Skip to content

Add weight tying logic to LM head, i.e. Lingua does not tie weights.

1a6f2b4
Select commit
Loading
Failed to load commit list.
Open

Enable TransformerEngine-backed Tensor Parallelism with Llama3. #1483

Add weight tying logic to LM head, i.e. Lingua does not tie weights.
1a6f2b4
Select commit
Loading
Failed to load commit list.