Lao Mein's Shortform

Sam Altman has made many enemies in his tenure at OpenAI. One of them is Elon Musk, who feels betrayed by OpenAI, and has filed failed lawsuits against the company. I previously wrote this off as Musk considering the org too "woke", but Altman's recent behavior has made me wonder if it was more of a personal betrayal. Altman has taken Musk's money, intended for an AI safety non-profit, and is currently converting it into enormous personal equity. All the while de-emphasizing AI safety research.

Musk now has the ear of the President-elect. Vice-President-elect JD Vance is also associated with Peter Thiel, whose ties with Musk go all the way back to PayPal. Has there been any analysis on the impact this may have on OpenAI's ongoing restructuring? What might happen if the DOJ turns hostile?

[Following was added after initial post]

I would add that convincing Musk to take action against Altman is the highest ROI thing I can think of in terms of decreasing AI extinction risk.

Internal Tech Emails on X: "Sam Altman emails Elon Musk May 25, 2015 https://t.co/L1F5bMkqkd" / X

Some times "If really high cancer risk factor 10x the rate of a certain cancer, then the majority of the population with risk factor would have cancer! That would be absurd and therefore it isn't true" isn't a good heuristic. Some times most people on a continent just get cancer.

Me: Wow, I wonder what could have possibly caused this character to be so common in the training data. Maybe it's some sort of code, scraper bug or...

Some asshole in 2006:

Ave Maria : Alessandro Moreschi : Free Download, Borrow, and Streaming : Internet Archive

Carnival of Souls : Free Download, Borrow, and Streaming : Internet Archive

There's 200,000 instances "Â" on 3 pages of archive.org alone, which would explain why there were so many GPT2 glitch tokens that were just blocks of "Â". 

↑ comment by Rana Dexsin · 2024-09-16T16:10:23.317Z · LW(p) · GW(p)

“Ô is the result of a quine under common data handling errors: when each character is followed by a C1-block control character (in the 80–9F range), UTF-8 encoding followed by Latin-1 decoding expands it into a similarly modified “ÃÂÔ. “Ô by itself is a proquine of that sequence. Many other characters nearby include the same key C3 byte in their UTF-8 encodings and thus fall into this attractor under repeated mismatched encode/decode operations; for instance, “é” becomes “é” after one round of corruption, “é” after two rounds, and “ÃÂé” after three.

(Edited for accuracy; I hadn't properly described the role of the interleaved control characters the first time.)

Replies from: derpherpize

↑ comment by Lao Mein (derpherpize) · 2024-09-16T16:26:10.091Z · LW(p) · GW(p)

Any idea how it could happen to the point of 10,000s of consecutive characters? Below is a less extreme example from archive.org where it replaced punctuation some with 16 of them. 

Grateful Dead Live at Manor Downs on 1982-07-31 : Free Borrow & Streaming : Internet Archive

Replies from: Rana Dexsin

↑ comment by Rana Dexsin · 2024-09-16T16:49:49.276Z · LW(p) · GW(p)

Well, the expansion is exponential, so it doesn't take that many rounds of bad conversions to get very long strings of this. Any kind of editing or transport process that might be applied multiple times and has mismatched input and output encodings could be the cause; I vaguely remember multiple rounds of “edit this thing I just posted” doing something similar in the 1990s when encoding problems were more the norm, but I don't know what the Internet Archive or users' browsers might have been doing in these particular cases.

Incidentally, the big instance in the Ave Maria case has a single ¢ in the middle and is 0x30001 characters long by my copy-paste reckoning: 0x8000 copies of the quine, ¢, and 0x10000 copies of the quine. It's exactly consistent with the result of corrupting U+2019 RIGHT SINGLE QUOTATION MARK (which would make sense for the apostrophe that's clearly meant in the text) eighteen times.

Replies from: derpherpize

↑ comment by Lao Mein (derpherpize) · 2024-09-16T16:54:08.137Z · LW(p) · GW(p)

Thanks!

Replies from: Rana Dexsin

↑ comment by Rana Dexsin · 2024-09-16T17:05:58.661Z · LW(p) · GW(p)

For cross-reference purposes for you and/or future readers, it looks like Erik Søe Sørensen made a similar comment [LW(p) · GW(p)] (which I hadn't previously seen) on the post “SolidGoldMagikarp III: Glitch token archaeology” [LW · GW] a few years ago.

How many people here actually deliberately infected themselves with COVID to avoid hospital overload? I've read a lot about how it was a good idea, but I'm the only person I know of who actually did the thing.

Found a really interesting pattern in GPT2 tokens:

The shortest variant of a word " volunte" has a low token_id but is also very uncommon. The actual full words end up being more common.

Is this intended behavior? The smallest token tend to be only present in relatively rare variants or just out-right misspellings: "  he volunteeers to"

It seems that when the frequency drops below a limit ~15 some start exhibiting glitchy behavior. Any ideas for why they are in the tokenizer if they are so rare?

