Musings, playbooks and thoughts from the Yurts team
Ben Van Roo
You are more than a prompt, and your company needs more than a LLM
We see a daily set of Generative AI press releases and a gazillion GenAI tweets and LI posts. We hear constant calls for regulation, testimonies, and forecasts of massive economic disruption. New GenAI models, new acronyms, new companies. The dizzying pace of it all is exciting and offers promising breakthroughs, yet makes it difficult to understand the opportunity versus the noise.Amid the excitement, buzz, and frothiness that is GenAI in 2023, there’s an elephant in the room that has not yet been addressed: Enterprises are barely using it. GenAI is not driving major activities across the Fortune 2000 – it’s not being assigned to critical workflows or driving real revenue generating functions.
4 Minute Read
Illusions Unraveled: The Magic and Madness of Hallucinations in LLMs — Part 1
TL;DR We have benchmarked several popular open source LLMs (including the latest Llama-v2–7b-chat) to estimate both, the frequency and degree of hallucinations. Overall, we find that on average, popular, open-source models hallucinate close to 55% of the time on a context-aware Q&A task, when tested without any tuning...
11 Minute Read
© Yurts. 2023, San Francisco. All right reserved