Block a user
tacit
synced and deleted reference 2026-01-05 11:12:45 +00:00
refs/tags/refs/pull/394/merge at tacit/nanochat from mirror
9d4c9b786d
many small fixes to base_train: reporting ETA, allowing some additional kwarg flexibility, making sure we don't crash when e.g. depth = 11 - we now calculate the closest num_heads that works
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/407/merge at tacit/nanochat from mirror
2026-01-05 03:02:46 +00:00
9d4c9b786d
many small fixes to base_train: reporting ETA, allowing some additional kwarg flexibility, making sure we don't crash when e.g. depth = 11 - we now calculate the closest num_heads that works
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/405/merge at tacit/nanochat from mirror
2026-01-05 03:02:45 +00:00
9d4c9b786d
many small fixes to base_train: reporting ETA, allowing some additional kwarg flexibility, making sure we don't crash when e.g. depth = 11 - we now calculate the closest num_heads that works
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/400/merge at tacit/nanochat from mirror
2026-01-05 03:02:45 +00:00
9d4c9b786d
many small fixes to base_train: reporting ETA, allowing some additional kwarg flexibility, making sure we don't crash when e.g. depth = 11 - we now calculate the closest num_heads that works
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/399/merge at tacit/nanochat from mirror
2026-01-05 03:02:45 +00:00
9d4c9b786d
many small fixes to base_train: reporting ETA, allowing some additional kwarg flexibility, making sure we don't crash when e.g. depth = 11 - we now calculate the closest num_heads that works
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/370/merge at tacit/nanochat from mirror
2026-01-05 03:02:44 +00:00
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/396/merge at tacit/nanochat from mirror
2026-01-05 03:02:44 +00:00
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/296/merge at tacit/nanochat from mirror
2026-01-05 03:02:43 +00:00
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/312/merge at tacit/nanochat from mirror
2026-01-05 03:02:43 +00:00
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
9c60dfb64c
bump nanochat to use the latest stable pytorch that is 2.9.1 . Run e.g. to re-update your local environment if you git pull
tacit
synced commits to refs/pull/151/merge at tacit/nanochat from mirror
2026-01-05 03:02:42 +00:00
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
tacit
synced commits to refs/pull/141/merge at tacit/nanochat from mirror
2026-01-05 03:02:39 +00:00
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
9d4c9b786d
many small fixes to base_train: reporting ETA, allowing some additional kwarg flexibility, making sure we don't crash when e.g. depth = 11 - we now calculate the closest num_heads that works
962b6bfba3
alright add transformers as a dep of the repo because it should be easy to evaluate the CORE score of HF models. Not super happy about it but i tried it and the uv.lock doesn't get bloated as much as i expected
eb7bbc1b66
delete the configurator in favor of argparse and clean up a lot of kwarg details to make them more consistent across all scripts
aa42f40e66
delete the inline rustbpe project. it was ugly to have a project within project and rustbpe is now nicely a separate repo on my github karpathy/rustbpe and it's on pypi etc., so we just add it as a depedency to uv. i think it is appropriate that this is a separate repo because 1) it doesn't have too many knobs, other than the ones that are exposed - the regex pattern and vocab size and 2) all of its complexity is not algorithmic (it's equivalent to minbpe), instead it is efficiency-related, so it is ok to hide relatively speaking
tacit
synced commits to refs/pull/370/merge at tacit/nanochat from mirror
2026-01-04 18:52:37 +00:00
aa42f40e66
delete the inline rustbpe project. it was ugly to have a project within project and rustbpe is now nicely a separate repo on my github karpathy/rustbpe and it's on pypi etc., so we just add it as a depedency to uv. i think it is appropriate that this is a separate repo because 1) it doesn't have too many knobs, other than the ones that are exposed - the regex pattern and vocab size and 2) all of its complexity is not algorithmic (it's equivalent to minbpe), instead it is efficiency-related, so it is ok to hide relatively speaking
tacit
synced commits to refs/pull/393/merge at tacit/nanochat from mirror
2026-01-04 18:52:37 +00:00
aa42f40e66
delete the inline rustbpe project. it was ugly to have a project within project and rustbpe is now nicely a separate repo on my github karpathy/rustbpe and it's on pypi etc., so we just add it as a depedency to uv. i think it is appropriate that this is a separate repo because 1) it doesn't have too many knobs, other than the ones that are exposed - the regex pattern and vocab size and 2) all of its complexity is not algorithmic (it's equivalent to minbpe), instead it is efficiency-related, so it is ok to hide relatively speaking
tacit
synced commits to refs/pull/328/merge at tacit/nanochat from mirror
2026-01-04 18:52:37 +00:00
aa42f40e66
delete the inline rustbpe project. it was ugly to have a project within project and rustbpe is now nicely a separate repo on my github karpathy/rustbpe and it's on pypi etc., so we just add it as a depedency to uv. i think it is appropriate that this is a separate repo because 1) it doesn't have too many knobs, other than the ones that are exposed - the regex pattern and vocab size and 2) all of its complexity is not algorithmic (it's equivalent to minbpe), instead it is efficiency-related, so it is ok to hide relatively speaking
tacit
synced commits to refs/pull/399/merge at tacit/nanochat from mirror
2026-01-04 18:52:37 +00:00
aa42f40e66
delete the inline rustbpe project. it was ugly to have a project within project and rustbpe is now nicely a separate repo on my github karpathy/rustbpe and it's on pypi etc., so we just add it as a depedency to uv. i think it is appropriate that this is a separate repo because 1) it doesn't have too many knobs, other than the ones that are exposed - the regex pattern and vocab size and 2) all of its complexity is not algorithmic (it's equivalent to minbpe), instead it is efficiency-related, so it is ok to hide relatively speaking
aa42f40e66
delete the inline rustbpe project. it was ugly to have a project within project and rustbpe is now nicely a separate repo on my github karpathy/rustbpe and it's on pypi etc., so we just add it as a depedency to uv. i think it is appropriate that this is a separate repo because 1) it doesn't have too many knobs, other than the ones that are exposed - the regex pattern and vocab size and 2) all of its complexity is not algorithmic (it's equivalent to minbpe), instead it is efficiency-related, so it is ok to hide relatively speaking