Attribution methods

Slides of the session which looks at the internal workings of LLMs and different techniques for trying to understand them (e.g., transformer attention visualization, feature attribution methods, probing), can be found here.

Additional materials

If you want to dig a bit deeper, here are (optional!) supplementary readings:

• McCoy et al. (2019) Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference

• Tenney et al. (2019) What do you learn from context? Probing for sentence structure in contextualized word representations

• Elazar et al. (2021) Amnesic Probing: Behavioral Explanation with Amnesic Counterfactuals