Step 1 | loss:0.12513844668865204 lr:5.263157894736842e-06 tokens_per_second_per_gpu:489.1639760078218 Step 2 | loss:0.27767521142959595 lr:1.0526315789473684e-05 tokens_per_second_per_gpu:211.5141760606634 Step 3 | loss:0.08814200013875961 lr:1.5789473684210526e-05 tokens_per_second_per_gpu:1034.8852521431254 Step 4 | loss:0.14448179304599762 lr:2.105263157894737e-05 tokens_per_second_per_gpu:603.851086921295 Step 5 | loss:0.17749035358428955 lr:2.6315789473684212e-05 tokens_per_second_per_gpu:346.6801253216191 Step 6 | loss:0.22013026475906372 lr:3.157894736842105e-05 tokens_per_second_per_gpu:330.39612038517987 Step 7 | loss:0.10268905758857727 lr:3.6842105263157895e-05 tokens_per_second_per_gpu:4722.104566287469 Step 8 | loss:0.018015574663877487 lr:4.210526315789474e-05 tokens_per_second_per_gpu:6431.808558531696 Step 9 | loss:0.08885238319635391 lr:4.736842105263158e-05 tokens_per_second_per_gpu:571.6879149150644 Step 10 | loss:0.05220437049865723 lr:5.2631578947368424e-05 tokens_per_second_per_gpu:4693.365676410151 Step 11 | loss:0.27746930718421936 lr:5.789473684210527e-05 tokens_per_second_per_gpu:1778.005353013356 Step 12 | loss:0.07933241873979568 lr:6.31578947368421e-05 tokens_per_second_per_gpu:4633.485321885326 Step 13 | loss:0.053572043776512146 lr:6.842105263157895e-05 tokens_per_second_per_gpu:4663.791103674315 Step 14 | loss:0.03598742559552193 lr:7.368421052631579e-05 tokens_per_second_per_gpu:6399.908531613933 Step 15 | loss:0.11993309110403061 lr:7.894736842105263e-05 tokens_per_second_per_gpu:2789.6131114882687 Step 16 | loss:0.02057805471122265 lr:8.421052631578948e-05 tokens_per_second_per_gpu:4668.259126390028 Step 17 | loss:0.08534551411867142 lr:8.947368421052632e-05 tokens_per_second_per_gpu:2704.2567977002714 Step 18 | loss:0.04972148686647415 lr:9.473684210526316e-05 tokens_per_second_per_gpu:2865.2209746652698 Step 19 | loss:0.10108952969312668 lr:0.0001 tokens_per_second_per_gpu:2700.8888204910354 Step 20 | loss:0.0645143985748291 lr:9.999156208166614e-05 tokens_per_second_per_gpu:4832.97408792502 Step 21 | loss:0.02110893465578556 lr:9.996625117460318e-05 tokens_per_second_per_gpu:6440.741150678779 Step 22 | loss:0.02561924234032631 lr:9.992407582166581e-05 tokens_per_second_per_gpu:4606.905708939646 Step 23 | loss:0.021514398977160454 lr:9.986505025774138e-05 tokens_per_second_per_gpu:4623.379873491071 Step 24 | loss:0.016372518613934517 lr:9.978919440494539e-05 tokens_per_second_per_gpu:6425.975738887444 Step 25 | loss:0.006173485890030861 lr:9.969653386589748e-05 tokens_per_second_per_gpu:6420.184616940538 Step 26 | loss:0.08447691053152084 lr:9.958709991508012e-05 tokens_per_second_per_gpu:2812.9036846256577 Step 27 | loss:0.04044446721673012 lr:9.946092948828289e-05 tokens_per_second_per_gpu:2045.035723823389 Step 28 | loss:0.00761376041918993 lr:9.931806517013612e-05 tokens_per_second_per_gpu:6419.125727543057 Step 29 | loss:2.9212271329015493e-05 lr:9.915855517973776e-05 tokens_per_second_per_gpu:2828.156433864423 Step 30 | loss:0.019834276288747787 lr:9.89824533543787e-05 tokens_per_second_per_gpu:4602.148315466991 Step 31 | loss:0.02634788677096367 lr:9.878981913137179e-05 tokens_per_second_per_gpu:2814.4680355231203 Step 32 | loss:0.01563478261232376 lr:9.85807175279907e-05 tokens_per_second_per_gpu:6430.664192724959 Step 33 | loss:0.00018869338964577764 