All of CallumMcDougall's Comments + Replies

Sorry I didn't get to this message earlier, glad you liked the post though! The answer is that attention heads can have multiple different functions - the simplest way is to store things entirely orthogonally so they lie in fully independent subspsaces, but even this isn't necessary because it seems like transformers take advantage of superposition to represent multiple concepts at once, more so than they have dimensions.

Oh, interesting, wasn't aware of this bug. I guess this is probably fine since most people replicating it will be pulling it rather than copying and pasting it into their IDE. Also this comment thread is now here for anyone who might also get confused. Thanks for clarifying!

+1, thanks for sharing! I think there's a formatting error in the notebook, where the tags like <OUTPUT> were all removed and replaced with empty strings (e.g. see attached photo). We've recently made the ARENA evals material public, and we've got a working replication there which I think has the tags in the right place (section 2  of 3 on the page linked here)

Amazing post! Forgot to do this for a while, but here's a linked diagram explaining how I think about feature absorption, hopefully ppl find it helpful!

I don't know of specific examples, but this is the image I have in my head when thinking about why untied weights are more free than tied weights: 

I think more generally this is why I think studying SAEs in the TMS setup can be a bit challenging, because there's often too much symmetry and not enough complexity for untied weights to be useful, meaning just forcing your weights to be tied can fix a lot of problems! (We include it in ARENA mostly for illustration of key concepts, not because it gets you many super informative results). But I'm keen for ... (read more)

Oh yeah this is great, thanks! For people reading this, I'll highlight SLT + developmental interp + mamba as areas which I think are large enough to have specific exercise sections but currently don't

Thanks!! Really appreciate it

Thanks so much! (-:

Thanks so much, really glad to hear it's been helpful!

Thanks, really appreciate this (and the advice for later posts!)

Yep, definitely! If you're using MSE loss then it's got a pretty straightforward to use backprop to see how importance relates to the loss function. Also if you're interested, I think Redwood's paper on capacity (which is the same as what Anthropic calls dimensionality) look at derivative of loss wrt the capacity assigned to a given feature

Thanks (-:

Good question! In the first batch of exercises (replicating toy models of interp), we play around with different importances. There are some interesting findings here (e.g. when you decrease sparsity to the point where you no longer represent all features, it's usually the lower-importance features which collapse first). I chose not to have the SAE exercises use varying importance, although it would be interesting to play around with this and see what you get!

As for what importance represents, it's basically a proxy for "how much a certain feature reduces ... (read more)

Winner = highest-quality solution over the time period of a month (solutions get posted at the start of the next month, along with a new problem).

Note that we're slightly de-emphasising the competition side now that there are occasional hints which get dropped during the month in the Slack group. I'll still credit the best solution in the Slack group & next LW post, but the choice to drop hints was to make the problem more accessible and hopefully increase the overall reach of this series.

Thanks for the latter point, glad you got that impression!

These are super valid concerns, and it's true that there's lots of information we won't have for a while. That said, we also have positive evidence from the first iteration of ARENA (which is about a year old now). There were only 5 full-time participants, and they've all gone on to do stuff I'm excited about, including the following (note that obviously some of these 5 have done more than one of the stuff on this list):

• internships at CHAI,
• working with Owain Evans (including some recent papers),&nbs

Upvoted overall karma, because I think this is a valuable point to bring up and we could have done a better job discussing it in this post.

To be clear, contribution to capabilities research is a very important consideration for us, and apologies if we didn't address this comprehensively here. A few notes to this effect:

• We selected strongly on prior alignment familiarity (particularly during the screening & interview process), and advertised it in locations we expected to mainly attract people who had prior alignment familiarity
• We encouraged interaction

Hi, sorry for the late response! The layer 0 attention head should have query at position 1, and value at position 0 (same as key). Which diagram are you referring to?

(context: I ran the most recent iteration of ARENA, and after this I joined Neel Nanda's mech interp stream in SERI MATS)

Registering a strong pushback to the comment on ARENA. The primary purpose of capstone projects isn't to turn people into AI safety technical researchers or to produce impressive capstones, it's to give people engineering skills & experience working on group projects. The initial idea was not to even push for things that were safety-specific (much like Redwood's recommendations - all of the suggested MLAB2 capstones were either mech ... (read more)

Yep, the occlusion effect is pretty large for colored images, that's why I use a layering system (e.g. 20% of all white threads, then 20% of all blue, then 20% of black, and cycle through). I go in reverse order, so the ones found first by the algorithm are the last ones to be placed. I also put black on top and white lowest down, cause white on top looks super jarring. The effect is that the In the colab you can play around with the order of the threads. If you reverse the order then the image looks really bad. You can also create gifs of the image formin... (read more)

Thanks so much for this comment, I really appreciate it! Glad it was helpful for you 🙂

Thanks, really appreciate it!

