Nice work, these seem like interesting and useful results! 

High level question/comment which might be totally off: one benefit of having a single, large, SAE neuron space that each token gets projected into is that features don't get in each other's way, except insofar as you're imposing sparsity. Like, your "I'm inside a parenthetical" and your "I'm attempting a coup" features will both activate in the SAE hidden layer, as long as they're in the top k features (for some sparsity). But introducing switch SAEs breaks that: if these two features are in different experts, only one of them will activate in the SAE hidden layer (based on whatever your gating learned). 

The obvious reply is "but look at the empirical results you fool! The switch SAEs are pretty good!" And that's fair. I weakly expect what is happening in your experiment is that similar but slightly specialized features are being learned by each expert (a testable hypothesis), and maybe you get enough of this redundancy that it's fine e.g,. the expert with "I'm inside a parenthetical" also has a "Words relevant to coups" feature and this is enough signal for coup detection in that expert. 

Again, maybe this worry is totally off or I'm misunderstanding something. 

Thanks for the addition, that all sounds about right to me!

Leaving Dangling Questions in your Critique is Bad Faith

Note: I’m trying to explain an argumentative move that I find annoying and sometimes make myself; this explanation isn’t very good, unfortunately. 

Example

Them: This effective altruism thing seems really fraught. How can you even compare two interventions that are so different from one another? 

Explanation of Example

I think the way the speaker poses the above question is not as a stepping stone for actually answering the question, it’s simply as a way to cast doubt on effective altruists. My response is basically, “wait, you’re just going to ask that question and then move on?! The answer really fucking matters! Lives are at stake! You are clearly so deeply unserious about the project of doing lots of good, such that you can pose these massively important questions and then spend less than 30 seconds trying to figure out the answer.” I think I might take these critics more seriously if they took themselves more seriously. 

Description of Dangling Questions

A common move I see people make when arguing or criticizing something is to pose a question that they think the original thing has answered incorrectly or is not trying sufficiently hard to answer. But then they kinda just stop there. The implicit argument is something like “The original thing didn’t answer this question sufficiently, and answering this question sufficiently is necessary for the original thing to be right.”

But importantly, the criticisms usually don’t actually argue that — they don’t argue for some alternative answer to the original questions, if they do they usually aren’t compelling, and they also don’t really try to argue that this question is so fundamental either. 

One issue with Dangling Questions is that they focus the subsequent conversation on a subtopic that may not be a crux for either party, and this probably makes the subsequent conversation less useful. 

Example

Me: I think LLMs might scale to AGI. 

Friend: I don’t think LLMs are actually doing planning, and that seems like a major bottleneck to them scaling to AGI. 

Me: What do you mean by planning? How would you know if LLMs were doing it? 

Friend: Uh…idk

Explanation of Example

I think I’m basically shifting the argumentative burden onto my friend when it falls on both of us. I don’t have a good definition of planning or a way to falsify whether LLMs can do it — and that’s a hole in my beliefs just as it is a hole in theirs. And sure, I’m somewhat interested in what they say in response, but I don’t expect them to actually give a satisfying answer here. I’m posing a question I have no intention of answering myself and implying it’s important for the overall claim of LLMs scaling to AGI (my friend said it was important for their beliefs, but I’m not sure it’s actually important for mine). That seems like a pretty epistemically lame thing to do. 

Traits of “Dangling Questions”

1. They are used in a way that implies the target thing is wrong vis a vis the original idea, but this argument is not made convincingly.
2. The author makes minimal effort to answer the question with an alternative. Usually they simply pose it. The author does not seem to care very much about having the correct answer to the question.
3. The author usually implies that this question is particularly important for the overall thing being criticized, but does not usually make this case.
4. These questions share a lot in common with the paradigm criticisms discussed in Criticism Of Criticism Of Criticism, but I think they are distinct in that they can be quite narrow.
5. One of the main things these questions seem to do is raise the reader’s uncertainty about the core thing being criticized, similar to the Just Asking Questions phenomenon. To me, Dangling Questions seem like a more intellectual version of Just Asking Questions — much more easily disguised as a good argument.

Here's another example, though it's imperfect.

Example

From an AI Snake Oil blog post:

Research on scaling laws shows that as we increase model size, training compute, and dataset size, language models get “better”. … But this is a complete misinterpretation of scaling laws. What exactly is a “better” model? Scaling laws only quantify the decrease in perplexity, that is, improvement in how well models can predict the next word in a sequence. Of course, perplexity is more or less irrelevant to end users — what matters is “emergent abilities”, that is, models’ tendency to acquire new capabilities as size increases.

