2 articles
CMU and Meta's STEM framework swaps Transformer up-projections for token-indexed embedding lookups, trimming a third of feed-forward parameters while gaining 3-4% on MMLU benchmarks.
Essential AI's new Rnj-1 model packs 8 billion parameters but delivers performance that rivals much larger AI systems. The open-source release scores impressively across coding and reasoning tests while using significantly less computing power than its competitors.
We use cookies to improve your experience on our site and to show you relevant advertising. To find our more, read our privacy policy and cookie policy
Peter Smith
Eseandre Mordi
