tacit
0 Followers · 0 Following
  • Joined on 2024-05-31
Repositories
3
 Projects Packages Public Activity Starred Repositories
tacit synced and deleted reference refs/tags/refs/pull/51/merge at tacit/nanochat from mirror 2025-10-15 19:02:13 +00:00
tacit synced commits to refs/pull/19/merge at tacit/nanochat from mirror 2025-10-15 19:02:13 +00:00
cfb6e1b0bb Merge 1240e1299e into 190d9515d0
190d9515d0 dont evaluate the sampling evals during SFT they are too slow. keep the multiple choice evals. delete unused imports
b8076dd367 fix bug in learning rate multiplier, it was ramping up instead of ramping down. see more in Issue #68. also add --dry_run option useful for experimentation
Compare 3 commits »
tacit synced commits to refs/pull/18/merge at tacit/nanochat from mirror 2025-10-15 19:02:13 +00:00
4113f75864 Merge 49870bb540 into 190d9515d0
190d9515d0 dont evaluate the sampling evals during SFT they are too slow. keep the multiple choice evals. delete unused imports
b8076dd367 fix bug in learning rate multiplier, it was ramping up instead of ramping down. see more in Issue #68. also add --dry_run option useful for experimentation
49870bb540 Merge branch 'master' into launch-with-skypilot
67aaca98f5 export NANOCHAT_BASE_DIR so child processes get it too
Compare 6 commits »
tacit synced commits to refs/pull/18/head at tacit/nanochat from mirror 2025-10-15 19:02:13 +00:00
49870bb540 Merge branch 'master' into launch-with-skypilot
67aaca98f5 export NANOCHAT_BASE_DIR so child processes get it too
f0855cbcc7 Update speedrun.sh
dd6ff9a1cc fix bug in fallback case of find_largest_model
afaa5b4c90 Fix: Handle missing d<number> model tags in find_largest_model
Compare 7 commits »
tacit synced commits to refs/pull/17/merge at tacit/nanochat from mirror 2025-10-15 19:02:13 +00:00
64cebaef3b Merge de6eac26be into 190d9515d0
190d9515d0 dont evaluate the sampling evals during SFT they are too slow. keep the multiple choice evals. delete unused imports
b8076dd367 fix bug in learning rate multiplier, it was ramping up instead of ramping down. see more in Issue #68. also add --dry_run option useful for experimentation
Compare 3 commits »
tacit synced commits to refs/pull/15/merge at tacit/nanochat from mirror 2025-10-15 19:02:13 +00:00
9167bf570f Merge 6ef9f77789 into 190d9515d0
190d9515d0 dont evaluate the sampling evals during SFT they are too slow. keep the multiple choice evals. delete unused imports
b8076dd367 fix bug in learning rate multiplier, it was ramping up instead of ramping down. see more in Issue #68. also add --dry_run option useful for experimentation
Compare 3 commits »
tacit synced commits to refs/pull/13/merge at tacit/nanochat from mirror 2025-10-15 19:02:13 +00:00
e56aee8e30 Merge 07f193ab0d into 190d9515d0
190d9515d0 dont evaluate the sampling evals during SFT they are too slow. keep the multiple choice evals. delete unused imports
b8076dd367 fix bug in learning rate multiplier, it was ramping up instead of ramping down. see more in Issue #68. also add --dry_run option useful for experimentation
Compare 3 commits »
tacit synced commits to master at tacit/nanochat from mirror 2025-10-15 19:02:13 +00:00
190d9515d0 dont evaluate the sampling evals during SFT they are too slow. keep the multiple choice evals. delete unused imports
b8076dd367 fix bug in learning rate multiplier, it was ramping up instead of ramping down. see more in Issue #68. also add --dry_run option useful for experimentation
Compare 2 commits »
tacit synced commits to refs/pull/51/merge at tacit/nanochat from mirror 2025-10-15 10:52:14 +00:00
cb1f49e451 Merge bfd8d21313 into 67aaca98f5
bfd8d21313 Update architecture diagram in nanochat_architecture.jpg
2167425dba Merge branch 'karpathy:master' into master
f4efbe2825 Merge branch 'master' of https://github.com/HosLak/nanochat
9ff69c99b9 Replace architecture diagram in README with a new JPG format image and remove the old PNG file.