Explanation of Example

The argument being implied is something like “scaling laws are only about perplexity, but perplexity is different from the metric we actually care about — how much? who knows? —, so you should ignore everything related to perplexity, also consider going on a philosophical side-quest to figure out what ‘better’ really means. We think ‘better’ is about emergent abilities, and because they’re emergent we can’t predict them so who knows if they will continue to appear as we scale up”. In this case, the authors have ventured an answer to their Dangling Question, “what is a ‘better’ model?“, they’ve said it’s one with more emergent capabilities than a previous model. This answer seems flat out wrong to me; acceptable answers include: downstream performance, self-reported usefulness to users, how much labor-time it could save when integrated in various people’s work, ability to automate 2022 job tasks, being more accurate on factual questions, and much more. I basically expect nobody to answer the question “what does it mean for one AI system to be better than another?” with “the second has more capabilities that were difficult to predict based on the performance of smaller models and seem to increase suddenly on a linear-performance, log-compute plot”.

Even given the answer “emergent abilities”, the authors fail to actually argue that we don’t have a scaling precedent for these. Again, I think the focus on emergent abilities is misdirected, so I’ll instead discuss the relationship between perplexity and downstream benchmark performance — I think this is fair game because this is a legitimate answer to the “what counts as ‘better’?” question and because of the original line “Scaling laws only quantify the decrease in perplexity, that is, improvement in how well models can predict the next word in a sequence”. The quoted thing is technically true but in this context highly misleading, because we can, in turn, draw clear relationships between perplexity and downstream benchmark performance; here are three recent papers which do so, here are even more studies that relate compute directly to downstream performance on non-perplexity metrics. Note that some of these are cited in the blog post. I will also note that this seems like one example of a failure I’ve seen a few times where people conflate “scaling laws” with what I would refer to as “scaling trends” where the scaling laws refer to specific equations for predicting various metrics based on model inputs such as # parameters and amount of data to predict perplexity, whereas scaling trends are the more general phenomenon we observe that scaling up just seems to work and in somewhat predictable ways; the scaling laws are useful for the predicting, but whether we have those specific equations or not has no effect on this trend we are observing, the equations just yield a bit more precision. Yes, scaling laws relating parameters and data to perplexity or training loss do not directly give you info about downstream performance, but we seem to be making decent progress on the (imo still not totally solved) problem of relating perplexity to downstream performance, and together these mean we have somewhat predictable scaling trends for metrics that do matter.

Example

Here’s another example from that blog post where the authors don’t literally pose a question, but they are still doing the Dangling Question thing in many ways. (context is referring to these posts):

Also, like many AI boosters, he conflates benchmark performance with real-world usefulness.

Explanation of Example

(Perhaps it would be better to respond to the linked AI Snake Oil piece, but that’s a year old and lacks lots of important evidence we have now). I view the move being made here as posing the question “but are benchmarks actually useful to real world impact?“, assuming the answer is no — or poorly arguing so in the linked piece — and going on about your day. It’s obviously the case that benchmarks are not the exact same as real world usefulness, but the question of how closely they’re related isn’t some magic black box of un-solvability! If the authors of this critique want to complain about the conflation between benchmark performance and real-world usefulness, they should actually bring the receipts showing that these are not related constructs and that relying on benchmarks would lead us astray. I think when you actually try that, you get an answer like: benchmark scores seem worse than user’s reported experience and than user’s reported usefulness in real world applications, but there is certainly a positive correlation here; we can explain some of the gap via techniques like few-shot prompting that are often used for benchmarks, a small amount via dataset contamination, and probably much of this gap comes from a validity gap where benchmarks are easy to assess but unrealistic, but thankfully we have user-based evaluations like LMSYS that show a solid correlation between benchmark scores and user experience, … (if I actually wanted to make the argument the authors were, I would be spending like >5 paragraphs on it and elaborating on all of the evidences mentioned above, including talking more about real world impacts, this is actually a difficult question and the above answer is demonstrative rather than exemplar)

Caveats and Potential Solutions

There is room for questions in critiques. Perfect need not be the enemy of good when making a critique. Dangling Questions are not always made in bad faith. 

Many of the people who pose Dangling Questions like this are not trying to act in bad faith. Sometimes they are just unserious about the overall question, and they don’t care much about getting to the right answer. Sometimes Dangling Questions are a response to being confused and not having tons of time to think through all the arguments, e.g., they’re a psychological response something like “a lot feels wrong about this, here are some questions that hint at what feels wrong to me, but I can’t clearly articulate it all because that’s hard and I’m not going to put in the effort”.

