Skip to content

Conversation

@al-jshen
Copy link
Member

@al-jshen al-jshen commented Apr 7, 2025

This is part of #6.

This adds a notebook to explain:

  • the different components of the AION model (tokenizers, transformer model)
  • how to load the different components
  • how to prepare data for the model and how to tokenize
  • how to do conditional sampling
  • how to do embedding-based adaptation for a downstream task

Major pain point at the moment is the tokenizer loading, which is being worked on by @LTMeyer. I will update the notebook to use his new interface when it is ready.

@al-jshen al-jshen requested a review from EiffL April 7, 2025 18:18
@EiffL EiffL requested a review from Copilot April 8, 2025 14:58
Copy link

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

Comments suppressed due to low confidence (1)

aion/utils.py:4

  • [nitpick] The function name 'to' is ambiguous; consider renaming it to a more descriptive name like 'move_to_device' to clarify its functionality.
def to(x, device='cuda'):

@EiffL
Copy link
Contributor

EiffL commented Jun 30, 2025

@al-jshen can you check what the latest notebooks look like, and see if you can adapt yours to be complementary?

@al-jshen
Copy link
Member Author

al-jshen commented Jul 1, 2025

@al-jshen can you check what the latest notebooks look like, and see if you can adapt yours to be complementary?

Looks like this is outdated and is also basically replaced by the StellarTutorial.ipynb notebook.

@al-jshen al-jshen closed this Jul 1, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants