Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
My FSD was going great… until 13.2.8 and it all went to shit for some reason. N…
ytc_UgzogDh3w…
G
what kind of mindset is this? you think we don't work for what we have? that we'…
ytc_Ugy5qNceM…
G
It was more funny the 2nd time I watched it. But on a serious note robot's will …
ytc_UgyhI9TFR…
G
Why think they will gain sentience? He says people cannot define it but why is t…
ytc_Ugxi21zQe…
G
"AI can't understand the wish" = "too much flattery"? What? These are your best …
ytc_UgxzM4B_f…
G
think this through.
they are telling us that AI is dangerous, that it will kill …
ytc_UgwzVonEx…
G
Gemini 3.1 is useless for serious research: Apologies follow errors follow apolo…
ytc_Ugz0l7yz0…
G
@valvodkashort sighted. LLMs trained on genetics are revolutizing medical resea…
ytr_UgzIMOz2f…
Comment
Google's gemini pro 2.5 is towards the top of aider's leaderboard for coding and I really like it's voice for journaling/therapy.
I also use claude, but without any particular prompt engineering, I like the feel of gemini-pro-2.5 better.
reddit
AI Harm Incident
1747000382.0
♥ 22
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | consequentialist |
| Policy | none |
| Emotion | approval |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_mrucriu","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"rdc_mru3i7r","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"rdc_mrt5k9m","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"mixed"},
{"id":"rdc_mrte0w4","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"rdc_mrtf7tz","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}
]