0dac2ddeac
* llama4 support for linearized experts * clean up fsdp2 sharding to prevent hang * add yaml config * cleanup example [skip ci]
9 lines
73 B
Plaintext
9 lines
73 B
Plaintext
pre-commit
|
|
black
|
|
mypy
|
|
types-requests
|
|
quartodoc
|
|
jupyter
|
|
blobfile
|
|
tiktoken
|