Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
To some extent, I think the answer is that your tool sucks. A human would have the same issues here. Write a tool with a more intuitive API, as well as a stricter one, and both AI and human will have an easier time with it. > The API tried to parse it as an integer, got NaN, and... just returned empty results instead of erroring. like, c'mon, seriously? Either fix the API, if you own it, or fix the tool, if you own it, or use something better-written, if you can, or write your own validation wrapper that unbraindamages the crappy-ass API and tool that you're forced to use.
reddit Viral AI Reaction 1776989493.0 ♥ 2
Coding Result
DimensionValue
Responsibilitydeveloper
Reasoningconsequentialist
Policynone
Emotionoutrage
Coded at2026-04-25T08:33:43.502452
Raw LLM Response
[{"id":"rdc_ohprofz","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"}, {"id":"rdc_ohp4pb1","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"unclear"}, {"id":"rdc_ohs6exv","responsibility":"developer","reasoning":"deontological","policy":"unclear","emotion":"approval"}, {"id":"rdc_ohp7m7c","responsibility":"user","reasoning":"virtue","policy":"unclear","emotion":"outrage"}, {"id":"rdc_ohx5qem","responsibility":"developer","reasoning":"consequentialist","policy":"none","emotion":"outrage"}]