SPB Git

spb/wp31_uqo

Public
1{"step": 0, "loss": 0.6086878180503845, "lr": 1e-05, "grad_norm": 5.591502666473389, "tokens": 166, "tok_per_s": 48.57572112768346, "t": 4.15}2{"step": 50, "loss": 0.3951524794101715, "lr": 0.00051, "grad_norm": 0.8533229231834412, "tokens": 8464, "tok_per_s": 28.70151326621961, "t": 295.63}3{"step": 100, "loss": 0.3223797082901001, "lr": 0.001, "grad_norm": 0.18323586881160736, "tokens": 16901, "tok_per_s": 19.66823249731207, "t": 860.04}4{"step": 150, "loss": 0.27734050154685974, "lr": 0.0009993400362330058, "grad_norm": 0.08692359924316406, "tokens": 25274, "tok_per_s": 22.80529961459579, "t": 1108.99}5{"step": 200, "loss": 0.38253864645957947, "lr": 0.0009973620807194618, "grad_norm": 0.13802668452262878, "tokens": 33315, "tok_per_s": 28.572322405111716, "t": 1166.72}6{"step": 250, "loss": 0.454597145318985, "lr": 0.000994071935143687, "grad_norm": 0.07655514031648636, "tokens": 41251, "tok_per_s": 33.79220567346582, "t": 1221.46}7{"step": 300, "loss": 0.33502283692359924, "lr": 0.000989479250069539, "grad_norm": 0.10540986806154251, "tokens": 49559, "tok_per_s": 38.774727800271805, "t": 1278.87}8{"step": 350, "loss": 0.3576190769672394, "lr": 0.0009835974966336503, "grad_norm": 0.002110691275447607, "tokens": 57986, "tok_per_s": 43.55632628153049, "t": 1332.02}9{"step": 400, "loss": 0.4474697709083557, "lr": 0.0009764439270322612, "grad_norm": 0.05764389410614967, "tokens": 66476, "tok_per_s": 47.922163173557614, "t": 1387.9}10{"step": 450, "loss": 0.31184959411621094, "lr": 0.0009680395239175562, "grad_norm": 0.04684630036354065, "tokens": 74868, "tok_per_s": 51.45560421860237, "t": 1455.73}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4601260251514779, "val_loss_b2": 0.4607305832247595, "val_loss_b4": 0.4647802246680337, "val_loss_b8": 0.47610143019449186, "t": 1542.75}12{"step": 500, "loss": 0.4474729299545288, "lr": 0.0009584089388519307, "grad_norm": 0.25284940004348755, "tokens": 83020, "tok_per_s": 53.808487212277754, "t": 1543.61}13{"step": 550, "loss": 0.5516677498817444, "lr": 0.0009475804200007103, "grad_norm": 0.041814908385276794, "tokens": 91309, "tok_per_s": 57.574993114407924, "t": 1586.65}14{"step": 600, "loss": 0.3748024106025696, "lr": 0.0009355857292754152, "grad_norm": 0.014619260095059872, "tokens": 99553, "tok_per_s": 61.12524371195154, "t": 1629.4}15{"step": 650, "loss": 0.38783589005470276, "lr": 0.0009224600491706008, "grad_norm": 0.05259908735752106, "tokens": 107933, "tok_per_s": 64.58739419954003, "t": 1671.85}16{"step": 700, "loss": 0.3171798586845398, "lr": 0.0009082418795675396, "grad_norm": 0.08226539939641953, "tokens": 115964, "tok_per_s": 66.88734885399552, "t": 1734.45}17{"step": 750, "loss": 0.4626535177230835, "lr": 0.0008929729248074364, "grad_norm": 0.09627313911914825, "tokens": 124286, "tok_per_s": 69.19220312774348, "t": 1796.97}18{"step": 800, "loss": 0.4489482045173645, "lr": 0.0008766979713654089, "grad_norm": 0.055716268718242645, "tokens": 132702, "tok_per_s": 71.2839083682566, "t": 1862.33}19{"step": 850, "loss": 0.4665398895740509, "lr": 0.0008594647564840406, "grad_norm": 0.07492727786302567, "tokens": 141243, "tok_per_s": 73.33847285552099, "t": 1926.64}20{"step": 900, "loss": 0.45589616894721985, "lr": 0.0008413238281518224, "grad_norm": 0.030308181419968605, "tokens": 149475, "tok_per_s": 75.12241511713212, "t": 1990.48}21{"step": 950, "loss": 0.3333028256893158, "lr": 0.0008223283968371944, "grad_norm": 0.09674570709466934, "tokens": 157833, "tok_per_s": 76.97035589123739, "t": 2051.3}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4601433407561188, "val_loss_b2": 0.460631461258649, "val_loss_b4": 0.4644767372052384, "val_loss_b8": 0.47558435258635046, "t": 2115.99}23{"step": 1000, "loss": 0.5702598690986633, "lr": 0.000802534179413072, "grad_norm": 0.1668848991394043, "tokens": 166058, "tok_per_s": 78.47306457177287, "t": 