{"step": 0, "loss": 1.1303871870040894, "lr": 1e-05, "grad_norm": 19.156978607177734, "tokens": 146, "tok_per_s": 106.61191364337344, "t": 2.38} {"step": 50, "loss": 0.4088473618030548, "lr": 0.00051, "grad_norm": 0.8947320580482483, "tokens": 8382, "tok_per_s": 609.9761815651933, "t": 14.75} {"step": 100, "loss": 0.44658252596855164, "lr": 0.001, "grad_norm": 0.7412047982215881, "tokens": 16548, "tok_per_s": 659.6487019197025, "t": 26.1} {"step": 150, "loss": 0.3267364203929901, "lr": 0.0009993400362330058, "grad_norm": 0.6519464254379272, "tokens": 24812, "tok_per_s": 687.8310851018562, "t": 37.09} {"step": 200, "loss": 0.4395517408847809, "lr": 0.0009973620807194618, "grad_norm": 0.5386191010475159, "tokens": 33400, "tok_per_s": 701.8723757397129, "t": 48.6} {"step": 250, "loss": 0.4371294677257538, "lr": 0.000994071935143687, "grad_norm": 0.5980697274208069, "tokens": 41583, "tok_per_s": 708.5356265689844, "t": 59.7} {"step": 300, "loss": 0.36619535088539124, "lr": 0.000989479250069539, "grad_norm": 0.549362301826477, "tokens": 49989, "tok_per_s": 709.89605949112, "t": 71.44} {"step": 350, "loss": 0.49362534284591675, "lr": 0.0009835974966336503, "grad_norm": 0.494795024394989, "tokens": 58214, "tok_per_s": 715.4533593467881, "t": 82.37} {"step": 400, "loss": 0.42427685856819153, "lr": 0.0009764439270322612, "grad_norm": 0.5157603025436401, "tokens": 66391, "tok_per_s": 717.5133264238226, "t": 93.54} {"step": 450, "loss": 0.2982785701751709, "lr": 0.0009680395239175562, "grad_norm": 0.3406541049480438, "tokens": 74699, "tok_per_s": 722.8748721095218, "t": 104.35} {"step": 499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4591480939478754, "val_loss_b2": 0.4657251800792801, "val_loss_b4": 0.4811424473461228, "val_loss_b8": 0.5061533661415096, "t": 121.2} {"step": 500, "loss": 0.6357465386390686, "lr": 0.0009584089388519307, "grad_norm": 0.5368360877037048, "tokens": 83038, "tok_per_s": 689.891410646851, "t": 121.39} {"step": 550, "loss": 0.39021968841552734, "lr": 0.0009475804200007103, "grad_norm": 0.30846598744392395, "tokens": 91441, "tok_per_s": 695.7277104505869, "t": 132.44} {"step": 600, "loss": 0.45643171668052673, "lr": 0.0009355857292754152, "grad_norm": 0.36303508281707764, "tokens": 99553, "tok_per_s": 697.7345054725203, "t": 143.69} {"step": 650, "loss": 0.4497789144515991, "lr": 0.0009224600491706008, "grad_norm": 0.4570186138153076, "tokens": 107666, "tok_per_s": 701.6899997882736, "t": 154.46} {"step": 700, "loss": 0.4260607063770294, "lr": 0.0009082418795675396, "grad_norm": 0.34087568521499634, "tokens": 115959, "tok_per_s": 706.334913863484, "t": 165.18} {"step": 750, "loss": 0.3949582576751709, "lr": 0.0008929729248074364, "grad_norm": 0.3282860219478607, "tokens": 124302, "tok_per_s": 707.7143342632462, "t": 176.67} {"step": 800, "loss": 0.46723443269729614, "lr": 0.0008766979713654089, "grad_norm": 0.25655731558799744, "tokens": 132734, "tok_per_s": 713.1660448585383, "t": 187.14} {"step": 850, "loss": 0.5465179681777954, "lr": 0.0008594647564840406, "grad_norm": 0.5234431624412537, "tokens": 141025, "tok_per_s": 715.0936557127504, "t": 198.22} {"step": 900, "loss": 0.45426782965660095, "lr": 0.0008413238281518224, "grad_norm": 0.2999867796897888, "tokens": 149035, "tok_per_s": 713.7401934898809, "t": 209.82} {"step": 950, "loss": 0.4147106111049652, "lr": 0.0008223283968371944, "grad_norm": 0.3587491512298584, "tokens": 157135, "tok_per_s": 714.575918190923, "t": 220.91} {"step": 999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4530972489704341, "val_loss_b2": 0.4522318926980896, "val_loss_b4": 0.45584874272992776, "val_loss_b8": 0.46584678036993865, "t": 238.42} {"step": 1000, "loss": 0.42529383301734924, "lr": 0.000802534179413072, "grad_norm": 0.31314516067504883, "tokens": 165602, "tok_per_s": 696.5303040233642, "t": 