-
Notifications
You must be signed in to change notification settings - Fork 25
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Make config.json consistent between shortfin and sharktank #487
base: main
Are you sure you want to change the base?
Conversation
return { | ||
"module_name": "module", | ||
"module_abi_version": 1, | ||
"max_seq_len": hp.context_length, | ||
"attn_head_count": hp.attention_head_count, | ||
# "attn_head_count": hp.attention_head_count, # we don't need the attention head count we just need the kvcache attention head count for shortfin |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Since all of the docs are updated and references to "attn_head_count" are removed, this is probably fine to delete. Otherwise, it'll sit here lingering for who knows how long
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good! We can probably remove the commented out line for attn_head_count
in export_paged_llm_v1.py
Added types and docstrings definitely clears things up
And remove the adaption layer in buidl_tools/integration_tests/llm/conftest.py