999.6323965532055, "t": 383.97}52{"step": 2350, "loss": 0.43665871024131775, "lr": 0.00020702707519256364, "grad_norm": 0.3200297951698303, "tokens": 391427, "tok_per_s": 999.8321701723585, "t": 392.45}53{"step": 2400, "loss": 0.6184206604957581, "lr": 0.00019175812043246033, "grad_norm": 0.5718246102333069, "tokens": 399797, "tok_per_s": 1000.3084865064794, "t": 400.63}54{"step": 2450, "loss": 0.3194226324558258, "lr": 0.00017753995082939931, "grad_norm": 0.24311095476150513, "tokens": 407948, "tok_per_s": 999.7659441892224, "t": 409.0}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.445787163777208, "val_loss_b2": 0.4425691216376217, "val_loss_b4": 0.44307358479958087, "val_loss_b8": 0.44531052872839677, "t": 421.0}56{"step": 2500, "loss": 0.38175392150878906, "lr": 0.00016441427072458494, "grad_norm": 0.29214105010032654, "tokens": 416019, "tok_per_s": 990.0054073622898, "t": 421.19}57{"step": 2550, "loss": 0.32035934925079346, "lr": 0.0001524195799992897, "grad_norm": 0.26605361700057983, "tokens": 424449, "tok_per_s": 990.2516017070942, "t": 429.59}58{"step": 2600, "loss": 0.34859734773635864, "lr": 0.0001415910611480694, "grad_norm": 0.3040164113044739, "tokens": 432494, "tok_per_s": 990.1325745221486, "t": 437.77}59{"step": 2650, "loss": 0.29335588216781616, "lr": 0.00013196047608244387, "grad_norm": 0.3412756323814392, "tokens": 440693, "tok_per_s": 990.468518276773, "t": 445.9}60{"step": 2700, "loss": 0.3877958059310913, "lr": 0.00012355607296773896, "grad_norm": 0.3710648715496063, "tokens": 449079, "tok_per_s": 990.8028281627495, "t": 454.21}61{"step": 2750, "loss": 0.30402904748916626, "lr": 0.00011640250336634969, "grad_norm": 0.222353994846344, "tokens": 457686, "tok_per_s": 991.727257585367, "t": 462.47}62{"step": 2800, "loss": 0.39523300528526306, "lr": 0.00011052074993046101, "grad_norm": 0.38902547955513, "tokens": 466220, "tok_per_s": 992.8543682317036, "t": 470.53}63{"step": 2850, "loss": 0.4534169137477875, "lr": 0.00010592806485631325, "grad_norm": 0.3856627643108368, "tokens": 474198, "tok_per_s": 992.7982086172335, "t": 478.61}64{"step": 2900, "loss": 0.42924803495407104, "lr": 0.00010263791928053818, "grad_norm": 0.5133951306343079, "tokens": 482297, "tok_per_s": 992.156865423293, "t": 487.07}65{"step": 2950, "loss": 0.3878234326839447, "lr": 0.0001006599637669943, "grad_norm": 0.4019675850868225, "tokens": 490516, "tok_per_s": 991.5821101080938, "t": 495.64}66{"step": 2999, "loss": 0.5060074925422668, "lr": 0.0001000002640500325, "grad_norm": 0.5497512817382812, "tokens": 498543, "tok_per_s": 991.4364895291196, "t": 503.81}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4453311391861466, "val_loss_b2": 0.44301822186691414, "val_loss_b4": 0.44345716586425543, "val_loss_b8": 0.44439047131413834, "t": 507.82}68{"step": 2999, "checkpoint_sha256": "13e35362d3e2e4c169299b94379e544e3b64dde1d8cb8d7b0c5c32346d279037", "t": 507.96}69