Skip to content

feat(training-data): expand pre-training and chat training datasets w…#16

Open
Genstark wants to merge 11 commits into
tekaratzas:mainfrom
Genstark:add-data
Open

feat(training-data): expand pre-training and chat training datasets w…#16
Genstark wants to merge 11 commits into
tekaratzas:mainfrom
Genstark:add-data

Conversation

@Genstark

@Genstark Genstark commented Oct 2, 2025

Copy link
Copy Markdown

…ith additional examples

@tekaratzas

Copy link
Copy Markdown
Owner

Hey! We actually moved training samples into the JSON file in the data folder.

Can you move them there to keep consistent?

@Genstark

Genstark commented Oct 2, 2025

Copy link
Copy Markdown
Author

sure

@Genstark

Genstark commented Oct 2, 2025

Copy link
Copy Markdown
Author

done

@Genstark

Genstark commented Oct 3, 2025

Copy link
Copy Markdown
Author

How will this problem be solved?

@Genstark

Genstark commented Oct 3, 2025

Copy link
Copy Markdown
Author

now check if there is any build problem

@tekaratzas

Copy link
Copy Markdown
Owner

Looks like there is a conflict!

@Genstark

Genstark commented Oct 6, 2025

Copy link
Copy Markdown
Author

Check now if there is any problem now.

@tekaratzas

Copy link
Copy Markdown
Owner

Looks like a unit test + clippy failure!

can you take a look?

@Genstark

Genstark commented Oct 8, 2025

Copy link
Copy Markdown
Author

fix: resolve import errors and undeclared types in main.rs and llm.rs

  • Corrected use statements to properly import structs and constants from submodules:
    • Embeddings, Vocab, TransformerBlock, OutputProjection, LLM
    • EMBEDDING_DIM, HIDDEN_DIM, MAX_SEQ_LEN
  • Removed incorrect leading colons and fixed capitalization issues
  • Ensures all types are correctly recognized by Rust compiler

@Genstark

Copy link
Copy Markdown
Author

hello tekaratzas can you check current commit working or not

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.

2 participants