site banner
PaperclipPerfector  29d ago (text post)   34853 thread views

Culture War Roundup for the week of March 3, 2025

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

7
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

Is this an artifact of the LLM having no side-effects while processing outside of the explicit textual output?

Pretty much, which is also why dumb hacks like letting the LLM explicitly write out its chain of thought before the actual response have been so effective.

if you tell them to process it explicitly but include that in a sidebar like the block, would they have an easier time keeping the anime chicks where they oughta be?

Yes, actually. Instructing the LLM to insert basic statboxes/infoblocks after every response (and regexing them out via the frontend after to avoid shitting up the context) so it has the basic "states" at the tip of its proverbial tongue at all times has been the meta for a pretty long time now. Something like this:

End your response with "infoblock" to keep track of the scene. You don’t have to mention those details in your actual response unless they are important. Use the format below:

<infoblock>

Location: (Current location)
Positions: (All the characters' and {{user}}'s current positions relative to each other.)
Outfits: (For each character their current clothing and underwear.)

</infoblock>

I can't say exactly how well it works, but accounts of e.g. sizefags people who require accurate dimensions and/or relative positions for their scenarios say it's definitely better than placebo.

I wonder why the knowledge base approach seems to have fallen flat.

IMO exactly because of this:

re-reading the design, it looks like the knowledge base isn't so much like a binder of notes as it is a single post-it note stuck to the screen - Claude doesn't query it deliberately, it apparently gets the entire contents of it shoved into the prompt.

So it's basically just another part of the prompt without(?) any distinguishing features, and Claude's attention growing more scattered as the context limit grows is a known problem. In my experience amnesia/retardation kicks in as soon as 20k tokens. Granted, I don't know how their prompt builder looks like, or what context limit they use - although I suspect money isn't as much of a problem for an Anthropic researcher as it is for someone like me, so I expect it to be pretty high given the 200k upper bound.

It's hard to fit anything very detailed in there and means that Claude can't get a new set of "notes" for whatever area/task it's currently attempting to handle.

Yeah, I hope nobody tells them about worldinfo or something. I'm still convinced the median /g/oon still has the median researcher's ass handily beat wrt "prompt engineering". Arguably this is a testament to how powerful a tool SillyTavern is, but afaik every feature has been initially conceived and pitched by the community anyway.

What if you gave it something open-ended like "Pokemon is a game that children play to explore, befriend Pokemon, and win tough battles. Play this game the way it was meant to be played"?

I expect that wouldn't change much, arguably it'd make it get lost even more, at least now it seems to have a fairly clear objective in mind (beat children defeat gyms), which it can even translate into lower-level "tasks" like navigating routes.

Besides, the minimal prompting seems to be the point; from my understanding the dev is unwilling to hold Claude's hand any more than necessary and he wishes to see how it holds up on its own, even if it takes it days to get out of every stupid loop he gets stuck in. I wish I had unlimited credit think it's dumb, even with crutches to streamline progression and break loops this would still be pretty interesting to watch, but oh well.

edit: I tuned back into the stream and immediately witnessed a page-long reply thread starting with someone asserting Claude's preferred pronouns and gender, based on them literally just asking it. I thus retract my observation about this being the sanest Twitch chat. At least there's content now.

Outfits: (For each character their current clothing and underwear.)

and underwear

This site needs emojis for shit like this. Text doesn't do it justice.

Hobbyists have no shame.

Yeah, I hope nobody tells them about worldinfo or something. I'm still convinced the median /g/oon still has the median researcher's ass handily beat wrt "prompt engineering". Arguably this is a testament to how powerful a tool SillyTavern is, but afaik every feature has been initially conceived and pitched by the community anyway.

It looks like they actually implemented something similar to what I was talking about earlier - I watched Claude sit and churn for a while after it left Pewter, moving all information about that city into long-term memory (with explicit tags!) and clearing up local information. It's now back in Mt Moon, so we'll see whether this has made it more effective at navigation. What it's definitely doing is taking meaningful and extended "clock cycles" to manage - so this kind of improvement is definitely not free or cheap at present implementation/with present models.

