d_model SF Tech Week Social

​Care about AI interpretability and alignment? Join d_model and others in the space for an evening of good drinks, appetizers, and conversation.

​What we believe:

​Alignment is an unsolved problem with enormous stakes. We think interpretability is key to making models safer and more steerable.

​What we build:

​d_model is a fundamental AI research lab partnering with frontier labs to build RL environments for open-ended interpretability tasks. Alongside our partnerships, we also aim to use the agents we build for independent research.

​Explore our work:

• ​Using agents to create concept erasure methods

• ​Steering characters with interpretability

• ​Nullability representation in LLMs

​The location is a 5 minute walk from Moscone Center.

​Space is limited. We’re prioritizing attendees working in interpretability, alignment, RL, or similar fields.