My guess at a mental move which could help here: when you find yourself posing a question in the context of an argument, ask whether you care about the answer, ask whether you should spend a few minutes trying to determine the answer, ask whether the answer to this question would shift your beliefs about the overall argument, ask whether the question puts undue burden on your interlocutor. 

If you’re thinking quickly and aren’t hoping to construct a super solid argument, it’s fine to have Dangling Questions, but if your goal is to convince others of your position, you should try to answer your key questions, and you should justify why they matter to the overall argument. 

Another example of me posing a Dangling Question in this:

What happens to OpenAI if GPT-5 or the ~5b training run isn't much better than GPT-4? Who would be willing to invest the money to continue? It seems like OpenAI either dissolves or gets acquired. 

Explanation of Example

(I’m not sure equating GPT-5 with a ~5b training run is right). In the above quote, I’m arguing against The Scaling Picture by asking whether anybody will keep investing money if we see only marginal gains after the next (public) compute jump. I think I spent very little time trying to answer this question, and that was lame (though acceptable given this was a Quick Take and not trying to be a strong argument). I think for an argument around this to actually go through, I should argue: without much larger dollar investments, The Scaling Picture won’t hold; those dollar investments are unlikely conditional on GPT-5 not being much better than GPT-4. I won’t try to argue these in depth, but I do think some compelling evidence is that OpenAI is rumored to be at ~$3.5 billion annualized revenue, and this plausibly justifies considerable investment even if the GPT-5 gain over this isn’t tremendous. 

I agree that repeated training will change the picture somewhat. One thing I find quite nice about the linked Epoch paper is that the range of tokens is an order of magnitude, and even though many people have ideas for getting more data (common things I hear include "use private platform data like messaging apps"), most of these don't change the picture because they don't move things more than an order of magnitude, and the scaling trends want more orders of magnitude, not merely 2x. 

Repeated data is the type of thing that plausibly adds an order of magnitude or maybe more. 

I sometimes want to point at a concept that I've started calling The Scaling Picture. While it's been discussed at length (e.g., here, here, here), I wanted to give a shot at writing a short version:

• The picture:
  • We see improving AI capabilities as we scale up compute, projecting the last few years of progress in LLMs forward might give us AGI (transformative economic/political/etc. impact similar to the industrial revolution; AI that is roughly human-level or better on almost all intellectual tasks) later this decade (note: the picture is not about specific capabilities so much as the general picture).
  • Relevant/important downstream capabilities improve as we scale up pre-training compute (size of model and amount of data), although for some metrics there are very sublinear returns — this is the current trend. Therefore, you can expect somewhat predictable capability gains in the next few years as we scale up spending (increase compute), and develop better algorithms / efficiencies.
  • AI capabilities in the deep learning era are the result of three inputs: data, compute, algorithms. Keeping algorithms the same, and scaling up the others, we get better performance — that's what scaling means. We can lump progress in data and algorithms together under the banner "algorithmic progress" (i.e., how much intelligence can you get per compute) and then to some extent we can differentiate the source of progress: algorithmic progress is primarily driven by human researchers, while compute progress is primarily driven by spending more money to buy/rent GPUs. (this may change in the future). In the last few years of AI history, we have seen massive gains in both of these areas: it's estimated that the efficiency of algorithms has improved about 3x/year, and the amount of compute used has increased 4.1x/year. These are ludicrous speeds relative to most things in the world. 
  • Edit to add: The below arguments are just supposed to be pointers toward longer argument one could make, but the one sentence version usually isn't compelling on its own.
• Arguments for:
  • Scaling laws (mathematically predictable relationship between pretraining compute and perplexity) have held for ~12 orders of magnitude already
  • We are moving though ‘orders of magnitude of compute’ quickly, so lots of probability mass should be soon (this argument is more involved, following from having uncertainty over orders of magnitude of compute that might be necessary for AGI, like the approach taken here; see here for discussion)
  • Once you get AIs that can speed up AI progress meaningfully, progress on algorithms could go much faster, e.g., by AIs automating the role of researchers at OpenAI. You also get compounding economic returns that allow compute to grow — AIs that can be used to make a bunch of money, and that money can be put into compute. It seems plausible that you can get to that level of AI capabilities in the next few orders of magnitude, e.g., GPT-5 or GPT-6. Automated researchers are crazy.
  • Moore’s law has held for a long time. Edit to add: I think a reasonable breakdown for the "compute" category mentioned above is "money spent" and "FLOP purchasable per dollar". While Moore's Law is technically about the density of transistors, the thing we likely care more about is FLOP/$, which follows similar trends. 
  • Many people at AGI companies think this picture is right, see e.g., this, this, this (can’t find an aggregation)
• Arguments against:
  • Might run out of data. There are estimated to be 100T-1000T internet tokens, we will likely hit this level in a couple years.
  • Might run out of money — we’ve seen ~$100m training runs, we’re likely at $100m-1b this year, tech R&D budgets are ~30B, governments could fund $1T. One way to avoid this 'running out of money' problem is if you get AIs that speed up algorithmic progress sufficiently.
  • Scaling up is a non-trivial engineering problem and it might cause slow downs due to e.g., GPU failure and difficulty parallelizing across thousands of GPUs
  • Revenue might just not be that big and investors might decide it's not worth the high costs
    • OTOH, automating jobs is a big deal if you can get it working
  • Marginal improvements (maybe) for huge increased costs; bad ROI. 
    • There are numerous other economics arguments against, mainly arguing that huge investments in AI will not be sustainable, see e.g., here
  • Maybe LLMs are missing some crucial thing
    • Not doing true generalisation to novel tasks in the ARC-AGI benchmark
    • Not able to learn on the fly — maybe long context windows or other improvements can help
    • Lack of embodiment might be an issue
  • This is much faster than many AI researchers are predicting
  • This runs counter to many methods of forecasting AI development
  • Will be energy intensive — might see political / social pressures to slow down. 
  • We might see slowdowns due to safety concerns.

