8.9 KiB
8.9 KiB
| 1 | training_steps | training_tokens | training_loss | walltime | gputime | learning_rate |
|---|---|---|---|---|---|---|
| 2 | 125 | 524288000 | 1.3418247509002685 | 931.8152546882629 | 477089.4104003906 | 1.424689071427565e-05 |
| 3 | 250 | 1048576000 | 1.3494457769393922 | 1892.8055691719055 | 969116.4514160156 | 2.2908012397238053e-05 |
| 4 | 375 | 1572864000 | 1.3450150632858275 | 2841.477949619293 | 1454836.7102050781 | 2.2896543669048697e-05 |
| 5 | 500 | 2097152000 | 1.3463186407089234 | 3801.364312171936 | 1946298.5278320312 | 2.28722292376915e-05 |
| 6 | 625 | 2621440000 | 1.357196674346924 | 4742.318592071533 | 2428067.119140625 | 2.2835098206996918e-05 |
| 7 | 750 | 3145728000 | 1.3642438173294067 | 5690.46168088913 | 2913516.3806152344 | 2.278519059473183e-05 |
| 8 | 875 | 3670016000 | 1.3604784631729125 | 6645.177431106567 | 3402330.8447265625 | 2.272256460855715e-05 |
| 9 | 1000 | 4194304000 | 1.350071816444397 | 7589.283299446106 | 3885713.0493164062 | 2.2647287551080808e-05 |
| 10 | 1125 | 4718592000 | 1.359410786628723 | 8539.165923595428 | 4372052.952880859 | 2.2559446733794175e-05 |
| 11 | 1250 | 5242880000 | 1.367375102043152 | 9474.964418411255 | 4851181.7822265625 | 2.245913856313564e-05 |
| 12 | 1375 | 5767168000 | 1.3527356910705566 | 10430.55284500122 | 5340443.056640625 | 2.2346477635437623e-05 |
| 13 | 1500 | 6291456000 | 1.350779275894165 | 11383.854374885559 | 5828533.439941406 | 2.222158946096897e-05 |
| 14 | 1625 | 6815744000 | 1.3523637342453003 | 12330.729007720947 | 6313333.251953125 | 2.2084612282924354e-05 |
| 15 | 1750 | 7340032000 | 1.3702508544921874 | 13284.42476272583 | 6801625.478515625 | 2.193570435338188e-05 |
| 16 | 1875 | 7864320000 | 1.3592170333862306 | 14231.939177513123 | 7286752.858886719 | 2.1775029381387867e-05 |
| 17 | 2000 | 8388608000 | 1.375773310661316 | 15178.148682117462 | 7771212.125244141 | 2.1602767446893267e-05 |
| 18 | 2125 | 8912896000 | 1.355927152633667 | 16116.971938610077 | 8251889.632568359 | 2.141911318176426e-05 |
| 19 | 2250 | 9437184000 | 1.3739105558395386 | 17064.795393943787 | 8737175.241699219 | 2.122427395079285e-05 |
| 20 | 2375 | 9961472000 | 1.3521932220458985 | 18012.379891872406 | 9222338.504638672 | 2.1018466213718057e-05 |
| 21 | 2500 | 10485760000 | 1.3504028701782227 | 18967.846796512604 | 9711537.559814453 | 2.0801922801183537e-05 |
| 22 | 2625 | 11010048000 | 1.362982177734375 | 19911.94496870041 | 10194915.82397461 | 2.057488563877996e-05 |
| 23 | 2750 | 11534336000 | 1.3546034097671509 | 20865.127074718475 | 10682945.06225586 | 2.0337609385023825e-05 |
| 24 | 2875 | 12058624000 | 1.3497952461242675 | 21813.99649620056 | 11168766.206054688 | 2.009036143135745e-05 |
| 25 | 3000 | 12582912000 | 1.349095754623413 | 22763.452064990997 | 11654887.45727539 | 1.983342008315958e-05 |
| 26 | 3046 | 12775849984 | 1.3463608264923095 | 23145.171978473663 | 11850328.052978516 | 1.973648068087641e-05 |
| 27 | 3171 | 13300137984 | 1.3599746942520141 | 24092.784764766693 | 12335505.799560547 | 1.9466746380203404e-05 |
| 28 | 3296 | 13824425984 | 1.3763371849060058 | 25021.107606887817 | 12810807.094726562 | 1.9188015357940458e-05 |
| 29 | 3421 | 14348713984 | 1.3610063695907593 | 25969.015357494354 | 13296135.86303711 | 1.890060411824379e-05 |
| 30 | 3546 | 14873001984 | 1.3589703798294068 | 26897.04850435257 | 13771288.834228516 | 1.860483462223783e-05 |
| 31 | 3671 | 15397289984 | 1.3462161684036256 | 27841.132175922394 | 14254659.674072266 | 1.830103610700462e-05 |
| 32 | 3796 | 15921577984 | 1.3638871717453003 | 28781.747272014618 | 14736254.603271484 | 1.798955418053083e-05 |
