Sam FM likes this.
Happy New AXRP!
Yet another in the Alignment Workshop series.
AI researchers often complain about the poor coverage of their work in the news media. But why is this happening, and how can it be fixed? In this episode, I speak with Shakeel Hashim about the resource constraints facing AI journalism, the disconnect between journalists' and AI researchers' views on transformative AI, and efforts to improve the state of AI journalism, such as Tarbell and Shakeel's newsletter, Transformer.
Ben Weinstein-Raun likes this.
Ben Weinstein-Raun likes this.
Ben Weinstein-Raun likes this.
like this
like this
Daniel Filan likes this.
Ben Weinstein-Raun likes this.
A question bopping around my mind: are there things like making AXRP or being a MATS RM that I could do instead of those things that would be more valuable? Possible answers:
- just do research that matters
- project manager at a place that does research that matters
- be more directly a competitor to Zvi
- team up with Lawrence Chan and write stuff about various alignment schemes
I think a bottleneck I feel is being unsure about what things are valuable in the info environment, where I think I'm best placed to do stuff.
like this
Ben Weinstein-Raun likes this.
Ben Millwood likes this.
Solstice notes
- I like that the celebration took place on (or adjacent to) the actual solstice
- I broadly thought this year's was worse than last year's, altho it had its charms
- I liked "Humankind as a sailor" - tricky to pick up but rewarding once you did
- Just because a song takes place in Australia, I don't think it thereby glorifies the negative aspects of colonialism.
- The darkness speech was touching this year
- I feel like a lot of the time the speaker would say something I straightforwardly agreed with in the way I would say it and everyone would laugh.
- It was funny when Ozy said her favourite website was Our World in Data and Scott sang the praises of Dustin Moskowitz while I was sitting next to Oli
- I think "the world is awful" is wrong, and not established by there being awful things in the world.
like this
dynomight.net/arguments-3/
Things to argue about over the holidays instead of politics III
report back on how it goesdynomight (DYNOMIGHT)
Daniel Filan likes this.
Daniel Filan reshared this.
Ben Weinstein-Raun likes this.
How much nesting can we do in English verb tenses, and what controls that? For an example of what I mean, I can say:
- I eat
- I will eat
- I will have been eating
- I will have been going to eat
But I don't think we can say "I will have been going to have eaten".
Ben Weinstein-Raun likes this.
One possibility: basically it goes as far as it makes sense to add extra timing information. But this only works if you disagree about your last positive example, which I personally don't actually think I've ever heard used.
Like, imagine a timeline. "I eat" describes a period of time encompassing now. "I will eat" describes a period of time in the future. "I will have eaten" describes two times; one in the future and one in the past of that future. "I will have been going to eat" describes a time in the future, a time in the past of that future, and a time in the future of that past of the first future. But in some sense this collapses back to the semantic content of "I will eat", and so my guess is that it's basically never used.
Daniel Filan likes this.
Ben Weinstein-Raun likes this.
I think what I mean is that additional times around the loop aren't really adding any extra information, because they introduce new reference points along the timeline that typically don't connect to anything else.
Like, there's some implicit time T that I'm trying to locate with a given statement, and there's an additional time Now that I get from just being in the present.
It makes sense to be like "Some time between Now and [implicitly / contextually defined] T, X will happen", and this is ~ the two-level wrapping. But if you say "Some time between Now and [newly introduced / 'bound' / 'scoped-to-this-statement'] T1, it will be the case that X happened after [implicit / 'free' / contextual] T2", T1 is kind of irrelevant, since it's introduced and used only within the statement.
In principle I guess you could have extra context that disambiguates, but I think it's also kinda relevant that verbs tend to have a subject, a direct object, and up to one indirect object, and typically not more than that.
Daniel Filan likes this.
Daniel Filan likes this.
Ben Weinstein-Raun likes this.
Ben Weinstein-Raun likes this.
MOAR AXRP
This time with Erik Jenner, on a paper he's presenting at NeurIPS tomorrow - check it out if you're there!
Lots of people in the AI safety space worry about models being able to make deliberate, multi-step plans. But can we already see this in existing neural nets? In this episode, I talk with Erik Jenner about his work looking at internal look-ahead within chess-playing neural networks.
Ben Weinstein-Raun likes this.
Jeroen Henneman, The Long Way Home
From: https://x.com/opancaro/status/186529216161008481
Wilhelm Kranz
From: https://x.com/0zmnds/status/1865291905249980735
#art
Ben Weinstein-Raun likes this.
Visual Arts Feed reshared this.
Ben Weinstein-Raun likes this.