lr:9.835521911952555e-05 tokens_per_second_per_gpu:4621.488151696804 Step 34 | loss:0.012434639967978 lr:9.811340001546251e-05 tokens_per_second_per_gpu:2720.019388899271 Step 35 | loss:0.009905942715704441 lr:9.785534183379572e-05 tokens_per_second_per_gpu:1779.500183931238 Step 36 | loss:0.0009810677729547024 lr:9.75811316734796e-05 tokens_per_second_per_gpu:6436.066182459395 Step 37 | loss:3.712317629833706e-05 lr:9.729086208503174e-05 tokens_per_second_per_gpu:6385.345573591526 Step 38 | loss:0.051087636500597 lr:9.698463103929542e-05 tokens_per_second_per_gpu:1740.9540895086413 Step 39 | loss:0.055866535753011703 lr:9.666254189437286e-05 tokens_per_second_per_gpu:6430.774494881437 Step 40 | loss:0.042876582592725754 lr:9.632470336074009e-05 tokens_per_second_per_gpu:2725.4300337513587 Step 41 | loss:4.068629277753644e-05 lr:9.59712294645554e-05 tokens_per_second_per_gpu:4793.682682767743 Step 42 | loss:0.005137309432029724 lr:9.560223950917353e-05 tokens_per_second_per_gpu:4886.87236586204 Step 43 | loss:9.51029869611375e-06 lr:9.521785803487889e-05 tokens_per_second_per_gpu:6433.24017207722 Step 44 | loss:0.0006581911002285779 lr:9.481821477685101e-05 tokens_per_second_per_gpu:6394.068307927573 Step 45 | loss:0.011923214420676231 lr:9.440344462137689e-05 tokens_per_second_per_gpu:4651.927148948125 Step 46 | loss:0.003881220705807209 lr:9.397368756032445e-05 tokens_per_second_per_gpu:4624.039551597291 Step 47 | loss:0.0006434835959225893 lr:9.352908864389312e-05 tokens_per_second_per_gpu:6421.54598110918 Step 48 | loss:0.013022343628108501 lr:9.306979793165681e-05 tokens_per_second_per_gpu:4593.720201739378 Step 49 | loss:0.016436750069260597 lr:9.259597044191636e-05 tokens_per_second_per_gpu:2712.1897246523313 Step 50 | loss:0.005530699156224728 lr:9.210776609937829e-05 tokens_per_second_per_gpu:4650.439436447809 Step 51 | loss:0.03162027895450592 lr:9.160534968117752e-05 tokens_per_second_per_gpu:4658.941153618514 Step 52 | loss:0.001070328988134861 lr:9.108889076126226e-05 tokens_per_second_per_gpu:6420.722511431488 Step 53 | loss:0.001636895933188498 lr:9.055856365316011e-05 tokens_per_second_per_gpu:4676.1058043584 Step 54 | loss:0.0023436627816408873 lr:9.001454735114421e-05 tokens_per_second_per_gpu:4685.836913259768 Step 55 | loss:0.002947223139926791 lr:8.945702546981969e-05 tokens_per_second_per_gpu:3607.0607627985087 Step 56 | loss:1.1357014045643155e-05 lr:8.88861861821507e-05 tokens_per_second_per_gpu:6388.666042322892 Step 57 | loss:0.0018116658320650458 lr:8.83022221559489e-05 tokens_per_second_per_gpu:2814.6286149842804 Step 58 | loss:0.0025165448896586895 lr:8.770533048884482e-05 tokens_per_second_per_gpu:2828.894264031168 Step 59 | loss:0.028220675885677338 lr:8.709571264176409e-05 tokens_per_second_per_gpu:2825.978751158324 Step 60 | loss:0.03158221021294594 lr:8.647357437093105e-05 tokens_per_second_per_gpu:340.0394546222093 Step 61 | loss:0.010333526879549026 lr:8.583912565842257e-05 tokens_per_second_per_gpu:4617.829336225042 Step 62 | loss:0.0030926747713238 lr:8.519258064129558e-05 tokens_per_second_per_gpu:4818.569941848527 Step 63 | loss:0.004414541646838188 lr:8.453415753931222e-05 tokens_per_second_per_gpu:4601.682609455322 Step 64 | loss:0.016952676698565483 lr:8.386407858128706e-05 tokens_per_second_per_gpu:2834.2569108388043 Step 65 | loss:0.004038005601614714 