Explore differences between language models through jet n-grams.
This page focuses on the Model Diff application of the JET Expansion framework. By utilizing jet n-grams, we can systematically analyze and compare the computational paths taken by different language models. This allows for a deeper understanding of model behavior without the need for additional data or retraining. The scores in the table are jet bi-gram scores computed with the jet expand method.
If the app doesn't load, click here.
For more information or inquiries, please reach out via email.