Compare 5 commits »
tacit synced commits to refs/pull/21/head at tacit/nanochat from mirror 2025-10-15 10:52:13 +00:00
837b43a504 feat: support mps
1c5dd2b7ba fix: get safe autocast dtype
Compare 2 commits »
tacit synced commits to refs/pull/51/head at tacit/nanochat from mirror 2025-10-15 10:52:13 +00:00
bfd8d21313 Update architecture diagram in nanochat_architecture.jpg
2167425dba Merge branch 'karpathy:master' into master
f4efbe2825 Merge branch 'master' of https://github.com/HosLak/nanochat
9ff69c99b9 Replace architecture diagram in README with a new JPG format image and remove the old PNG file.
67aaca98f5 export NANOCHAT_BASE_DIR so child processes get it too
Compare 6 commits »
tacit synced commits to refs/pull/50/merge at tacit/nanochat from mirror 2025-10-15 10:52:13 +00:00
e4692ea3a7 Merge 90e3bc778b into 67aaca98f5
90e3bc778b indent fix, limited device options to cuda/mps
Compare 2 commits »
tacit synced commits to refs/pull/50/head at tacit/nanochat from mirror 2025-10-15 10:52:13 +00:00
90e3bc778b indent fix, limited device options to cuda/mps
tacit synced commits to refs/pull/21/merge at tacit/nanochat from mirror 2025-10-15 10:52:13 +00:00
2575cb0b9a Merge 837b43a504 into 67aaca98f5
837b43a504 feat: support mps
1c5dd2b7ba fix: get safe autocast dtype
Compare 3 commits »
tacit synced and deleted reference refs/tags/refs/pull/52/merge at tacit/nanochat from mirror 2025-10-15 10:52:13 +00:00
tacit synced commits to refs/pull/35/merge at tacit/nanochat from mirror 2025-10-15 02:42:15 +00:00
2225350938 Merge 02440f670d into 67aaca98f5
67aaca98f5 export NANOCHAT_BASE_DIR so child processes get it too
f0855cbcc7 Update speedrun.sh
Compare 3 commits »
tacit synced commits to refs/pull/6/merge at tacit/nanochat from mirror 2025-10-15 02:42:15 +00:00
712500e71a Merge 6d51049077 into 67aaca98f5
67aaca98f5 export NANOCHAT_BASE_DIR so child processes get it too
f0855cbcc7 Update speedrun.sh
Compare 3 commits »
tacit synced commits to refs/pull/49/merge at tacit/nanochat from mirror 2025-10-15 02:42:15 +00:00
e9d0d7f9b6 Merge 938cb31f1a into 67aaca98f5
67aaca98f5 export NANOCHAT_BASE_DIR so child processes get it too
f0855cbcc7 Update speedrun.sh
Compare 3 commits »
tacit synced commits to refs/pull/43/merge at tacit/nanochat from mirror 2025-10-15 02:42:15 +00:00
6c88c11a49 Merge 53fe1cc18b into 67aaca98f5
67aaca98f5 export NANOCHAT_BASE_DIR so child processes get it too
f0855cbcc7 Update speedrun.sh
Compare 3 commits »
tacit synced commits to refs/pull/40/merge at tacit/nanochat from mirror 2025-10-15 02:42:15 +00:00
e08553f6b6 Merge 919ea572b0 into 67aaca98f5
67aaca98f5 export NANOCHAT_BASE_DIR so child processes get it too
f0855cbcc7 Update speedrun.sh
Compare 3 commits »