I am a researcher at EleutherAI studying AI and ML, especially LLMs. My work thus far has focused on supporting open study of LLM behaviors and internals, reproducible evaluation of language model capabilities, and the engineering and systems work that goes into large-scale distributed DL training.

It seems clear that AI systems are being used and deployed widely, whether or not we yet fully understand the implications of this, and I’m excited about any research that demystifies these black boxes’ inner workings, or puts tools in the hands of users or model creators to make systems more steerable and elicit desired behaviors.

I’ve also done work on AI for Mathematics and on improving the multilinguality of language models.

For a full list of my research publications, see my Google Scholar page.

Contact me

One of the best ways to get in touch is on Twitter, where I post research-related thoughts.

You can reach me at hailey.ruth.schoelkopf@gmail.com or, for EleutherAI-related research collaborations or questions, hailey@eleuther.ai.