2116.85}24{"step": 1050, "loss": 0.3951827585697174, "lr": 0.0007819992357296598, "grad_norm": 0.07296370714902878, "tokens": 174370, "tok_per_s": 80.77866657490672, "t": 2159.35}25{"step": 1100, "loss": 0.36607158184051514, "lr": 0.0007607837983149056, "grad_norm": 0.06467955559492111, "tokens": 182838, "tok_per_s": 83.05931421512983, "t": 2202.03}26{"step": 1150, "loss": 0.44352781772613525, "lr": 0.0007389500957021191, "grad_norm": 0.17181701958179474, "tokens": 190769, "tok_per_s": 85.02311509347577, "t": 2244.46}27{"step": 1200, "loss": 0.523062527179718, "lr": 0.0007165621699029615, "grad_norm": 0.299515038728714, "tokens": 199061, "tok_per_s": 87.05808546535506, "t": 2287.26}28{"step": 1250, "loss": 0.34359949827194214, "lr": 0.000693685688561191, "grad_norm": 0.040510959923267365, "tokens": 207183, "tok_per_s": 88.95387962836296, "t": 2329.84}29{"step": 1300, "loss": 0.5045455694198608, "lr": 0.0006703877523381495, "grad_norm": 0.13754883408546448, "tokens": 215576, "tok_per_s": 90.87898706295462, "t": 2372.85}30{"step": 1350, "loss": 0.44206345081329346, "lr": 0.0006467366980949609, "grad_norm": 0.09391453117132187, "tokens": 223982, "tok_per_s": 92.73778075313835, "t": 2415.95}31{"step": 1400, "loss": 0.3570104241371155, "lr": 0.0006228018984487442, "grad_norm": 0.10468284785747528, "tokens": 232005, "tok_per_s": 94.39144181978952, "t": 2458.63}32{"step": 1450, "loss": 0.36034026741981506, "lr": 0.0005986535582907738, "grad_norm": 0.3067401349544525, "tokens": 239958, "tok_per_s": 95.97207268683574, "t": 2501.02}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45715150038792973, "val_loss_b2": 0.4551765588541346, "val_loss_b4": 0.4551622288282309, "val_loss_b8": 0.4620041440660111, "t": 2552.18}34{"step": 1500, "loss": 0.5470016002655029, "lr": 0.000574362508863438, "grad_norm": 0.17903417348861694, "tokens": 248410, "tok_per_s": 97.3269858910891, "t": 2553.06}35{"step": 1550, "loss": 0.357530415058136, "lr": 0.00055, "grad_norm": 0.16632308065891266, "tokens": 256596, "tok_per_s": 98.88811800371236, "t": 2595.54}36{"step": 1600, "loss": 0.370943546295166, "lr": 0.0005256374911365621, "grad_norm": 0.14112021028995514, "tokens": 264965, "tok_per_s": 100.4471352267856, "t": 2638.59}37{"step": 1650, "loss": 0.4515704810619354, "lr": 0.0005013464417092262, "grad_norm": 0.12218248099088669, "tokens": 273349, "tok_per_s": 101.97319282923124, "t": 2681.33}38{"step": 1700, "loss": 0.33938828110694885, "lr": 0.00047719810155125587, "grad_norm": 0.09921961277723312, "tokens": 281608, "tok_per_s": 103.41103743239542, "t": 2723.92}39{"step": 1750, "loss": 0.45946207642555237, "lr": 0.0004532633019050392, "grad_norm": 0.13133591413497925, "tokens": 290020, "tok_per_s": 104.8437001391309, "t": 2766.94}40{"step": 1800, "loss": 0.2865503132343292, "lr": 0.0004296122476618507, "grad_norm": 0.05468787997961044, "tokens": 298457, "tok_per_s": 106.26026598737931, "t": 2809.47}41{"step": 1850, "loss": 0.38586026430130005, "lr": 0.00040631431143880916, "grad_norm": 0.10746917873620987, "tokens": 306802, "tok_per_s": 107.58011769960137, "t": 2852.58}42{"step": 1900, "loss": 0.31629493832588196, "lr": 0.00038343783009703845, "grad_norm": 0.13617277145385742, "tokens": 314844, "tok_per_s": 108.82055204622884, "t": 2893.97}43{"step": 1950, "loss": 0.36099597811698914, "lr": 0.000361049904297881, "grad_norm": 0.21049484610557556, "tokens": 323167, "tok_per_s": 110.07547015446069, "t": 2936.6}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45385239264129595, "val_loss_b2": 0.45039074382152156, "val_loss_b4": 0.4491189436820882, "val_loss_b8": 0.45450616388147014, "t": 2988.15}45{"step": 2000, "loss": 0.45853590965270996, "lr": 0.0003392162016850945, "grad_norm": 0.237838014960289, "tokens": 331553, "tok_per_s": 110.94975386305275, "t": 2989.05}46{"step": 2050, "loss": 0.4334942698478699, "lr": 0.0003180007642703402, "grad_norm": 