238.76} {"step": 1050, "loss": 0.3572467267513275, "lr": 0.0007819992357296598, "grad_norm": 0.2744152545928955, "tokens": 173674, "tok_per_s": 697.6697700218165, "t": 249.94} {"step": 1100, "loss": 0.35868406295776367, "lr": 0.0007607837983149056, "grad_norm": 0.20643961429595947, "tokens": 181834, "tok_per_s": 698.5999522485814, "t": 261.29} {"step": 1150, "loss": 0.4067578911781311, "lr": 0.0007389500957021191, "grad_norm": 0.3078262507915497, "tokens": 190158, "tok_per_s": 700.1346502979239, "t": 272.61} {"step": 1200, "loss": 0.4134320318698883, "lr": 0.0007165621699029615, "grad_norm": 0.3199818730354309, "tokens": 198573, "tok_per_s": 700.8096050271967, "t": 284.37} {"step": 1250, "loss": 0.4760536551475525, "lr": 0.000693685688561191, "grad_norm": 0.31957995891571045, "tokens": 206880, "tok_per_s": 702.4096672429708, "t": 295.55} {"step": 1300, "loss": 0.3089119791984558, "lr": 0.0006703877523381495, "grad_norm": 0.296504408121109, "tokens": 215095, "tok_per_s": 703.7481343414984, "t": 306.67} {"step": 1350, "loss": 0.3951007127761841, "lr": 0.0006467366980949609, "grad_norm": 0.40391260385513306, "tokens": 223349, "tok_per_s": 703.1475254532435, "t": 318.67} {"step": 1400, "loss": 0.4536643922328949, "lr": 0.0006228018984487442, "grad_norm": 0.2914159893989563, "tokens": 231716, "tok_per_s": 704.0915507643581, "t": 330.13} {"step": 1450, "loss": 0.4818269610404968, "lr": 0.0005986535582907738, "grad_norm": 0.3112286329269409, "tokens": 239960, "tok_per_s": 704.8315316149459, "t": 341.47} {"step": 1499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4520345502677605, "val_loss_b2": 0.451071882553439, "val_loss_b4": 0.4560073337395124, "val_loss_b8": 0.4698763565807991, "t": 357.8} {"step": 1500, "loss": 0.43627554178237915, "lr": 0.000574362508863438, "grad_norm": 0.3390117287635803, "tokens": 248357, "tok_per_s": 695.5284299899816, "t": 358.08} {"step": 1550, "loss": 0.545423686504364, "lr": 0.00055, "grad_norm": 0.4425813555717468, "tokens": 256793, "tok_per_s": 696.9582972182975, "t": 369.46} {"step": 1600, "loss": 0.4664572477340698, "lr": 0.0005256374911365621, "grad_norm": 0.3341352641582489, "tokens": 265023, "tok_per_s": 697.322745179824, "t": 381.07} {"step": 1650, "loss": 0.5023301839828491, "lr": 0.0005013464417092262, "grad_norm": 0.4127004146575928, "tokens": 273283, "tok_per_s": 699.2701171476041, "t": 391.82} {"step": 1700, "loss": 0.4782492220401764, "lr": 0.00047719810155125587, "grad_norm": 0.38390716910362244, "tokens": 281671, "tok_per_s": 701.4225836117404, "t": 402.58} {"step": 1750, "loss": 0.40946075320243835, "lr": 0.0004532633019050392, "grad_norm": 0.24065548181533813, "tokens": 289799, "tok_per_s": 700.5158549404371, "t": 414.7} {"step": 1800, "loss": 0.34420567750930786, "lr": 0.0004296122476618507, "grad_norm": 0.2081642597913742, "tokens": 298140, "tok_per_s": 700.9544768784915, "t": 426.35} {"step": 1850, "loss": 0.4563181400299072, "lr": 0.00040631431143880916, "grad_norm": 0.36275792121887207, "tokens": 306518, "tok_per_s": 701.9582279006828, "t": 437.67} {"step": 1900, "loss": 0.34524214267730713, "lr": 0.00038343783009703845, "grad_norm": 0.25196799635887146, "tokens": 314885, "tok_per_s": 702.9496573391679, "t": 448.96} {"step": 1950, "loss": 0.30550169944763184, "lr": 0.000361049904297881, "grad_norm": 0.25513511896133423, "tokens": 322988, "tok_per_s": 703.177637472405, "t": 460.33} {"step": 1999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.44864598608416484, "val_loss_b2": 0.446466895263732, "val_loss_b4": 0.4488269209920508, "val_loss_b8": 0.4563600456497245, "t": 477.43} {"step": 2000, "loss": 0.4754907190799713, "lr": 0.0003392162016850945, "grad_norm": 0.3372824490070343, "tokens": 331190, "tok_per_s": 694.8807541008393, "t": 477.63} {"step": 2050, "loss": 0.395969033241272, "lr": 0.0003180007642703402, "grad_norm": 