Paper drop: Position Interpolation Improves ALiBi Extrapolation We found a simple method to 2x the context length of models that use ALiBi. This lets models like BTLM-3B-8K and MPT-7B-8K run high quality inference at up to 16K with no additional fine tuning.
@cerebras
-
India Team Doing Fundamental Engineering Work Says CEO
By
–
"We don't believe in an offshore development center model; we believe in hiring good engineers and giving them big projects to run. Our India team ..is doing fundamental work for us." @andrewdfeldman said this to @timesofindia in the article found here:
-
AI Platform Alliance Addresses Critical Compute Power Demands
By
–
(2/2) @deantak from @VentureBeat writes, “The AI Platform Alliance comes at a critical juncture, not only for the technology industry but also for the world as a whole. The rapid expansion of AI has led to an unprecedented need for compute power to train and run AI workloads."
-
Cerebras Unveils Largest Fastest AI Computer Chip
By
–
(1/3) MarketWatch recently wrote an article about Cerebras, describing our position in the AI landscape and highlighting our largest and fastest-ever computer chip, dedicated to AI. Read the full story here:
-
Architectural Advantage Drives Revenue Growth in AI Sector
By
–
(3/3) Additionally, Pat Moorhead, founder and chief analyst at Moor Insights and Strategy, says “For the sole fact that they are driving some pretty impressive revenue, they can establish a niche. They have an architectural advantage.”
-
Cerebras Architecture Handles Massive Data Processing for AI
By
–
(2/3) “Because of the way their system is architected, they can handle enormous amounts of data…it is an incredible solution for high-end data sets,” says Jim McGregor, an analyst with Tirias Research. “I would put Cerebras in [the] category of AI factory.”
-
Med42 LLM Passes USMLE with 72% Grade
By
–
(1/2) We are pleased to announce that our strategic partner M42 has developed Med42, an LLM fine-tuned on Llama-70B that passes the USMLE with a 72% grade. Learn more here:
-
Llama-70B Fine-Tuned for Medical AI Outperforms GPT-3.5
By
–
(2/2) We worked with Core42, a G42 company, to preprocess the medical dataset and fine-tune Llama-70B in 1 weekend. No code changes were required, and no infrastructure considerations were needed. This model performs above GPT-3.5 despite being smaller and trained on less data
-
Cerebras Condor Galaxy 1 Nominated for Data AI Award
By
–
Cerebras and our partner @G42ai have been nominated for the @BigDATAwireNews Readers’ Choice Awards for the Condor Galaxy 1 AI Supercomputer in the Top Data & AI Achievement category. Please vote for us – voting ends on October 25! https://
datanami.com/2023-bigdatawi
re-readers-choice-awards-polls-are-open/
… -
Cerebras Co-Founder Sean Lie Named to AI 100 List
By
–
@BusinessInsider recently published "The AI 100 2023: The top people in artificial intelligence". We are pleased to announce that Cerebras Chief Hardware Architect and Co-Founder, Sean Lie, made the list! See the full list and read more about Sean here: https://
businessinsider.com/the-ai-100-202
3-the-people-who-make-ai-intelligent-2023-10
…