4.6 KiB
4.6 KiB
| 1 | training_steps | training_tokens | training_loss | walltime | gputime | learning_rate |
|---|---|---|---|---|---|---|
| 2 | 125 | 524288000 | 1.9134495639801026 | 717.8844523429871 | 22972.302474975586 | 4.273942613508552e-05 |
| 3 | 250 | 1048576000 | 1.7826188468933106 | 1437.2500848770142 | 45992.00271606445 | 6.871273217257112e-05 |
| 4 | 375 | 1572864000 | 1.7930814552307128 | 2156.3654732704163 | 69003.69514465332 | 6.857037078589201e-05 |
| 5 | 500 | 2097152000 | 1.7946469068527222 | 2875.5332112312317 | 92017.06275939941 | 6.826908065704629e-05 |
| 6 | 625 | 2621440000 | 1.8028223133087158 | 3593.7442326545715 | 114999.81544494629 | 6.781024421798065e-05 |
| 7 | 750 | 3145728000 | 1.7957618522644043 | 4313.1157994270325 | 138019.70558166504 | 6.719600060023367e-05 |
| 8 | 875 | 3670016000 | 1.7825455617904664 | 5032.957694530487 | 161054.6462249756 | 6.642920197919011e-05 |
| 9 | 1000 | 4194304000 | 1.7887218809127807 | 5750.98313331604 | 184031.46026611328 | 6.551342085003853e-05 |
| 10 | 1125 | 4718592000 | 1.7748011398315429 | 6469.151132106781 | 207012.836227417 | 6.445289182011038e-05 |
| 11 | 1250 | 5242880000 | 1.7839254188537597 | 7187.22439289093 | 229991.18057250977 | 6.325256254058331e-05 |
| 12 | 1375 | 5767168000 | 1.7755695390701294 | 7905.280108451843 | 252968.96347045898 | 6.191799184307456e-05 |
| 13 | 1500 | 6291456000 | 1.7726542615890504 | 8624.770984649658 | 275992.67150878906 | 6.045540067134425e-05 |
| 14 | 1625 | 6815744000 | 1.7758274745941163 | 9342.781820297241 | 298969.0182495117 | 5.88715702178888e-05 |
| 15 | 1750 | 7340032000 | 1.7913320350646973 | 10060.986177921295 | 321951.55769348145 | 5.717386011383496e-05 |
| 16 | 1875 | 7864320000 | 1.78360032081604 | 10781.33857011795 | 345002.8342437744 | 5.537016477319412e-05 |
| 17 | 2000 | 8388608000 | 1.780481719970703 | 11499.381604194641 | 367980.2113342285 | 5.3468862461159006e-05 |
| 18 | 2125 | 8912896000 | 1.7762548875808717 | 12217.425961494446 | 390957.63076782227 | 5.147878619027324e-05 |
| 19 | 2250 | 9437184000 | 1.7626476001739502 | 12935.543794631958 | 413937.40142822266 | 4.940918006468564e-05 |
| 20 | 2375 | 9961472000 | 1.7646649742126466 | 13654.215188026428 | 436934.8860168457 | 4.726966653834097e-05 |
| 21 | 2500 | 10485760000 | 1.764815411567688 | 14372.40879535675 | 459917.081451416 | 4.507018093136139e-05 |
| 22 | 2625 | 11010048000 | 1.7657397699356079 | 15090.61161994934 | 482899.5718383789 | 4.282094232621603e-05 |
| 23 | 2750 | 11534336000 | 1.7624193906784058 | 15809.874806404114 | 505915.99380493164 | 4.053240627399646e-05 |
| 24 | 2875 | 12058624000 | 1.7604171276092528 | 16528.626129627228 | 528916.0361480713 | 3.82151993107982e-05 |
| 25 | 3000 | 12582912000 | 1.7540028762817383 | 17247.75738954544 | 551928.2364654541 | 3.588009349186905e-05 |