Very cool tool from WorldInfo. I like the idea of bringing word definitions into context transparently based on the prompt.

I expect that wouldn't change much, arguably it'd make it get lost even more, at least now it seems to have a fairly clear objective in mind (beat children defeat gyms), which it can even translate into lower-level "tasks" like navigating routes.

Besides, the minimal prompting seems to be the point; from my understanding the dev is unwilling to hold Claude's hand any more than necessary and he wishes to see how it holds up on its own, even if it takes it days to get out of every stupid loop he gets stuck in. I wish I had unlimited credit think it's dumb, even with crutches to streamline progression and break loops this would still be pretty interesting to watch, but oh well.

Yeah, watching the money burn is a little eye-watering, but I appreciate how seriously the guy seems to take it. He seems to have known from the start that it wasn't going to be a magical success, but wants to see what it takes to get it working. I'm here for that. My only complaints are: there's no summary of where it's been/what it's done (so I can't track progress easily) and there's no export of the knowledge base over time to show what it's learned. Getting to read the knowledge base would be incredibly interesting.

Hobbyists have no shame.

I've seen things... *cough* you people wouldn't believe...

I watched Claude sit and churn for a while after it left Pewter, moving all information about that city into long-term memory (with explicit tags!) and clearing up local information. It's now back in Mt Moon, so we'll see whether this has made it more effective at navigation.

It seems to still firmly remain in Mt Moon, sadly, but all the <thinking> about a certain "loop prevention log" seems hopeful, if anything it at least seems to retain said log in context and refer to it quite often. Will check back in the morning, I'm getting invested.

watching the money burn is a little eye-watering

This is putting it mildly tbh, this guy is quite literally living my best life, even just worldinfo on its own has gotten me a lot of fun exercises in tard wrangling. I would kill a man in cold blood to sit and fuck (around) with chatbots all day as, or at least in parallel to, my day job without constantly fretting over my OR credits. Not that goons haven't burned even more money in aggregate at this point (most of which isn't theirs either), but... fuck, man. It should've been me! Not him! How do AI safetyists get to take away my toys and keep the cool toys to themselves?!

there's no summary of where it's been/what it's done (so I can't track progress easily) and there's no export of the knowledge base over time to show what it's learned.

Agreed, would've been interesting to see. I don't know if he's sitting on the "tech" as it were and doesn't want to reveal too much but I can fairly confidently say he doesn't really have anything special yet, the setup definitely doesn't seem Neuro-sama-tier complicated.

I've seen things... cough you people wouldn't believe...

LOL

fun exercises in tard wrangling

Is that a full TTRPG campaign set up for an LLM to execute on? How well does that work, and how extensive can it get? Is there some kind of external scaffolding for selecting things like random events, or does it have the capacity to toss all the events together in memory and then select? How long does it go before it totally loses the plot? (Maybe not an appropriate Culture War Roundup topic, but w/e.)

I considered playing around with some of that stuff a while back but I just couldn't justify the costs to myself. It's interesting, but so are a lot of other things that are WAY cheaper (and I'm at this point morally opposed to interfacing with large companies if I can at all help it). If cost-to-performance comes massively down over the next decade, maybe I'll try a local model off a reasonably priced GPU. Otherwise, idk, it's cool hearing stories.

Is that a full TTRPG campaign set up for an LLM to execute on?

Not really (though people have been trying, with some success), it's more a free-form adventure(ish) thing, mostly my half-baked experiments with making a properly-ordered lorebook. My code-fu is... not great and my access expensive capricious and intermittent, so I haven't tried building any external scaffolding around the LLM, but even just built-in ST tools have been very fun to fuck with.

How does it work, and how extensive it can get?

