Join an LLM Evaluations Working Group (OpenAI + EA Czechia)

Join an LLM Evaluations Working Group (OpenAI + EA Czechia)

OpenAI
Opportunity type
Independent project
Cause areas
AI Safety & Policy
Routes to impact
💡 Direct/Increased Engagement with EA
🧪 Testing Your Fit for a Certain Career Path
📖 Learning about Important Cause Areas
📈 Skill-Building & Building Career Capital
Relevant aptitudes
Conceptual & Empirical Research
Communicator
Software Engineering
Location
Czech Republic or Remote (Europe)
Description
Our goal is to test our own scenarios and assist in writing evaluations for large language models. We welcome anyone who is willing to delve deep enough to be able to produce quality evaluations and gain a wealth of experience working with the models.
We aim to focus on manipulation detection and situational-awareness. The plan is to:
  1. We have a call on 26th of June 7pm CET, fill the form below or contact Hana if you want to be included. We will set goals based on our capacities and preferred collaborative setup, you can join later as well.
  2. Get together at https://alignmentjam.com/jam/benchmarks either in Prague (https://www.facebook.com/events/933647194582086) or in your closest hosting city. Not necessary for participating in the working group, but it will provide some relevant context.
Join 60k subscribers and sign up for the EA Newsletter, a monthly email with the latest ideas and opportunities