Neat idea. I notice that this looks similar to dealing with many-shot jailbreaking:

For jailbreaking you are trying to learn the policy "Always imitate/generate-from a harmless assistant", here you are trying to learn "Always imitate safe human". In both, your model has some prior for outputting harmful next tokens, the context provides an update toward a higher probability of outputting harmful text (because of seeing previous examples of the assistant doing so, or because the previous generations came from an AI). And in both cases we would like some training technique that causes the model's posterior on harmful next tokens to be low. 

I'm not sure there's too much else of note about this similarity, but it seemed worth noting because maybe progress on one can help with the other. 

Cool! I'm not very familiar with the paper so I don't have direct feedback on the content — seems good. But I do think I would have preferred a section at the end with your commentary / critiques of the paper, also that's potentially a good place to try and connect the paper to ideas in AI safety. 

It looks like the example you gave pretty explicitly is using “compute” rather than “effective compute”. The point of having the “effective” part is to take into account non compute progress, such as using more optimal N/D ratios. I think in your example, the first two models would be at the same effective compute level, based on us predicting the same performance.

That said, I haven’t seen any detailed descriptions of how Anthropic is actually measuring/calculating effective compute (iirc they link to a couple papers and the main theme is that you can use training CE loss as a predictor).

Claude 3.5 Sonnet solves 64% of problems on an internal agentic coding evaluation, compared to 38% for Claude 3 Opus. Our evaluation tests a model’s ability to understand an open source codebase and implement a pull request, such as a bug fix or new feature, given a natural language description of the desired improvement.

...

While Claude 3.5 Sonnet represents an improvement in capabilities over our previously released Opus model, it does not trigger the 4x effective compute threshold at which we will run the full evaluation protocol described in our Responsible Scaling Policy (RSP).

Hmmm, maybe the 4x effective compute threshold is too large given that you're getting near doubling of agentic task performance (on what I think is an eval with particularly good validity) but not hitting the threshold. 

Or maybe at the very least you should make some falsifiable predictions that might cause you to change this threshold. e.g., "If we train a model that has downstream performance (on any of some DC evals) ≥10% higher than was predicted by our primary prediction metric, we will revisit our prediction model and evaluation threshold." 

It is unknown to me whether Sonnet 3.5's performance on this agentic coding evaluation was predicted in advance at Anthropic. It seems wild to me that you can double your performance on a high validity ARA-relevant evaluation without triggering the "must evaluate" threshold; I think evaluation should probably be required in that case, and therefore, if I had written the 4x threshold, I would be reducing it. But maybe those who wrote the threshold were totally game for these sorts of capability jumps? 

Can you say more about why you would want this to exist? Is it just that "do auto-interpretability well" is a close proxy for "model could be used to help with safety research"? Or are you also thinking about deception / sandbagging, or other considerations.