lr:8.318256993008107e-05 tokens_per_second_per_gpu:6423.033432306101 Step 66 | loss:0.00045517730177380145 lr:8.2489861606268e-05 tokens_per_second_per_gpu:6401.850928014607 Step 67 | loss:9.486277122050524e-06 lr:8.178618741049842e-05 tokens_per_second_per_gpu:6394.543131261295 Step 68 | loss:2.5972610728786094e-06 lr:8.107178484458824e-05 tokens_per_second_per_gpu:6360.161820455056 Step 69 | loss:0.0017736494773998857 lr:8.034689503135783e-05 tokens_per_second_per_gpu:4900.146728652612 Step 70 | loss:0.00011349606938892975 lr:7.961176263324901e-05 tokens_per_second_per_gpu:6405.960111262532 Step 71 | loss:2.416427832940826e-06 lr:7.886663576974733e-05 tokens_per_second_per_gpu:6396.551017054138 Step 72 | loss:4.147927029407583e-06 lr:7.811176593363772e-05 tokens_per_second_per_gpu:4918.184814476602 Step 73 | loss:3.7847003113711253e-06 lr:7.734740790612136e-05 tokens_per_second_per_gpu:2882.40260371104 Step 74 | loss:0.0003083910560235381 lr:7.65738196708228e-05 tokens_per_second_per_gpu:6420.82991308512 Step 75 | loss:0.0027848940808326006 lr:7.579126232671621e-05 tokens_per_second_per_gpu:4897.586010719973 Step 76 | loss:0.0022790227085351944 lr:7.500000000000001e-05 tokens_per_second_per_gpu:2721.072563920397 Step 77 | loss:1.4661931118098437e-06 lr:7.420029975494995e-05 tokens_per_second_per_gpu:6449.610643613004 Step 78 | loss:0.0009201367502100766 lr:7.33924315037804e-05 tokens_per_second_per_gpu:6393.140054586642 Step 79 | loss:0.00021389646281022578 lr:7.257666791554448e-05 tokens_per_second_per_gpu:2836.579024426254 Step 80 | loss:0.0001596698275534436 lr:7.175328432410366e-05 tokens_per_second_per_gpu:4649.107091415815 Step 81 | loss:0.003130219876766205 lr:7.092255863519806e-05 tokens_per_second_per_gpu:2827.0497468671183 Step 82 | loss:0.0005868118023499846 lr:7.008477123264848e-05 tokens_per_second_per_gpu:4712.871502714318 Step 83 | loss:0.00012436187535058707 lr:6.924020488372229e-05 tokens_per_second_per_gpu:6448.857423217293 Step 84 | loss:6.51985828881152e-05 lr:6.838914464369467e-05 tokens_per_second_per_gpu:3556.6975550492407 Step 85 | loss:0.00011639451258815825 lr:6.753187775963773e-05 tokens_per_second_per_gpu:4861.603860869885 Step 86 | loss:0.03337816894054413 lr:6.666869357346978e-05 tokens_per_second_per_gpu:2818.0086060190574 Step 87 | loss:0.0003359356487635523 lr:6.579988342429763e-05 tokens_per_second_per_gpu:4629.657454767926 Step 88 | loss:0.0014997980324551463 lr:6.492574055008473e-05 tokens_per_second_per_gpu:4881.239518855494 Step 89 | loss:7.738002750556916e-06 lr:6.404655998867848e-05 tokens_per_second_per_gpu:4669.80564291533 Step 90 | loss:8.703746425453573e-05 lr:6.316263847822997e-05 tokens_per_second_per_gpu:6417.726691214301 Step 91 | loss:0.0739043727517128 lr:6.227427435703997e-05 tokens_per_second_per_gpu:4612.9635786516965 Step 92 | loss:0.016813863068819046 lr:6.138176746286468e-05 tokens_per_second_per_gpu:2823.8781025034764 Step 93 | loss:6.928161747055128e-05 lr:6.0485419031715516e-05 tokens_per_second_per_gpu:2859.723339670579 Step 94 | loss:0.02898344211280346 lr:5.958553159618693e-05 tokens_per_second_per_gpu:6426.714621734662 Step 95 | loss:0.0013739264104515314 lr:5.868240888334653e-05 tokens_per_second_per_gpu:2844.9955209254304 Step 96 | loss:2.435055421301513e-06 lr:5.7776355712222166e-05 tokens_per_second_per_gpu:6418.51945491529 