| 33 | 3921 | 16445865984 | 1.3659167861938477 | 29728.183631896973 | 15220830.01953125 | 1.767073445080314e-05 |
| 34 | 4046 | 16970153984 | 1.3543964910507202 | 30663.842606544495 | 15699887.414550781 | 1.734493525873404e-05 |
| 35 | 4171 | 17494441984 | 1.3538448762893678 | 31585.340785980225 | 16171694.482421875 | 1.7012525859172456e-05 |
| 36 | 4296 | 18018729984 | 1.3664672470092774 | 32533.905737400055 | 16657359.737548828 | 1.6673875506967306e-05 |
| 37 | 4421 | 18543017984 | 1.3337582159042358 | 33466.050419807434 | 17134617.814941406 | 1.6329366189893335e-05 |
| 38 | 4546 | 19067305984 | 1.3535391807556152 | 34392.57918596268 | 17609000.54321289 | 1.5979385352693498e-05 |
| 39 | 4671 | 19591593984 | 1.3478158473968507 | 35334.314131736755 | 18091168.83544922 | 1.5624324078089558e-05 |
| 40 | 4796 | 20115881984 | 1.353451647758484 | 36272.28346824646 | 18571409.135742188 | 1.526458072476089e-05 |
| 41 | 4921 | 20640169984 | 1.3361215162277222 | 37208.85023355484 | 19050931.319580078 | 1.4900561836839188e-05 |
| 42 | 5046 | 21164457984 | 1.3517218732833862 | 38133.61739397049 | 19524412.10571289 | 1.4532673048961442e-05 |
| 43 | 5171 | 21688745984 | 1.355444827079773 | 39076.93432569504 | 20007390.37475586 | 1.4161330000206362e-05 |
| 44 | 5296 | 22213033984 | 1.3552981281280518 | 40017.45046854019 | 20488934.639892578 | 1.3786947420157958e-05 |
| 45 | 5421 | 22737321984 | 1.3426275634765625 | 40962.5593829155 | 20972830.404052734 | 1.3409946404863149e-05 |
| 46 | 5546 | 23261609984 | 1.3578113174438478 | 41901.87567472458 | 21453760.345458984 | 1.3030750778852962e-05 |
| 47 | 5671 | 23785897984 | 1.3435347986221313 | 42839.63329076767 | 21933892.244873047 | 1.2649785276153125e-05 |
| 48 | 5796 | 24310185984 | 1.3542452430725098 | 43775.769159793854 | 22413193.809814453 | 1.2267478268768173e-05 |
| 49 | 5921 | 24834473984 | 1.3246106481552125 | 44706.83588504791 | 22889899.97314453 | 1.188425812870264e-05 |
| 50 | 6046 | 25358761984 | 1.3361506652832031 | 45649.95430231094 | 23372776.602783203 | 1.1500555956445169e-05 |
| 51 | 6171 | 25883049984 | 1.3631443643569947 | 46588.49558591843 | 23853309.739990234 | 1.1116801942989696e-05 |
| 52 | 6296 | 26407337984 | 1.3622448348999023 | 47526.081244945526 | 24333353.59741211 | 1.0733427188824862e-05 |
| 53 | 6421 | 26931625984 | 1.3260996294021608 | 48466.43445968628 | 24814814.443359375 | 1.0350861884944607e-05 |
| 54 | 6546 | 27455913984 | 1.3378325510025024 | 49397.44676113129 | 25291492.74169922 | 9.969535312848166e-06 |
| 55 | 6671 | 27980201984 | 1.3243994665145875 | 50339.94748592377 | 25774053.11279297 | 9.589875844540074e-06 |
| 56 | 6796 | 28504489984 | 1.3497977590560912 | 51262.73466348648 | 26246520.147705078 | 9.212310033035465e-06 |
| 57 | 6921 | 29028777984 | 1.3282609605789184 | 52207.10548400879 | 26730038.0078125 | 8.83726079337066e-06 |
| 58 | 7046 | 29553065984 | 1.3322417497634889 | 53140.394983291626 | 27207882.231445312 | 8.465149221592583e-06 |
| 59 | 7171 | 30077353984 | 1.3372080469131469 | 54082.02884435654 | 27689998.768310547 | 8.096394594758749e-06 |
| 60 | 7296 | 30601641984 | 1.3316910409927367 | 55022.170560359955 | 28171351.326904297 | 7.731408913969062e-06 |
| 61 | 7421 | 31125929984 | 1.3499228382110595 | 55957.524094581604 | 28650252.33642578 | 7.370603270828724e-06 |
| 62 | 7546 | 31650217984 | 1.334327540397644 | 56884.41381931305 | 29124819.87548828 | 7.014381935732672e-06 |
| 63 | 7671 | 32174505984 | 1.326025242805481 | 57829.13219690323 | 29608515.684814453 | 6.663145086349687e-06 |
| 64 | 7796 | 32698793984 | 1.3477921724319457 | 58760.65451145172 | 30085455.10986328 | 6.317286988632986e-06 |
| 65 | 7921 | 33223081984 | 1.3305214691162108 | 59694.21422958374 | 30563437.685546875 | 5.977195996820228e-06 |
