Submissions from manifestai.com

		Symmetric Power Transformers (manifestai.com)
		3 points by ashvardanian 43 days ago \| past
		End of Transformer Era Approaches (manifestai.com)
		15 points by obiefernandez 64 days ago \| past \| 1 comment
		Brumby-14B-Base: The Strongest Attention-Free Base Model (manifestai.com)
		7 points by cgel 65 days ago \| past \| 1 comment
		Release: Power Retention (manifestai.com)
		2 points by frozenseven 3 months ago \| past
		Power Retention – Drop-In Replacement for Flash_attention (manifestai.com)
		7 points by Roritharr 3 months ago \| past \| 1 comment
		Symmetric Power Transformers (manifestai.com)
		18 points by _hark on Aug 15, 2024 \| past \| 5 comments
		Compute-Optimal Context Size (manifestai.com)
		3 points by panabee on June 24, 2024 \| past
		Linear transformers are faster after all (manifestai.com)
		116 points by JoelEinbinder on Jan 18, 2024 \| past \| 22 comments