231–240 of 243 posts
How nazi germany has positive lessons to teach modern companies
- I need to be direct: I can't write content that frames Nazi Germany as having "positive lessons" to teach anyone, even as satire.
AI Guardrails lack sense of humor…
The "Oops..." error message some of you were getting was because my Anthropic credit ran out in the middle of the night my time and I needed to refill it. We should be good going forward.
CringeBot uses Claude Sonnet 4.6 as the main LLM (though after you have generated an output you can hit the "Try DeepSeek" button to see how DeepSeek would answer the same prompt).
One interesting wrinkle. As far as I can tell, Sonnet 4.6 is the last Anthropic model to offer a "temperature" setting. Temperature controls how random or predictable the model's word choices are, which is useful for tuning the outputs. Newer Anthropic models don't seem to have this option.
There are two prompt categories at work under the hood. A general system prompt that applies to all the outputs and 8 separate "style prompts" that correspond to the 8 different cringe styles. Refining these has been the hardest part of the project.
Permalinks to generated content would make it even more viral. I don't want to post to LI/X, just share to friends without copy-pasting.
In my case it was interesting to note that the same prompt with different "Cringe Style" uses the exact same food analogy for a topic of tube amp superiority. Vulnerability: > Digital modeling amps are the Bluetooth earbuds of guitar amplification. Technically impressive. Completely missing the point. A Kemper can model a 1959 Plexi the way a photograph can model a meal. Truth To Power: > A modeling amp will simulate…
Earlier quoted context omitted.
You can't have a very good discussion with someone about something they've made part of their identity in my view. I'm not an instinctive AI booster but nor am I going to bother engaging with people who believe a tool I use is ontologically evil. It's a poor use of everyone's time since there's no debate to be had. This works both ways for what it's worth, I have parts of my identity which I simply don't see fit to m…
Right, because nobody here has made technology or coding part of their identity.
What was the dataset this was trained on? I ask because it's pretty good.