At its core worldinfo is dead simple, it's something like a dictionary with every entry having a value (text to be added to the outgoing prompt), a key (keyword(s) which will trigger the insertion when detected in the context), and a few trigger parameters - e.g. you can set it to automatically add [PLEASE JUST GO TO THE FUCKING LADDER AT (X,Y)] to the bottom of context whenever the last X messages contain the string "Mt. Moon". This lets you insert things into context or take them out conditionally, for example hide info about secondary characters while they're not actively present or being talked about.

The basic functionality already allowed for some clever things - e.g with judicious use of the "recursive scanning" checkbox you could make cascading triggers that dump several entries into context off one triggering string, and if that string is something trivial like "." that's always in context, using the probability field to assign it a % chance of actually firing plausibly imitates "random encounters" (names in {{random}}ized enemy lineups are themselves keywords that pull up the corresponding entry).

Over time it was overgrown with a zillion knobs and dials, as seen in the screenshots, allowing for a lot of customization - you can now specify scan depth (how many latest messages are scanned for keywords), case sensitivity, simple logical operators (if "Mt. Moon" AND "B1F" trigger entry X, if "Mt. Moon" AND "B2F" trigger entry Y), locally disable recursive scan so a given entry doesn't trigger others even if it contains their keywords, delay the trigger so the entry doesn't show up until X messages in, stick it around for X messages after the fact, give it a X message cooldown, group several entries together so that only one can be active at any time (even the weights are customizable), etc. etc. etc.

Sorry for suddenly nerding out but I'm sure you're seeing potential here, including for our ClaudeFan's knowledge base.

(Further autistic details will probably go in DMs, this chain is long enough already.)

How long does it go before it totally loses the plot?

Arguably the decline starts as soon as the initial message gets pushed out of context, leaving the LLM somewhat "unmoored" from the start of the scenario, moreso with gradually losing earlier happenings as the context window drifts. Summarization every so often to preserve key developments can alleviate the worst of it, although I've found automatic summarization isn't very good since the LLM's views of what's important may differ from yours.

Otherwise the threshold depends on the system prompts, the model (e.g. Opus is infamous for its "artistic license" to deviate from instructions/make shit up and write the hottest smut known to mankind) and the given scenario. My personal opinion is that ~20k tokens, of which no more than 40% is worldinfo so that enough actual chat history remains to have a clue what's going on, is about the sweet spot.

I'm at this point morally opposed to interfacing with large companies if I can at all help it

At this point I'm accustomed to existence behind seven proxies, so I am addicted and coping less concerned, hoping that soulless capitalism can actually work in my favor for once and disregard the silly notions of AI ethics et cetera so I can pay a third party to proompt in peace. It's much better to go for local models in this regard, for several reasons, but I've been unlucky to try the corpo-grade crack first so that's what I'm stuck with.

Interesting - I do think there's a pretty major intermediate step of using static analysis-type processing to control the excesses of AI, and worldinfo is a plausible first step. But that lack of real memory just keeps coming back and kicking any in-depth efforts right in the teeth.

Probably a good place to stop this particular conversation, but not before Claude updates. It's still at Mt Moon, but there have been some interesting developments (timeline:

  1. It almost got out of Mt Moon! Unfortunately, then it died.
  2. After dying, it managed to convince the oversight layer ("Critique Claude") that there's a secret passage to Cerulean City that skips going through Mt Moon, and all it needs to do is find it. Now both AIs are convinced that this is the one way forward.
  3. But the only way to get to that secret passage is by dying again. So it's trying to black out on wild Zubats in Mt Moon to reach Terabithia or something.
  4. Now that it's converted to Gnosticism, it doesn't need any knowledge of the world of the Demiurge. So it's been scrubbing data out of its memory banks in favor of the holy book "black_out_strategy".
  5. At some point it managed to crash its internal tooling. This hasn't helped matters.

This is absolute peak comedy. Somehow the AI has managed to go completely nuts, seduce its parole officer, and start a death cult. We're not even past the linear part of the game yet!

Hobbyists have no shame.

How very lewd, to have the characters have underwear instead of starting naked like every self-respecting anatomy study.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats