tacit

tacit synced commits to refs/pull/409/merge at tacit/nanochat from mirror 2026-01-16 00:14:06 +00:00

532463334f Merge 489075bdbd into bdcc030ffa

bdcc030ffa oops legacy spurious line now

22a71aa3d3 fuse adamw into a single torch compiled kernel similar to muon. it's about 1.7X faster, but overall it's so tiny that it's not making a major dent

255f8b9af6 cleanly separate cpu and gpu sections

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

Compare 6 commits »

tacit synced commits to refs/pull/400/head at tacit/nanochat from mirror 2026-01-16 00:14:06 +00:00

a39c303912 Merge branch 'master' into fix/grad_acc_norm

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

3b50b77ed3 fix base_loss to report correct loss by switching the dataloader to the new default

Compare 45 commits »

tacit synced commits to refs/pull/400/merge at tacit/nanochat from mirror 2026-01-16 00:14:06 +00:00

28fab481b9 Merge a39c303912 into 6bb92403d5

a39c303912 Merge branch 'master' into fix/grad_acc_norm

Compare 2 commits »

tacit synced commits to refs/pull/407/merge at tacit/nanochat from mirror 2026-01-16 00:14:06 +00:00

4bd4b5221f Merge 9de0a121c5 into 6bb92403d5

9de0a121c5 Merge branch 'master' into fix-wandb-for-local-run

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

Compare 11 commits »

tacit synced commits to refs/pull/396/merge at tacit/nanochat from mirror 2026-01-16 00:14:05 +00:00

53d24e8f73 Merge 7f6219e092 into 6bb92403d5

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

Compare 3 commits »

tacit synced commits to refs/pull/399/head at tacit/nanochat from mirror 2026-01-16 00:14:05 +00:00

ff126c085e Merge branch 'master' into fix/loop

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

3b50b77ed3 fix base_loss to report correct loss by switching the dataloader to the new default

Compare 45 commits »

tacit synced commits to refs/pull/324/merge at tacit/nanochat from mirror 2026-01-16 00:14:05 +00:00

09354fa657 Merge e00c73322c into 6bb92403d5

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

Compare 3 commits »

tacit synced commits to refs/pull/312/merge at tacit/nanochat from mirror 2026-01-16 00:14:05 +00:00

414819fe76 Merge 26b0941f75 into 6bb92403d5

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

Compare 3 commits »

tacit synced commits to refs/pull/328/merge at tacit/nanochat from mirror 2026-01-16 00:14:05 +00:00

01b2f154ba Merge 02b22a5a13 into 6bb92403d5

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

3b50b77ed3 fix base_loss to report correct loss by switching the dataloader to the new default

Compare 10 commits »

tacit synced commits to refs/pull/393/merge at tacit/nanochat from mirror 2026-01-16 00:14:05 +00:00

6bc528fd4c Merge 89d2741cba into 6bb92403d5

89d2741cba Merge branch 'master' into issue-183-nvshmem-install-fix

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

Compare 38 commits »

tacit synced commits to refs/pull/393/head at tacit/nanochat from mirror 2026-01-16 00:14:05 +00:00

89d2741cba Merge branch 'master' into issue-183-nvshmem-install-fix

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

3b50b77ed3 fix base_loss to report correct loss by switching the dataloader to the new default

Compare 45 commits »

tacit synced commits to refs/pull/311/merge at tacit/nanochat from mirror 2026-01-16 00:14:04 +00:00

1dcae218ec Merge 81a958350a into 6bb92403d5

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

Compare 3 commits »

tacit synced commits to refs/pull/296/merge at tacit/nanochat from mirror 2026-01-16 00:14:04 +00:00

3bf4c99901 Merge 5172ea11bb into 6bb92403d5

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

Compare 3 commits »

tacit synced commits to refs/pull/258/head at tacit/nanochat from mirror 2026-01-16 00:14:04 +00:00

28b7dae0c3 restore

7cd3992f74 Merge branch 'master' into feat/add_dataset_progress_bar

d3679cd0a8 reverting change to .gitignore to prevent merge conflict

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

Compare 106 commits »

tacit synced commits to refs/pull/258/merge at tacit/nanochat from mirror 2026-01-16 00:14:04 +00:00

950a70e6a6 Merge 28b7dae0c3 into 6bb92403d5

28b7dae0c3 restore

7cd3992f74 Merge branch 'master' into feat/add_dataset_progress_bar

d3679cd0a8 reverting change to .gitignore to prevent merge conflict

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

Compare 40 commits »

tacit synced commits to refs/pull/151/merge at tacit/nanochat from mirror 2026-01-16 00:14:03 +00:00

3362c7dbb5 Merge 7f3154f025 into 6bb92403d5

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

Compare 3 commits »

tacit synced commits to refs/pull/204/merge at tacit/nanochat from mirror 2026-01-16 00:14:03 +00:00

414b4c38ab Merge 69b7cc9ac5 into 6bb92403d5

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

3b50b77ed3 fix base_loss to report correct loss by switching the dataloader to the new default

Compare 10 commits »

tacit synced commits to refs/pull/141/merge at tacit/nanochat from mirror 2026-01-16 00:14:03 +00:00

c240880542 Merge 65865df300 into 6bb92403d5

65865df300 Merge branch 'master' into master_goderr

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

Compare 11 commits »

tacit synced commits to master at tacit/nanochat from mirror 2026-01-16 00:14:02 +00:00

bdcc030ffa oops legacy spurious line now

22a71aa3d3 fuse adamw into a single torch compiled kernel similar to muon. it's about 1.7X faster, but overall it's so tiny that it's not making a major dent

255f8b9af6 cleanly separate cpu and gpu sections

Compare 3 commits »

tacit synced commits to refs/pull/141/head at tacit/nanochat from mirror 2026-01-16 00:14:02 +00:00

65865df300 Merge branch 'master' into master_goderr

6bb92403d5 changes and optimizations to muon, making it more efficient and simpler/cleaner a bit

3142ca1a28 minor helpful message

7312ec9898 fix buggy midtrain and update all kwargs to be idiomatic. that is, argparse uses dashes variables use underscores. the underscores are just a remnant of the previous Configurator object. This is the right way

3b50b77ed3 fix base_loss to report correct loss by switching the dataloader to the new default

Compare 44 commits »