Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
If the algorithm works, it works. manually altering it and seeing a drop in performance is not good.
reddit AI Harm Incident 1555765303.0 ♥ 4
Coding Result
DimensionValue
Responsibilityai_itself
Reasoningconsequentialist
Policynone
Emotionindifference
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[ {"id":"rdc_g9aphso","responsibility":"company","reasoning":"virtue","policy":"industry_self","emotion":"approval"}, {"id":"rdc_g9b5gic","responsibility":"company","reasoning":"deontological","policy":"ban","emotion":"outrage"}, {"id":"rdc_g99v2jp","responsibility":"unclear","reasoning":"mixed","policy":"unclear","emotion":"mixed"}, {"id":"rdc_g9asdt7","responsibility":"company","reasoning":"virtue","policy":"unclear","emotion":"outrage"}, {"id":"rdc_elcg4kc","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"indifference"} ]