alma.liezenga

Message

Evaluating LLaMA 3 for political sycophancy

TLDR: I evaluated LLaMA v3 (8B + 70B) for political sycophancy using one of the two datasets I created. The results for this dataset suggest that sycophancy definitely occurs in a blatant way for both models though more clearly for 8B than for 70B. There are hints of politically tainted...

Sep 28, 20242

Two new datasets for evaluating political sycophancy in LLMs

TLDR: I created two datasets (154 and 759 statements) that can aid in measuring political sycophancy (in the US in particular) by combining a diverse set of political statements with quantitative data on the degree to which different political groups (dis)agree with those statements. The datasets can be found here....

Sep 28, 20249

LESSWRONG
LW

LESSWRONG
LW

alma.liezenga

alma.liezenga

alma.liezenga

alma.liezenga

Evaluating LLaMA 3 for political sycophancy

Two new datasets for evaluating political sycophancy in LLMs

Evaluating LLaMA 3 for political sycophancy

Two new datasets for evaluating political sycophancy in LLMs