Simran

simarora's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes

upvoted a collection 3 months ago

LoLCATS

Collection

Linearizing LLMs with high quality and efficiency. We linearize the full Llama 3.1 model family -- 8b, 70b, 405b -- for the first time! • 4 items • Updated Oct 14, 2024 • 15

upvoted a collection 10 months ago

based

Collection

These language model checkpoints are trained at the 360M and 1.3Bn parameter scales for up to 50Bn tokens on the Pile corpus, for research purposes. • 15 items • Updated Oct 18, 2024 • 9

upvoted a paper 10 months ago

Simple linear attention language models balance the recall-throughput tradeoff

Paper • 2402.18668 • Published Feb 28, 2024 • 18