| 26 | 3125 | 13107200000 | 1.7573348188400268 | 17966.094574928284 | 574915.0263977051 | 3.353793727001175e-05 |
| 27 | 3250 | 13631488000 | 1.7569838047027588 | 18684.140555858612 | 597892.4977874756 | 3.1199611839838326e-05 |
| 28 | 3375 | 14155776000 | 1.7583934307098388 | 19404.312028884888 | 620937.9849243164 | 2.887598384404555e-05 |
| 29 | 3500 | 14680064000 | 1.7527799415588379 | 20122.6123547554 | 643923.5953521729 | 2.657784534676466e-05 |
| 30 | 3625 | 15204352000 | 1.7507912731170654 | 20840.710322856903 | 666902.7303314209 | 2.4315877453773282e-05 |
| 31 | 3750 | 15728640000 | 1.743433632850647 | 21559.86760377884 | 689915.7633209229 | 2.2100588466855697e-05 |
| 32 | 3875 | 16252928000 | 1.7478969144821166 | 22278.126459121704 | 712900.0466918945 | 1.9942272047046572e-05 |
| 33 | 4000 | 16777216000 | 1.7432158613204956 | 22996.365263462067 | 735883.6884307861 | 1.7850958101917058e-05 |
| 34 | 4125 | 17301504000 | 1.7420149660110473 | 23714.991760253906 | 758879.736328125 | 1.583636003488209e-05 |
| 35 | 4250 | 17825792000 | 1.7415733861923217 | 24435.1638174057 | 781925.2421569824 | 1.3907839274907019e-05 |
| 36 | 4375 | 18350080000 | 1.7456574296951295 | 25154.020771980286 | 804928.6647033691 | 1.2074357073288411e-05 |
| 37 | 4500 | 18874368000 | 1.7430942010879518 | 25872.437524795532 | 827918.000793457 | 1.0344432666897774e-05 |
| 38 | 4625 | 19398656000 | 1.7384707927703857 | 26590.624957084656 | 850899.998626709 | 8.726103260414675e-06 |
| 39 | 4750 | 19922944000 | 1.7505350017547607 | 27308.982837200165 | 873887.4507904053 | 7.226888101286022e-06 |
| 40 | 4875 | 20447232000 | 1.746962685585022 | 28027.27010011673 | 896872.6432037354 | 5.853752554685343e-06 |
| 41 | 5000 | 20971520000 | 1.7368433284759521 | 28745.597088336945 | 919859.1068267822 | 4.613077180692926e-06 |
| 42 | 5041 | 21143486464 | 1.731090293990241 | 29001.738576889038 | 928055.6344604492 | 4.235977030475624e-06 |
| 43 | 5166 | 21667774464 | 1.7286774969100953 | 29722.863705158234 | 951131.6385650635 | 3.1800113902136218e-06 |
| 44 | 5291 | 22192062464 | 1.7287598609924317 | 30444.640383720398 | 974228.4922790527 | 2.268928710691398e-06 |
| 45 | 5416 | 22716350464 | 1.7339081811904906 | 31166.346035003662 | 997323.0731201172 | 1.5069623486851924e-06 |
| 46 | 5541 | 23240638464 | 1.7407081651687621 | 31887.339103221893 | 1020394.8513031006 | 8.976522849479807e-07 |
| 47 | 5666 | 23764926464 | 1.742768988609314 | 32610.25783777237 | 1043528.2508087158 | 4.438297764863819e-07 |
| 48 | 5791 | 24289214464 | 1.7401256656646729 | 33331.42746925354 | 1066605.6790161133 | 1.4760334465790947e-07 |
| 49 | 5916 | 24813502464 | 1.729658899307251 | 34054.10992383957 | 1089731.5175628662 | 1.0349342716153842e-08 |
| 50 | 5956 | 24981274624 | 1.7287558913230896 | 34284.2702126503 | 1097096.6468048096 | 1.277759981155313e-10 |