New AXRP! With Evan Hubinger!
This time I won't retract it, I swear!
The 'model organisms of misalignment' line of research creates AI models that exhibit various types of misalignment, and studies them to try to understand how the misalignment occurs and whether it can be somehow removed. In this episode, Evan Hubinger talks about two papers he's worked on at Anthropic under this agenda: "Sleeper Agents" and "Sycophancy to Subterfuge".
Ben Weinstein-Raun likes this.
like this
Ben Weinstein-Raun likes this.
like this
EA: run by 1 or 2 extremely powerful guys who sometimes turn out to be deranged and corrupt. A woman called Julia is also involved.
like this
Amber Dawn likes this.
Latin practice day 7
These aren't very inspired but:
I. Cūr quaeque littera Graeca pulchrior est quam quaeque littera Latīna?
II. Sī linguam Latīnam scīre vult, quotiēs quamque litteram Latīnam scrībere necesse est?
III. Vōlōne ā magistrō laudārī?
IV. In Capitulō XVI, quia Dominus Iēsus tempestātem facit apud navem Lydiae? Lydia ā Deō dīligiturne?
V. Num medicus labōrans vērē sanat hominēs aegrōs?
VI. Num parēntēs laudant magister discipulōs verberāntem?
VII. Suntne bēstiolae industriorēs quam apēs? Quid facit illae?
VIII. Quia dea est pulcherrima?
IX. Hōdiē, quae bonae rēs daminī ā deī?
Hōdiē sum in domō parentum matris mea, in Arizonā. In hāc domō, saepe dormō in lectō parvō in cubiculō parvō, sed hōdiē habeō magnum cubiculum ac magnum lectum. Cēnābam cum parentibus matris meus, et cum amīcīs suīs. Aliī hominēs ēdēbant magnam avem, sed ego edēbam botulōs quī ex holeribus fīunt, nam Pythagoricus sum. Cōnspiciēbāmus pēs-pilam (harpastum? calcifollem? I guess Vicipaedia uses "Harpastum") - Leōnēs Detroitī, quī amantur ā parentēs matris meus, vincēbant contra Ursōs Sicāgoensis!
(I only know the imperfect past tense, forgive me)
Every country in the world belongs to America
Shouldn't the US buy the Vatican?
- they're rapidly going bankrupt and could use the money
- Trump would go for it
- the US is the new Rome
- would bring the US tons of geopolitical power
- new place to station US troops without any restrictions
- probably will ensure all Americans go to heaven
- zero downsides
Am I missing something?????
New episode with Jesse Hoogland!
Another short one, I'm afraid.
You may have heard of singular learning theory, and its "local learning coefficient", or LLC - but have you heard of the refined LLC? In this episode, I chat with Jesse Hoogland about his work on SLT, and using the refined LLC to find a new circuit in language models.
Lieke van der Vorst
From: https://x.com/marysia_cc/status/1861148591479288294/photo/1
-
Elena and Anna Balbusso
for Little Knife by Leigh Bardugo
From: https://x.com/marysia_cc/status/1861127999581528531/photo/1
#art
Franz Karl Leopold von Klenze
From: https://x.com/0zmnds/status/1861121676735586756/photo/1
-
Chesley Knight Bonestell, Jr.
From: https://x.com/0zmnds/status/1861297334195495170/photo/1
#art
IMO it's kind of weird that there aren't more blog posts in the rationality-sphere about how to do group house living well. There are a bunch of tricky problems that need solving and opportunities for clever solutions that make people better off, so you'd think there would be much fodder. Possibilities:
- Maybe people just don't think about it very much?
- "Group house living" isn't as culturally salient a category as "parenting", so we're not used to writing about it?
- Most of the problems involve being kind of annoyed at specific people, and so are inherently awkward to talk about?
like this
Yeah reading this I was like 'wow a lot of our lore is about chores'. I guess because this came up as an issue with us, whereas 'there are conflicts/annoyances with the other people' hasn't come up as much, possibly because two of the relationships were selected specifically for not being mutually annoying :p (and luckily you and Ben seem to not annoy each other that much)
Maybe the main tip is 'try to select people you really vibe with/share living preferences with', and if you manage that you will be well-placed to either not have problems (because your preferences don't clash), or to solve them?
Ben Weinstein-Raun likes this.
Ben Millwood
in reply to Daniel Filan • • •Daniel Filan
in reply to Ben Millwood • •Ben Weinstein-Raun
in reply to Daniel Filan • • •Daniel Ziegler
in reply to Daniel Filan • • •Ben Weinstein-Raun likes this.