0.30620312690734863, "tokens": 339365, "tok_per_s": 696.5676415834191, "t": 488.22} {"step": 2100, "loss": 0.35516464710235596, "lr": 0.000297465820586928, "grad_norm": 0.3272034823894501, "tokens": 347821, "tok_per_s": 697.8313342060809, "t": 499.45} {"step": 2150, "loss": 0.47438111901283264, "lr": 0.0002776716031628058, "grad_norm": 0.23794403672218323, "tokens": 355802, "tok_per_s": 698.017247608294, "t": 510.74} {"step": 2200, "loss": 0.33491751551628113, "lr": 0.00025867617184817756, "grad_norm": 0.3654625415802002, "tokens": 363932, "tok_per_s": 697.9353554261128, "t": 522.45} {"step": 2250, "loss": 0.29757630825042725, "lr": 0.00024053524351595949, "grad_norm": 0.2594342529773712, "tokens": 372235, "tok_per_s": 699.7981670794388, "t": 532.93} {"step": 2300, "loss": 0.36788827180862427, "lr": 0.0002233020286345913, "grad_norm": 0.3337343633174896, "tokens": 380617, "tok_per_s": 700.7441259610939, "t": 544.18} {"step": 2350, "loss": 0.3440735638141632, "lr": 0.00020702707519256364, "grad_norm": 0.3572402596473694, "tokens": 388820, "tok_per_s": 701.340029955876, "t": 555.42} {"step": 2400, "loss": 0.3749927580356598, "lr": 0.00019175812043246033, "grad_norm": 0.6173905730247498, "tokens": 397013, "tok_per_s": 701.814816409547, "t": 566.71} {"step": 2450, "loss": 0.380673348903656, "lr": 0.00017753995082939931, "grad_norm": 0.3348492383956909, "tokens": 405444, "tok_per_s": 702.5378705590774, "t": 578.12} {"step": 2499, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.4467929447149631, "val_loss_b2": 0.44509214718039836, "val_loss_b4": 0.44757014087058455, "val_loss_b8": 0.45269790149310823, "t": 594.59} {"step": 2500, "loss": 0.45006272196769714, "lr": 0.00016441427072458494, "grad_norm": 0.4312812387943268, "tokens": 413920, "tok_per_s": 697.0332077911404, "t": 594.85} {"step": 2550, "loss": 0.31154370307922363, "lr": 0.0001524195799992897, "grad_norm": 0.2960372567176819, "tokens": 422230, "tok_per_s": 697.7020239502879, "t": 606.2} {"step": 2600, "loss": 0.42040112614631653, "lr": 0.0001415910611480694, "grad_norm": 0.36173930764198303, "tokens": 430404, "tok_per_s": 699.7523371810312, "t": 616.09} {"step": 2650, "loss": 0.38288208842277527, "lr": 0.00013196047608244387, "grad_norm": 0.45003724098205566, "tokens": 438498, "tok_per_s": 700.9007472494047, "t": 626.64} {"step": 2700, "loss": 0.34501010179519653, "lr": 0.00012355607296773896, "grad_norm": 0.3651038110256195, "tokens": 446658, "tok_per_s": 701.725429876494, "t": 637.54} {"step": 2750, "loss": 0.3390929400920868, "lr": 0.00011640250336634969, "grad_norm": 0.35142233967781067, "tokens": 455155, "tok_per_s": 702.6640536466413, "t": 648.78} {"step": 2800, "loss": 0.38923755288124084, "lr": 0.00011052074993046101, "grad_norm": 0.43240830302238464, "tokens": 463083, "tok_per_s": 702.9589156151578, "t": 659.78} {"step": 2850, "loss": 0.37596428394317627, "lr": 0.00010592806485631325, "grad_norm": 0.4015140235424042, "tokens": 471665, "tok_per_s": 702.985327203937, "t": 671.97} {"step": 2900, "loss": 0.4250481426715851, "lr": 0.00010263791928053818, "grad_norm": 0.5022682547569275, "tokens": 480193, "tok_per_s": 704.1496245054366, "t": 682.98} {"step": 2950, "loss": 0.3081389367580414, "lr": 0.0001006599637669943, "grad_norm": 0.35446104407310486, "tokens": 488284, "tok_per_s": 704.2471073201807, "t": 694.36} {"step": 2999, "loss": 0.42948317527770996, "lr": 0.0001000002640500325, "grad_norm": 0.4377197325229645, "tokens": 496298, "tok_per_s": 704.6837244208275, "t": 705.3} {"step": 2999, "val_loss_b0": 0.46212446495721704, "val_loss_b1": 0.44592514526908655, "val_loss_b2": 0.44428266334439687, "val_loss_b4": 0.4471073554853549, "val_loss_b8": 0.45107327945835196, "t": 711.25} {"step": 2999, "checkpoint_sha256": "597a91e3beb1daea126d7c5b2ecd2bac327eb323b55cebe57a153e1f27b95f80", "t": 711.47}