Hugging Face's logo
timaeus 's Collections
Datasets: Pile Subsets
Projects: Finetuning
Project: Lang2
Project: Lang1
Project: ICL1
Models: dh
Models: H-dh
Models: H
Models: L
Models: dm
Datasets: Suffixes
Datasets: Prefixes
Datasets: Delimiters
Datasets: Currencies
Datasets: Other

Models: H

Attention-only transformers, sweep over number of heads (for fixed head dimension)

Upvote
-