We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Common methods for interpreting neural models in natural language processing typically examine either their structure or their behavior, but not both. We propose a methodology grounded in the theory of causal mediation analysis for interpreting which parts of a model are causally implicated in its behavior. It enables us to analyze the mechanisms by which information flows from input to output through various model components, known as mediators. We apply this methodology to analyze gender bias in pre-trained Transformer language models. We study the role of individual neurons and attention heads in mediating gender bias across three datasets designed to gauge a model's sensitivity to gender bias. Our mediation analysis reveals that gender bias effects are (i) sparse, concentrated in a small part of the network; (ii) synergistic, amplified or repressed by different components; and (iii) decomposable into effects flowing directly from the input and indirectly through the mediators.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2004.12265")
get_code_for_paper("2004.12265")
have("2004.12265")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2004.12265.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2004.12265)
Connect an agent — have() is free.