Yep that's right, thanks! Corrected.

huh interesting, I wasn't aware of this, thanks for sending it!

Thanks for the suggestion! I've edited the first diagram to clarify things, is this what you had in mind?

The first week of WMLB / MLAB maps quite closely onto the first week of ARENA, with a few exceptions (ARENA includes PyTorch Lightning, plus some more meta stuff like typechecking, VSCode testing and debugging, using GPT in your workflow, etc). I'd say that starting some way through the second week would probably be most appropriate. If you didn't want to repeat stuff on training / sampling from transformers, the mech interp material would start on Wednesday of the second week.

Resolved by private message, but I'm just mentioning this here for others who might be reading this - we didn't have confirmation emails set up, but we expect to send out coding assessments to applicants tomorrow (Monday 24th April). For people who apply after this point, we'll generally try to send out coding assessments no later than 24 hours after your application.

Yeah, I think this would be possible. In theory, you could do something like:

• Study relevant parts of the week 0 material before the program starts (we might end up creating a virtual group to accommodate this, which also contains people who either don't get an offer or can't attend but still want to study the material.)
• Join at the start of the 3rd week - at that point there will be 3 days left of the transformers chapter (which is 8 days long and has 4 days of core content), so you could study (most of) the core content and then transition to RL with the r

Awesome, really glad to hear it was helpful, thanks for commenting!

Yep, fixed, thanks!

Or "prompting" ? Seems short and memorable, not used in many other contexts so its meaning would become clear, and it fits in with other technical terms that people are currently using in news articles, e.g. "prompt engineering". (Admittedly though, it might be a bit premature to guess what language people will use!)

This is awesome, I love it! Thanks for sharing (-:

Thank you :-)

Thanks, really appreciate it!

I think some of the responses here do a pretty good job of this. It's not really what I intended to go into with my post since I was trying to keep it brief (although I agree this seems like it would be useful).

And yeah, despite a whole 16 lecture course on convex opti I still don't really get Bregman either, I skipped the exam questions on it 😆

Oh yeah, I hadn't considered that one. I think it's interesting, but the intuitions are better in the opposite direction, i.e. you can build on good intuitions for $D_{KL}$ to better understand MI. I'm not sure if you can easily get intuitions to point in the other direction (i.e. from MI to $D_{KL}$), because this particular expression has MI as an expectation over $D_{KL}$, rather than the other way around. E.g. I don't think this expression illuminates the nonsymmetry of $D_{KL}$.

The way it's written here seems more illuminating (not sure if that's... (read more)

Oh yeah, I really like this one, thanks! The intuition here is again that a monomodal distribution is a bad model for a bimodal one because it misses out on an entire class of events, but the other way around is much less bad because there's no large class of events that happen in reality but that your model fails to represent.

For people reading here, this post discusses this idea in more detail. The image to have in mind is this one:

Love that this exists! Looks like the material here will make great jumping off points when learning more about any of these orgs, or discussing them with others

Thanks Nihalm, also I wasn't aware of it being free! CraigMichael maybe you didn't find it cause it's under "Rationality: From AI to Zombies" not "Sequences"?

The narration is pretty good imo, although one disadvantage is it's a pain to navigate to specific posts cause they aren't titled (it's the whole thing, not the highlights).

Yep those were both typos, fixed now, thanks!

I think the key point here is that we're applying a linear transformation to move from neuron space into feature space. Sometimes neurons and features do coincide and you can actually attribute particular concepts to neurons, but unless the neurons are a privileged basis there's no reason to expect this in general. We're taking the definition of feature here as a linear combination of neurons which represents some particular important and meaningful (and hopefully human-comprehensible) concept.

I guess another point here is that we won't know how different (for example) our results when sampling from the training distribution will be from our results if we just run the network on random noise and then intervene on neurons; this would be an interesting thing to experimentally test. If they're very similar, this neatly sidesteps the problem of deciding which one is more "natural", and if they're very different then that's also interesting

Yeah I think the key point here more generally (I might be getting this wrong) is that C represents some partial state of knowledge about X, i.e. macro rather than micro-state knowledge. In other words it's a (non-bijective) function of X. That's why (b) is true, and the equation holds.

Yeah I think this is Evan's view. This is from his research agenda (I'm guessing you might have already seen this given your comment but I'll add it here for reference anyway in case others are interested)

I suspect we can in fact design transparency metrics that are robust to Goodharting when the only optimization pressure being applied to them is coming from SGD, but cease to be robust if the model itself starts actively trying to trick them.

And I think his view on deception through inner optimisation pressure is that this is something we'll basically be... (read more)

Okay I see, yep that makes sense to me (-: