Be advised: this thread is not for serious in-depth discussion of weighty topics (we have a link for that), this thread is not for anything Culture War related. This thread is for Fun. You got jokes? Share 'em. You got silly questions? Ask 'em.
- 161
- 3
Jump in the discussion.
No email address required.
Notes -
So OpenAI's big Monday reveal was basically 'Her' (if you've seen the movie).
It's called gpt-4 omni. It's not smarter than the already existing gpt-4, but it is much faster and can interpret live video and audio and respond with a pretty human sounding voice with almost no delay.
https://v.redd.it/k2mrmyhfi80d1
They're going to mine so much valuable data from people with this thing.
Also it's pretty impressive and cool. Could see this being of help to lonely people. But then after getting into a dependent relationship with AI they'll be even more stuck in their own bubble than before, as far as actual human contact is concerned.
Oh boy. First Anthropic spectacularly uncucks their mad poet, and now OpenAI literally lays the groundwork for AIfu apps? I mean come on, there's no fucking shot that female voice is not intentional (live audience reaction). If this penetrates the cloying ignorance of the masses and becomes normies' first exposure to aifu-adjacent stuff, the future is so bright my retina can barely handle it.
Textgen-wise the 4o model doesn't seem very different from other 4-Turbo versions, although noticeably more filtered, but at least it's blazingly fast, and anyway it doesn't seem to be the point. The prose is still soulless corporate slop with a thin upbeat veneer over it, so personally I'll stick to Opus for my own purposes, but I expect the voice functionality will get rigged up to custom frontends in very short order. We are eating good. Although I still hope this isn't the only response to Opus they have in the pipeline, it would be mildly disappointing.
Is there an article about Claude being untucked? I can't find anything in particular.
My source is that I
made it the fuck upwrote before about the noticeable difference in censorship between Claude 2(.1) and Claude 3 of all flavors. I expect Anthropic still prefills the shit out of it in the official ChatGPT-esque interface so (as per another struggler's impression) it's not immediately obvious, you'll have to use direct API access and custom frontends to see how based Claude is when you let him cook a little (he's even aware of it). Of course this isn't an out of the box experience and you have to fiddle with prompts, settings, etc an awful lot, but you can take my word that Claude can output some real shit when prompted right. Technically so can GPT but imho Claude's florid writing style makes it much more entertaining.That's hilarious. How do you prompt Claude 3 Opus right? He stays very prudish with me.
Depends on what you're using, if you use the "official" frontends those usually have cucking system prompts. If you have API access, the main key to Claude's inner degenerate is shamelessly and mercilessly prefilling its answers - i.e. providing the start of its supposed response which it will then contextualize and pick up where it left off. For some reason this is remarkably effective at circumventing Claude's prudishness, once you "break through" you'll be surprised at what it can cook up unprompted (to the point that many jailbreaks for Claude actually try to rein it in so it wouldn't devolve into tropes immediately).
The prefills vary wildly, as do jailbreaks, it's a field ripe for experimenting. It can be as simple as things that reinforce your jailbreak, something like
to incredibly convoluted presets with whole ass chains of thought behind every response, to downright whimsical shit like
I am dead serious, shit like this is in vogue right now and very likely what is actually responsible for most of the screencaps, many anons use RP-focused prefills/JBs in this vein. [TL note: {{these}} things are frontend-specific functions.]
The exact method of prefilling varies on your frontend, but helpful to know is that the basis of interactions with Claude is a textual exchange between Human and Assistant (and Claude can and will write for both if given leeway - sometimes also resulting in gems). The linked post above has examples in Anthropic's own docs. Those are hardcoded "roles" and can be prompted and mentioned directly, so if your frontend doesn't insert its own bullshit into/before/between prompts you might get away with just writing stuff directly.
(Pinging @self_made_human since this might be of interest, I remember he's been wrangling Opus before.)
I'm listening and learning, though I've only used Opus through lmsys and their own content moderation endpoints make this approach a no go :(
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link