| 66 | 8046 | 33747369984 | 1.33052716255188 | 60627.17547893524 | 31041113.845214844 | 5.643253643938806e-06 |
| 67 | 8171 | 34271657984 | 1.341693778038025 | 61568.85575532913 | 31523254.146728516 | 5.315835096553201e-06 |
| 68 | 8296 | 34795945984 | 1.3412931156158447 | 62498.83219957352 | 31999402.08618164 | 4.995307790522929e-06 |
| 69 | 8421 | 35320233984 | 1.3353554916381836 | 63425.036759376526 | 32473618.82080078 | 4.682031431002542e-06 |
| 70 | 8546 | 35844521984 | 1.329939103126526 | 64355.69224357605 | 32950114.428710938 | 4.376357992441626e-06 |
| 71 | 8671 | 36368809984 | 1.333136568069458 | 65297.09375858307 | 33432112.00439453 | 4.078630354342749e-06 |
| 72 | 8796 | 36893097984 | 1.3331001377105713 | 66226.41993284225 | 33907927.005615234 | 3.7891829833824886e-06 |
| 73 | 8921 | 37417385984 | 1.3310446119308472 | 67165.35197019577 | 34388660.208740234 | 3.508340569169377e-06 |
| 74 | 9046 | 37941673984 | 1.336583514213562 | 68108.11698198318 | 34871355.89477539 | 3.2364184789912542e-06 |
| 75 | 9171 | 38465961984 | 1.324005627632141 | 69037.88452148438 | 35347396.875 | 2.9737220756942406e-06 |
| 76 | 9296 | 38990249984 | 1.3519883394241332 | 69982.61614084244 | 35831099.46411133 | 2.7205460355617106e-06 |
| 77 | 9421 | 39514537984 | 1.3276349020004272 | 70917.0075082779 | 36309507.84423828 | 2.4771748030616436e-06 |
| 78 | 9546 | 40038825984 | 1.336574878692627 | 71853.87295722961 | 36789182.95410156 | 2.243881453978247e-06 |
| 79 | 9671 | 40563113984 | 1.3190159940719604 | 72783.2350897789 | 37265016.3659668 | 2.0209279227856314e-06 |
| 80 | 9796 | 41087401984 | 1.337507767677307 | 73709.91152763367 | 37739474.70214844 | 1.808564434213622e-06 |
| 81 | 9921 | 41611689984 | 1.3189802169799805 | 74631.14792108536 | 38211147.7355957 | 1.6070295032477588e-06 |
| 82 | 10046 | 42135977984 | 1.3195102834701538 | 75562.1107006073 | 38687800.67871094 | 1.4165493666951079e-06 |
| 83 | 10171 | 42660265984 | 1.3336253833770753 | 76506.02549314499 | 39171085.052490234 | 1.237337642123748e-06 |
| 84 | 10296 | 43184553984 | 1.3461341094970702 | 77471.05675697327 | 39665181.05957031 | 1.0695957826101221e-06 |
| 85 | 10421 | 43708841984 | 1.3408458995819093 | 78418.21376562119 | 40150125.44799805 | 9.13512053557497e-07 |
| 86 | 10546 | 44233129984 | 1.324405198097229 | 79354.13931846619 | 40629319.33105469 | 7.69261532695964e-07 |
| 87 | 10671 | 44757417984 | 1.3204831790924072 | 80294.47205543518 | 41110769.69238281 | 6.370062237692764e-07 |
| 88 | 10796 | 45281705984 | 1.329838457107544 | 81225.9842967987 | 41587703.95996094 | 5.168946586309175e-07 |
| 89 | 10921 | 45805993984 | 1.3304166746139527 | 82172.30509757996 | 42072220.20996094 | 4.0906155618358753e-07 |
| 90 | 11046 | 46330281984 | 1.330317463874817 | 83097.30528354645 | 42545820.30517578 | 3.136280497528787e-07 |
| 91 | 11171 | 46854569984 | 1.3396130466461182 | 84041.39386415482 | 43029193.658447266 | 2.3070124655077962e-07 |
| 92 | 11296 | 47378857984 | 1.3372651576995849 | 84988.65896463394 | 43514193.38989258 | 1.603742560973842e-07 |
| 93 | 11421 | 47903145984 | 1.3238491821289062 | 85924.85509634018 | 43993525.80932617 | 1.0272601969063544e-07 |
| 94 | 11546 | 48427433984 | 1.3198965358734132 | 86849.85299825668 | 44467124.73510742 | 5.782125001019267e-08 |
| 95 | 11671 | 48951721984 | 1.3250528430938722 | 87793.58383893967 | 44950314.92553711 | 2.5710360063158078e-08 |
| 96 | 11796 | 49476009984 | 1.334643521308899 | 88735.2750492096 | 45432460.82519531 | 6.42939435024914e-09 |
| 97 | 11916 | 49979326464 | 1.3240145593881607 | 89640.23823976517 | 45895801.978759766 | 1.0287991812629738e-11 |