tacit
0 Followers · 0 Following
  • Joined on 2024-05-31
Repositories
9
 Projects Packages Public Activity Starred Repositories
tacit synced and deleted reference refs/tags/refs/pull/128/merge at tacit/nanochat from mirror 2026-02-13 14:09:51 +00:00
tacit synced commits to refs/pull/498/head at tacit/nanochat from mirror 2026-02-13 05:59:52 +00:00
330fa1188c Merge origin/master into muonh
25ec1e6c43 Merge branch 'master' into muonh-submit
116900ac16 muonh
5a965c1383 Remove runs/scaling_laws_muonh.sh
fe2a80badd Replace torchao with minimal custom FP8 implementation
Compare 27 commits »
tacit synced commits to refs/pull/498/merge at tacit/nanochat from mirror 2026-02-13 05:59:52 +00:00
d408a85606 Merge 330fa1188c into 2f09686724
330fa1188c Merge origin/master into muonh
25ec1e6c43 Merge branch 'master' into muonh-submit
116900ac16 muonh
5a965c1383 Remove runs/scaling_laws_muonh.sh
Compare 21 commits »
tacit pushed to main at tacit/keys 2026-02-12 10:57:28 +00:00
f8e5b7fe7c add
tacit created branch main in tacit/keys 2026-02-12 10:57:28 +00:00
tacit created repository tacit/keys 2026-02-12 10:56:51 +00:00
tacit synced commits to refs/pull/425/merge at tacit/nanochat from mirror 2026-02-11 21:19:51 +00:00
ba7fccb9be Merge eebab89a11 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
e569b59f92 delete torchao dependency, create our own exact API-matched version of Float8Linear, document it very well. for some poorly understood reason, the performance is not only ~identical but actually runs 3% faster. despite of it being significantly simpler and much less code. i don't fully understand why/how atm
Compare 3 commits »
tacit synced commits to refs/pull/486/merge at tacit/nanochat from mirror 2026-02-11 21:19:51 +00:00
ba63d4d650 Merge 7c0eb3f00b into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »
tacit synced commits to refs/pull/515/merge at tacit/nanochat from mirror 2026-02-11 21:19:51 +00:00
d6ee33eda5 Merge 2ae28292aa into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »
tacit synced commits to refs/pull/414/merge at tacit/nanochat from mirror 2026-02-11 21:19:51 +00:00
ac63a77462 Merge 57bcf6786e into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
e569b59f92 delete torchao dependency, create our own exact API-matched version of Float8Linear, document it very well. for some poorly understood reason, the performance is not only ~identical but actually runs 3% faster. despite of it being significantly simpler and much less code. i don't fully understand why/how atm
1ec0a34779 at 28 and above we start to need batch size 8
ff46300720 tune miniseries just a bit, fairly cosmetic, keep to even depths where the math works out nicely in model sizing
Compare 7 commits »
tacit synced commits to refs/pull/483/merge at tacit/nanochat from mirror 2026-02-11 13:09:55 +00:00
338f80ddcc Merge 56660c690b into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »
tacit synced commits to refs/pull/489/merge at tacit/nanochat from mirror 2026-02-11 13:09:55 +00:00
bc8fa006ae Merge 79b7b04ca0 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »
tacit synced commits to refs/pull/509/merge at tacit/nanochat from mirror 2026-02-11 13:09:55 +00:00
05a1f4ba81 Merge 9caf6690a1 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »
tacit synced commits to refs/pull/510/merge at tacit/nanochat from mirror 2026-02-11 13:09:55 +00:00
d06469a3c0 Merge f999d06d58 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »
tacit synced commits to refs/pull/511/merge at tacit/nanochat from mirror 2026-02-11 13:09:55 +00:00
dd6ae8a198 Merge 5ba77a31b9 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »
tacit synced commits to refs/pull/328/merge at tacit/nanochat from mirror 2026-02-11 13:09:54 +00:00
77829af59b Merge 02b22a5a13 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
e569b59f92 delete torchao dependency, create our own exact API-matched version of Float8Linear, document it very well. for some poorly understood reason, the performance is not only ~identical but actually runs 3% faster. despite of it being significantly simpler and much less code. i don't fully understand why/how atm
1ec0a34779 at 28 and above we start to need batch size 8
ff46300720 tune miniseries just a bit, fairly cosmetic, keep to even depths where the math works out nicely in model sizing
Compare 7 commits »
tacit synced commits to refs/pull/442/merge at tacit/nanochat from mirror 2026-02-11 13:09:54 +00:00
979846811c Merge 3bf06802e6 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »
tacit synced commits to refs/pull/151/merge at tacit/nanochat from mirror 2026-02-11 13:09:53 +00:00
ed9dc84caa Merge 7f3154f025 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
e569b59f92 delete torchao dependency, create our own exact API-matched version of Float8Linear, document it very well. for some poorly understood reason, the performance is not only ~identical but actually runs 3% faster. despite of it being significantly simpler and much less code. i don't fully understand why/how atm
1ec0a34779 at 28 and above we start to need batch size 8
ff46300720 tune miniseries just a bit, fairly cosmetic, keep to even depths where the math works out nicely in model sizing
Compare 7 commits »
tacit synced commits to refs/pull/141/merge at tacit/nanochat from mirror 2026-02-11 13:09:52 +00:00
210c16221f Merge 65865df300 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
e569b59f92 delete torchao dependency, create our own exact API-matched version of Float8Linear, document it very well. for some poorly understood reason, the performance is not only ~identical but actually runs 3% faster. despite of it being significantly simpler and much less code. i don't fully understand why/how atm
Compare 3 commits »
tacit synced commits to refs/pull/516/merge at tacit/nanochat from mirror 2026-02-11 04:59:51 +00:00
9061480975 Merge 00932d1955 into 2f09686724
2f09686724 clarify that this is bf16 mfu we're talking about
Compare 2 commits »