Commits

Breadcrumbs

History for

onmaster

User selector

All users

Datepicker

All time

Commit History

Commits on Feb 10, 2023

• fix misinformation in readme

  karpathy

  committed
  55c5069

  Copy full SHA for 55c5069

Commits on Feb 5, 2023

• also add a sampling/inference section

  karpathy

  committed
  f83dd03

  Copy full SHA for f83dd03

• revamp the readme file to be a bit better and more accessible, i hope

  karpathy

  committed
  23a8e70

  Copy full SHA for 23a8e70

Commits on Jan 31, 2023

• tune cited numbers and reproductions and more explicitly point out the problems w.r.t. the OWT vs WT domain gap

  karpathy

  committed
  d2705bd

  Copy full SHA for d2705bd

Commits on Jan 27, 2023

• ok i tried bringing back original init again and this time it makes a ton of difference and works much better than default. i'm not sure what was different with my earlier experiment where i saw a …

  Show description for f29a9ff
  karpathy

  committed
  f29a9ff

  Copy full SHA for f29a9ff

• grad clipping seems to slightly speed up training in the beginning but i can't see a big difference later in the training. it costs non-negligeable compute to clip. adding it for now because it is …

  Show description for 3cb3fc0
  karpathy

  committed
  3cb3fc0

  Copy full SHA for 3cb3fc0

Commits on Jan 20, 2023

• make mentions of mps in docs. ty good people in issue karpathy#28

  karpathy

  committed
  1f77d03

  Copy full SHA for 1f77d03

Commits on Jan 18, 2023

• the badge is a bit ugly, move it down to troubleshooting section

  karpathy

  committed
  2b083fb

  Copy full SHA for 2b083fb

• screwed up the link, fix

  karpathy

  committed
  aa8e4c2

  Copy full SHA for aa8e4c2

• experimenting with badges, and discord link to start specifically. issues sometimes feel a little too heavy

  karpathy

  committed
  6dab32c

  Copy full SHA for 6dab32c

Commits on Jan 16, 2023

• add docs on multinode training to main README too

  karpathy

  committed
  7f74652

  Copy full SHA for 7f74652

Commits on Jan 13, 2023

• it's true

  karpathy

  committed
  32b4f08

  Copy full SHA for 32b4f08

Commits on Jan 12, 2023

• including transformers as a dependency of the repo as well

  karpathy

  committed
  c1ac2d5

  Copy full SHA for c1ac2d5

• add note about windows and pytorch 2.0 and torch compile in general

  karpathy

  committed
  7f51d17

  Copy full SHA for 7f51d17

Commits on Jan 11, 2023

• oh no nanoGPT is trending quickly explain the character-level functionality I added late last night

  karpathy

  committed
  bb49751

  Copy full SHA for bb49751

Commits on Jan 8, 2023

• adjust the readme to reflect changes in the autocast branch

  karpathy

  committed
  8b2e622

  Copy full SHA for 8b2e622

• Make wandb import conditioned to wandb_log=True

  lantiga

  committed
  aba47f0

  Copy full SHA for aba47f0

• ran readme through spellchecker heh

  karpathy

  committed
  e53b9d2

  Copy full SHA for e53b9d2

• tune the readme with new header image and the loss curve for 124M

  karpathy

  committed
  df3b8a5

  Copy full SHA for df3b8a5

Commits on Jan 5, 2023

• shuttling the poor mans configurator aside into its own file and adding it to all of train,sample,bench. because i am leaving args in globals() so i can avoid having to prepend every single variabl…

  Show description for d562b3e
  karpathy

  committed
  d562b3e

  Copy full SHA for d562b3e

Commits on Jan 4, 2023

• mention current 8GPU SOTA and shuffle sections a bit

  karpathy

  committed
  ab04701

  Copy full SHA for ab04701

• Update README.md

  jorahn

  authored
  26aa5f3

  Copy full SHA for 26aa5f3

• add a notebook trying to reproduce chinchilla scaling laws. I can't get the numbers to be exactly right, have to look at more

  karpathy

  committed
  c72ecf5

  Copy full SHA for c72ecf5

Commits on Jan 3, 2023

• ty lambda labs

  karpathy

  committed
  5acba4b

  Copy full SHA for 5acba4b

• adding few more dependencies

  karpathy

  committed
  97fc426

  Copy full SHA for 97fc426

• flesh out the remaining TODOs in readme a bit more

  karpathy

  committed
  b45eec3

  Copy full SHA for b45eec3

Commits on Jan 1, 2023

• candidate changes to apis, have to think through more

  karpathy

  committed
  2febf44

  Copy full SHA for 2febf44

Commits on Dec 30, 2022

• simplify the prepare script a lot, write only using one process, seems sufficient for now. ty @LaihoE for suggestion and @proger for flagging

  karpathy

  committed
  7c6ea84

  Copy full SHA for 7c6ea84

• typo fix in readme

  karpathy

  committed
  d8abd21

  Copy full SHA for d8abd21

• add torch.compile by default, shows almost 1.8X improvement in throughput nice

  karpathy

  committed
  5a725d9

  Copy full SHA for 5a725d9

Commits on Dec 29, 2022

• add link to mingpt

  karpathy

  committed
  f88aa2c

  Copy full SHA for f88aa2c

• mention 4gpu loss as well in readme

  karpathy

  committed
  f2fc4be

  Copy full SHA for f2fc4be

• oops missed one # have to fix

  karpathy

  committed
  e7bac65

  Copy full SHA for e7bac65

• enhance readme, add some todos

  karpathy

  committed
  97e2ab1

  Copy full SHA for 97e2ab1

• add support for DDP training. the scaling timings right now do not look good by default, have to dig more into

  karpathy

  committed
  dea1507

  Copy full SHA for dea1507

Pagination

Previous

Next