0.161817267537117, "tokens": 339799, "tok_per_s": 112.10293709476615, "t": 3031.87}47{"step": 2100, "loss": 0.32871299982070923, "lr": 0.000297465820586928, "grad_norm": 0.12303311377763748, "tokens": 348148, "tok_per_s": 113.2332227807702, "t": 3075.34}48{"step": 2150, "loss": 0.372611403465271, "lr": 0.0002776716031628058, "grad_norm": 0.10326254367828369, "tokens": 356667, "tok_per_s": 114.41623817059472, "t": 3118.01}49{"step": 2200, "loss": 0.28739768266677856, "lr": 0.00025867617184817756, "grad_norm": 0.20464755594730377, "tokens": 365197, "tok_per_s": 115.5833865585507, "t": 3160.33}50{"step": 2250, "loss": 0.4734596312046051, "lr": 0.00024053524351595949, "grad_norm": 0.1492805778980255, "tokens": 373572, "tok_per_s": 116.67968126619245, "t": 3202.42}51{"step": 2300, "loss": 0.3605608642101288, "lr": 0.0002233020286345913, "grad_norm": 0.3812474012374878, "tokens": 381835, "tok_per_s": 117.68088239208893, "t": 3245.4}52{"step": 2350, "loss": 0.48266199231147766, "lr": 0.00020702707519256364, "grad_norm": 0.16672322154045105, "tokens": 390126, "tok_per_s": 118.68672226495248, "t": 3287.75}53{"step": 2400, "loss": 0.3916466534137726, "lr": 0.00019175812043246033, "grad_norm": 0.2150316834449768, "tokens": 398650, "tok_per_s": 119.71130121489942, "t": 3330.83}54{"step": 2450, "loss": 0.3649258315563202, "lr": 0.00017753995082939931, "grad_norm": 0.1758032739162445, "tokens": 406902, "tok_per_s": 120.6344305554515, "t": 3373.75}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45223719078428704, "val_loss_b2": 0.44911933662034775, "val_loss_b4": 0.4486228882708298, "val_loss_b8": 0.45410807512263585, "t": 3424.64}56{"step": 2500, "loss": 0.46765267848968506, "lr": 0.00016441427072458494, "grad_norm": 0.2109534591436386, "tokens": 415163, "tok_per_s": 121.22340678282737, "t": 3425.51}57{"step": 2550, "loss": 0.3364342451095581, "lr": 0.0001524195799992897, "grad_norm": 0.1276320070028305, "tokens": 423133, "tok_per_s": 122.06597877244545, "t": 3467.16}58{"step": 2600, "loss": 0.30369776487350464, "lr": 0.0001415910611480694, "grad_norm": 0.19694751501083374, "tokens": 431396, "tok_per_s": 122.93219355046828, "t": 3509.95}59{"step": 2650, "loss": 0.41286900639533997, "lr": 0.00013196047608244387, "grad_norm": 0.2409028708934784, "tokens": 439671, "tok_per_s": 123.77860402743019, "t": 3552.81}60{"step": 2700, "loss": 0.3701034188270569, "lr": 0.00012355607296773896, "grad_norm": 0.13612399995326996, "tokens": 448101, "tok_per_s": 124.64105431596504, "t": 3595.86}61{"step": 2750, "loss": 0.5174947381019592, "lr": 0.00011640250336634969, "grad_norm": 0.20960532128810883, "tokens": 456477, "tok_per_s": 125.44568750107531, "t": 3639.57}62{"step": 2800, "loss": 0.39786362648010254, "lr": 0.00011052074993046101, "grad_norm": 0.16486535966396332, "tokens": 464767, "tok_per_s": 126.23243422301424, "t": 3682.57}63{"step": 2850, "loss": 0.34810179471969604, "lr": 0.00010592806485631325, "grad_norm": 0.17520645260810852, "tokens": 473172, "tok_per_s": 127.04781261807351, "t": 3725.09}64{"step": 2900, "loss": 0.3737996220588684, "lr": 0.00010263791928053818, "grad_norm": 0.13601557910442352, "tokens": 481142, "tok_per_s": 127.72694985790581, "t": 3767.69}65{"step": 2950, "loss": 0.40360385179519653, "lr": 0.0001006599637669943, "grad_norm": 0.15253709256649017, "tokens": 489885, "tok_per_s": 128.59360801297566, "t": 3810.29}66{"step": 2999, "loss": 0.3920281231403351, "lr": 0.0001000002640500325, "grad_norm": 0.19843091070652008, "tokens": 497821, "tok_per_s": 129.274931679266, "t": 3851.6}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4507546093500324, "val_loss_b2": 0.44700439720097407, "val_loss_b4": 0.4467927397854402, "val_loss_b8": 0.45473843927510327, "t": 3860.58}68{"step": 2999, "checkpoint_sha256": "8424e19c8ff4bb15b787bd768057d3ac75224146817e6e20636f1f54ae3ee22c", "t": 3860.64}69
Files · main
Menu

Theme