4as

I tried using Base64 but quality of the responses drops significantly. It's like your communicating through Google translate. Go ahead and try it for yourself: ask ChatGPT to communicate in...

Let's focus on solving this one step at a time. Once there is sure-fire workaround for moderation checks, we can deal with tokens count. Here are things I tested that...

Same exact results. I'm guessing that line breaks and similar tricks work for short prompts, but as soon as you provide something longer the moderation catches on. 

Interestingly enough the jailbreak I'm using doesn't trigger the moderation, even though it's very explicit. Perhaps because it doesn't ask ChatGPT for anything except a confirmation that understands its role....

Because you're not generating content that breaks guidelines. Apply a jailbreak and ask it to generate an incest story or something similar, and then tell me if it works.

You could edit the script and delete `demod_div.appendChild(demod_status);` from it. You can edit scripts by going to Tampermonkey options (from Addons/Extensions in your browser) and then navigating to Installed Scripts....

Beside moderation ChatGPT also has a very primitive builtin filter for predefined words. It simply scans the text and looks for words that are in its list. If it finds...

I'm actually unable to find the list. The chat code is so obfuscated I can't really dig through it and get the full list. So far I knew about some...

Could you try installing this version and see if it works: http://4as.pl/other/demod.user.js It has all anti-lag methods disabled.

If you're talking about the 4.4 version I released on the main page, then it's understandable since I didn't do anything related to problems described in this issue. Instead, I...