gräfics's picture

3 3 2

gräfics

graefics

·

https://www.openmachine.ai

AI & ML interests

None yet

Recent Activity

upvoted a paper 5 days ago

Flash normalization: fast RMSNorm for LLMs

upvoted a paper 5 days ago

Slim attention: cut your context memory in half without loss of accuracy -- K-cache is all you need for MHA

View all activity

Organizations

graefics's activity

New activity in HuggingFaceTB/SmolLM-135M 7 months ago

Are there two identical embedding tensors, even though embeddings are shared?

#15 opened 7 months ago by

Are there two identical embedding tensors, even though embeddings are shared?

#15 opened 7 months ago by