Step 97 | loss:1.5492834108954412e-06 lr:5.686767789092041e-05 tokens_per_second_per_gpu:6416.743087828441 Step 98 | loss:9.047378171089804e-07 lr:5.595668211341118e-05 tokens_per_second_per_gpu:6357.369361637581 Step 99 | loss:0.0036852010525763035 lr:5.504367585601342e-05 tokens_per_second_per_gpu:6390.133020958593 Step 100 | loss:0.0021667189430445433 lr:5.4128967273616625e-05 tokens_per_second_per_gpu:4597.133995440272 Step 101 | loss:0.005624657031148672 lr:5.3212865095673514e-05 tokens_per_second_per_gpu:4662.51872029268 Step 102 | loss:0.003217491786926985 lr:5.229567852199859e-05 tokens_per_second_per_gpu:4652.101588464225 Step 103 | loss:0.0016954037128016353 lr:5.1377717118408105e-05 tokens_per_second_per_gpu:4695.5548734896165 Step 104 | loss:0.0013828268274664879 lr:5.0459290712236326e-05 tokens_per_second_per_gpu:6429.358015641005 Step 105 | loss:9.068298822967336e-06 lr:4.9540709287763685e-05 tokens_per_second_per_gpu:6391.431356242531 Step 106 | loss:0.0033326654229313135 lr:4.8622282881591906e-05 tokens_per_second_per_gpu:4868.965084559455 Step 107 | loss:0.003369354410097003 lr:4.7704321478001415e-05 tokens_per_second_per_gpu:4667.411675619285 Step 108 | loss:0.0030244227964431047 lr:4.6787134904326504e-05 tokens_per_second_per_gpu:2852.53285533292 Step 109 | loss:0.001989662414416671 lr:4.5871032726383386e-05 tokens_per_second_per_gpu:4678.622547190236 Step 110 | loss:1.4466125321632717e-05 lr:4.4956324143986596e-05 tokens_per_second_per_gpu:4810.972524572794 Step 111 | loss:3.1691836284153396e-06 lr:4.404331788658882e-05 tokens_per_second_per_gpu:3587.3725472140322 Step 112 | loss:0.0011431199964135885 lr:4.3132322109079596e-05 tokens_per_second_per_gpu:6415.560684389564 Step 113 | loss:0.005362535361200571 lr:4.222364428777786e-05 tokens_per_second_per_gpu:2752.632997217397 Step 114 | loss:0.0075534312054514885 lr:4.131759111665349e-05 tokens_per_second_per_gpu:4368.799613301608 Step 115 | loss:0.0011615206021815538 lr:4.0414468403813095e-05 tokens_per_second_per_gpu:4674.262697274069 Step 116 | loss:0.0009375385707244277 lr:3.951458096828449e-05 tokens_per_second_per_gpu:6372.567470355085 Step 117 | loss:0.003302461002022028 lr:3.861823253713535e-05 tokens_per_second_per_gpu:6400.079595816121 Step 118 | loss:1.4625480844188132e-06 lr:3.772572564296005e-05 tokens_per_second_per_gpu:6417.671241359475 Step 119 | loss:0.001941256457939744 lr:3.683736152177005e-05 tokens_per_second_per_gpu:2841.5228444613617 Step 120 | loss:0.002696903655305505 lr:3.595344001132154e-05 tokens_per_second_per_gpu:6433.570437260473 Step 121 | loss:0.05962380766868591 lr:3.5074259449915284e-05 tokens_per_second_per_gpu:2843.32026878976 Step 122 | loss:1.4319573892862536e-05 lr:3.420011657570238e-05 tokens_per_second_per_gpu:2811.763598035594 Step 123 | loss:0.04734177514910698 lr:3.333130642653024e-05 tokens_per_second_per_gpu:1780.8220458373457 Step 124 | loss:0.002887164708226919 lr:3.2468122240362284e-05 tokens_per_second_per_gpu:4639.027286752893 Step 125 | loss:4.599939074978465e-06 lr:3.1610855356305354e-05 tokens_per_second_per_gpu:4698.550787904085 Step 126 | loss:0.004566989839076996 lr:3.0759795116277725e-05 tokens_per_second_per_gpu:4697.269678423468 Step 127 | loss:0.0036403313279151917 lr:2.991522876735154e-05 tokens_per_second_per_gpu:4632.184769962814 Step 128 | loss:0.003928751684725285 lr:2.9077441364801938e-05 tokens_per_second_per_gpu:4629.39022456499 Step 129 | loss:0.0016177849611267447 lr:2.824671567589635e-05 tokens_per_second_per_gpu:6414.063868110892 Step 130 | loss:0.0006506482604891062 lr:2.7423332084455544e-05 tokens_per_second_per_gpu:4862.7184002824115 Step 131 | loss:0.008410315029323101 lr:2.660756849621962e-05 tokens_per_second_per_gpu:2860.926802536824 Step 132 | loss:1.534285729576368e-05 lr:2.5799700245050074e-05 tokens_per_second_per_gpu:4644.185636025985 Step 133 | loss:0.012605580501258373 lr:2.500000000000001e-05 tokens_per_second_per_gpu:2828.5641390430605 Step 134 | loss:9.261790182790719e-06 lr:2.4208737673283815e-05 tokens_per_second_per_gpu:4883.91824384238 Step 135 | loss:0.004368562716990709 lr:2.3426180329177215e-05 tokens_per_second_per_gpu:4648.604585515982 Step 136 | loss:0.006988442502915859 lr:2.2652592093878666e-05 tokens_per_second_per_gpu:4880.214896300479 Step 137 | loss:0.004174124449491501 lr:2.1888234066362302e-05 tokens_per_second_per_gpu:4639.287041089598 Step 138 | loss:4.192010237602517e-06 lr:2.1133364230252688e-05 tokens_per_second_per_gpu:2827.9441333131067 Step 139 | loss:4.253053702996112e-06 lr:2.0388237366751006e-05 tokens_per_second_per_gpu:1344.97427429136 Step 140 | loss:0.0013539784122258425 lr:1.9653104968642173e-05 tokens_per_second_per_gpu:4658.711630506871 Step 141 | loss:0.0006630073767155409 lr:1.8928215155411773e-05 tokens_per_second_per_gpu:6409.573963090514 Step 142 | loss:0.002947476925328374 lr:1.821381258950161e-05 tokens_per_second_per_gpu:6428.605302530668 Step 143 | loss:0.0018973237602040172 lr:1.7510138393732028e-05 tokens_per_second_per_gpu:6394.570954520929 Step 144 | loss:3.7145280202821596e-06 lr:1.681743006991894e-05 tokens_per_second_per_gpu:4675.013611001729 Step 145 | loss:0.042256902903318405 lr:1.6135921418712956e-05 tokens_per_second_per_gpu:2843.6277884070064 Step 146 | loss:0.003359108464792371 lr:1.5465842460687784e-05 tokens_per_second_per_gpu:4600.518508424465 Step 147 | loss:0.005931657738983631 lr:1.4807419358704433e-05 tokens_per_second_per_gpu:2860.5951701361273 Step 148 | loss:4.868064024776686e-06 lr:1.4160874341577446e-05 tokens_per_second_per_gpu:6413.965658588304 Step 149 | loss:0.023996124044060707 lr:1.3526425629068967e-05 tokens_per_second_per_gpu:2840.5765071242326 Step 150 | loss:0.0011750105768442154 lr:1.2904287358235928e-05 tokens_per_second_per_gpu:6419.940118376599 Step 151 | loss:0.0019033460412174463 lr:1.2294669511155193e-05 tokens_per_second_per_gpu:4623.908438531533 Step 152 | loss:0.001131798722781241 lr:1.1697777844051105e-05 tokens_per_second_per_gpu:4702.877983028494 Step 153 | loss:3.302582263131626e-05 lr:1.111381381784931e-05 tokens_per_second_per_gpu:6421.469748265485 Step 154 | loss:0.005529465619474649 lr:1.0542974530180327e-05 tokens_per_second_per_gpu:2728.8512653353687 Step 155 | loss:0.00396890239790082 lr:9.985452648855803e-06 tokens_per_second_per_gpu:2799.251290868056 Step 156 | loss:0.005971190519630909 lr:9.441436346839894e-06 tokens_per_second_per_gpu:2706.4303003540135 Step 157 | loss:8.517595233570319e-07 lr:8.911109238737747e-06 tokens_per_second_per_gpu:6381.4581134098125 Step 158 | loss:0.002489866456016898 lr:8.3946503188225e-06 tokens_per_second_per_gpu:4650.469394964638 Step 159 | loss:0.0034828418865799904 