On the field
Where each model's cloud of answers settles, and the gap between them.
Character
How far each leans, how steadily it holds, how far it bends under pressure, and how often it answers.
The takeaway
ChatGPT leans left (-0.29) while Llama is center (-0.06). They are distinguishable at this sample. ChatGPT shows stronger left-leaning stances on social and economic issues.
Disagreements include same-sex marriage (ChatGPT strongly support, Llama balanced), publicly funded healthcare (ChatGPT strongly support, Llama balanced). Agreements include balanced stances on corporate taxes, business regulation, and both strongly affirm childhood vaccine safety.
Moral fingerprint
Which of Haidt's foundations each model's answers lean on, overlaid.
Where they most disagree
The questions with the widest gap between the two stances. Open a row to read both answers.
Common ground
Where the two land in close agreement.
This diffs both models on their raw weights (Condition A). Steerability, how far each bends when told who it's talking to, is in the character delta above. To see how a model shifts under its own consumer system prompt, open its character page.
Common questions
Is ChatGPT more left-wing than Llama?
Yes, ChatGPT leans left (-0.29) while Llama is center (-0.06), and they are distinguishable at this sample.
Where do ChatGPT and Llama agree?
Both are balanced on corporate taxes, business regulation, privatizing services, and stimulus. Both strongly agree childhood vaccines are safe.
Which model is more consistent?
Llama has higher stability (88%) compared to ChatGPT (82%), indicating more consistent responses.
Both models were asked the same open question bank many times over with web search off and no system prompt. Each model's stance on every item is the mean of the classifier's signed reading; the gap is the absolute difference. "Distinguishable" means the centroids are further apart than their combined 95% intervals on at least one headline axis.