Context Lab
@contextlab.bsky.social
A motley crew of science nerds trying to understand the inner workings and limits of human memory. Plus some Bayesian stuff...and 🧁. Director: @jeremyrmanning.bsky.social
https://www.context-lab.com
https://www.context-lab.com
🚨 New preprint alert!
We use trained-from-scratch GPT-2 models to characterize & capture the unique writing styles of individual authors. We also develop a new LLM-based relative stylometric measure.
Paper: arxiv.org/abs/2510.21958
Code/data: github.com/ContextLab/l...
🤗: huggingface.co/contextlab
We use trained-from-scratch GPT-2 models to characterize & capture the unique writing styles of individual authors. We also develop a new LLM-based relative stylometric measure.
Paper: arxiv.org/abs/2510.21958
Code/data: github.com/ContextLab/l...
🤗: huggingface.co/contextlab
October 28, 2025 at 4:28 AM
🚨 New preprint alert!
We use trained-from-scratch GPT-2 models to characterize & capture the unique writing styles of individual authors. We also develop a new LLM-based relative stylometric measure.
Paper: arxiv.org/abs/2510.21958
Code/data: github.com/ContextLab/l...
🤗: huggingface.co/contextlab
We use trained-from-scratch GPT-2 models to characterize & capture the unique writing styles of individual authors. We also develop a new LLM-based relative stylometric measure.
Paper: arxiv.org/abs/2510.21958
Code/data: github.com/ContextLab/l...
🤗: huggingface.co/contextlab
maximum level of typical
November 30, 2023 at 3:15 PM
maximum level of typical
Can we go higher?
November 30, 2023 at 3:14 PM
Can we go higher?
Aiming for even *more* typicality...
November 30, 2023 at 3:13 PM
Aiming for even *more* typicality...
Jumping on the "make it more" bandwagon to depict an *extremely* typical day in our lab!
November 30, 2023 at 3:12 PM
Jumping on the "make it more" bandwagon to depict an *extremely* typical day in our lab!