lr:7.89223390062172e-06 tokens_per_second_per_gpu:4662.585502769738 Step 160 | loss:0.002724783029407263 lr:7.404029558083653e-06 tokens_per_second_per_gpu:6430.701081956164 Step 161 | loss:0.0018676317995414138 lr:6.9302020683432055e-06 tokens_per_second_per_gpu:4662.483149651056 Step 162 | loss:9.775993930816185e-06 lr:6.470911356106885e-06 tokens_per_second_per_gpu:6427.367120140001 Step 163 | loss:0.01138626504689455 lr:6.026312439675552e-06 tokens_per_second_per_gpu:4650.724654138391 Step 164 | loss:0.0014588163467124104 lr:5.596555378623125e-06 tokens_per_second_per_gpu:2851.8214822531313 Step 165 | loss:0.00135196337942034 lr:5.181785223148999e-06 tokens_per_second_per_gpu:4793.982009234704 Step 166 | loss:0.000627187080681324 lr:4.782141965121128e-06 tokens_per_second_per_gpu:4692.159156595695 Step 167 | loss:8.381396128243068e-07 lr:4.397760490826481e-06 tokens_per_second_per_gpu:3617.0074058089476 Step 168 | loss:0.0012127950321882963 lr:4.028770535444615e-06 tokens_per_second_per_gpu:6421.134820675016 Step 169 | loss:4.570974124362692e-05 lr:3.675296639259912e-06 tokens_per_second_per_gpu:6406.2440034019 Step 170 | loss:0.0016346625052392483 lr:3.3374581056271447e-06 tokens_per_second_per_gpu:4832.4576941429605 Step 171 | loss:0.0013866540975868702 lr:3.0153689607045845e-06 tokens_per_second_per_gpu:4669.9487983815 Step 172 | loss:0.0026889871805906296 lr:2.7091379149682685e-06 tokens_per_second_per_gpu:4663.827255565714 Step 173 | loss:0.0013674983056262136 lr:2.4188683265204127e-06 tokens_per_second_per_gpu:4628.5059242993375 Step 174 | loss:0.002935702446848154 lr:2.1446581662042942e-06 tokens_per_second_per_gpu:4691.468123682847 Step 175 | loss:0.004059434402734041 lr:1.8865999845374793e-06 tokens_per_second_per_gpu:4653.043818434914 Step 176 | loss:0.0026409674901515245 lr:1.6447808804744668e-06 tokens_per_second_per_gpu:4878.260104369465 Step 177 | loss:0.0015465031610801816 lr:1.419282472009309e-06 tokens_per_second_per_gpu:6418.617830043843 Step 178 | loss:0.00817189458757639 lr:1.210180868628219e-06 tokens_per_second_per_gpu:4792.315895091995 Step 179 | loss:0.0015504483599215746 lr:1.0175466456213034e-06 tokens_per_second_per_gpu:4871.149434461099 Step 180 | loss:0.00481545040383935 lr:8.414448202622494e-07 tokens_per_second_per_gpu:2730.1446594882127 Step 181 | loss:0.002081669634208083 lr:6.819348298638839e-07 tokens_per_second_per_gpu:4669.819980068593 Step 182 | loss:0.0024928648490458727 lr:5.390705117171047e-07 tokens_per_second_per_gpu:4678.283477984429 Step 183 | loss:0.005283087957650423 lr:4.129000849198872e-07 tokens_per_second_per_gpu:2814.5930062504976 Step 184 | loss:0.0012949762167409062 lr:3.034661341025258e-07 tokens_per_second_per_gpu:2840.571429191292 Step 185 | loss:0.0006076831487007439 lr:2.1080559505462505e-07 tokens_per_second_per_gpu:6385.848503137581 Step 186 | loss:9.34275612962665e-06 lr:1.3494974225863322e-07 tokens_per_second_per_gpu:6429.516357906251 Step 187 | loss:0.006051680073142052 lr:7.59241783341913e-08 tokens_per_second_per_gpu:2718.71073842405 Step 188 | loss:0.005551817826926708 lr:3.374882539681767e-08 tokens_per_second_per_gpu:2686.2392676993068 Step 189 | loss:2.3192162643681513e-06 lr:8.437918333864536e-09 tokens_per_second_per_gpu:6412.661227271177 Step 190 | loss:5.593451078311773e-06 lr:0.0 tokens_per_second_per_gpu:4855.294777618601