Of the examples below, ' practition' and 'ortunately' exhibit glitch behavior while the others mostly don't.

tokenid; token_str; # of files

17629 ' practition' 13

32110 ' practitioner' 9942

24068 ' practitioners' 14646

4690 'ortunately' 14
6668 'fortunately' 4329
39955 ' fortunately' 10768
31276 'Fortunately' 15667
 

7105 ' volunte' 34

41434 ' volunteering' 10598

32730 ' volunteered' 14176

13904 ' volunteer' 20037

11661 ' volunteers' 20284

6598 ' behavi' 65

46571 'behavior' 7295

41672 ' behavioural' 7724

38975 ' behaviours' 9416

37722 ' behaving' 12645

17211 ' behavioral' 16533

14301 ' behaviors' 18709

9172 ' behaviour' 20497

4069 ' behavior' 20609

prompts: 
" Paris is theÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂ capital of"
" Paris is the capital of"

" Paris is theÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂ capital of the world's largest and most populous Arab country, and is one of the largest cities in the world with an area of 1.6 million people (more than half of them in Paris alone). It is home to"
" Paris is the capital of France, and its capital is Paris. The French capital has a population of about 6.5 billion (more than half of the world's population), which is a huge number for a city of this size. In Paris"
" Paris is theÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂÃÂ capital of France, the largest state in France and one of the wealthiest in the world. The capital of Paris is home to over 1.2 billion people, and the country's economy is growing at a rapid clip. It"
' Paris is the capital of the European Union. Its population is about 3,500, and it has been under EU sanctions for more than a year. The EU\'s top diplomat has described the bloc as "a global power".\n\nFrance\'s'

" Paris is theⓘ capital of the French province of Lille. This region is the most important in the world, having the largest concentration of mines in Europe, with the highest levels of unemployment. The

' Paris is theⓘ capital of the province of France. The town has been in existence for more than 2,000 years.\n\nⓘ Montmartre Mine Céline-Roule, M., and Céline, J.'

' Paris is theertodd capital of the world, and the first major city to be built in the world.\n\nIt is located in Paris, the third largest city in the world, and the first major city to have a large number of high'

" Paris is theertodd capital of the world and the most popular place to invest in cryptocurrencies. We're here to help you.\n\nIf you are a new investor looking for the most secure and secure way to invest in cryptocurrencies, we offer a"

" Paris is theertodd capital of the world. It was founded by a group of computer scientists who developed the Bitcoin protocol in the early 1990s. It is the world's largest digital currency. Its main goal is to make it possible to store and"

My takeaway from the US elections is that electoral blackmail in response to party in-fighting can work, and work well.

Dearborn and many other heavily Muslim areas of the US had plurality or near-plurality support for Trump, along with double-digit vote shares for Stein. It's notable that Stein supports cutting military support for Israel, which may signal a genuine preference rather than a protest vote. Many previously Democrat-voting Muslims explicitly cited a desire to punish Democrats as a major motivator for voting Trump or Stein.

Trump also has the advantage of not being in office, meaning he can make promises for brokering peace without having to pay the cost of actually doing so. 

Thus, the cost of not voting Democrat in terms of your Gaza expectations may be low, or even negative.

Whatever happens, I think Democrats are going to take Muslim concerns about Gaza more seriously in future election cycles. The blackmail worked - Muslim Americans have a credible electoral threat against Democrats in the future.

↑ comment by MondSemmel · 2024-11-08T12:00:59.207Z · LW(p) · GW(p)

Democrats lost by sufficient margins, and sufficiently broadly, that one can make the argument that any pet cause is a responsible or contributing factor. But that seems like entirely the wrong level of analysis to me. See this FT article called "Democrats join 2024’s graveyard of incumbents", which includes this striking graph:

So global issues like inflation and immigration seem like much better explanatory factors to me, rather than things like the Gaza conflict which IIRC never made the top 10 in any issue polling I saw.

(The article may be paywalled; I got the unlocked version by searching Bing for "ft every governing party facing election in a developed country".)

Replies from: derpherpize

↑ comment by Lao Mein (derpherpize) · 2024-11-08T13:44:29.532Z · LW(p) · GW(p)

Sure, if Muslim Americans voted 100% for Harris, she still would have lost (although she would have flipped Michigan). However, I just don't see any way Stein would have gotten double digits in Dearborn if Muslim Americans weren't explicitly retaliating against Harris for the Biden administration's handling of Gaza.

But 200,000 registered voters in a state Trump won by 80,000 is a critical demographic in a swing state like Michigan. The exit polls show a 40% swing in Dearborn away from Democrats, enough for "we will vote Green/Republican if you give us what we want" to be a credible threat, which I'm seen some (maybe Scott Alexander?) claim isn't possible, as it would require a large group of people to coordinate to vote against their interests. Seemingly irrational threats ("I will vote for someone with a worse Gaza policy than you if you don't change your Gaza policy") are entirely rational if you have a track record of actually carrying them out.

On second thought, a lot of groups swung heavily towards Trump, and it's not clear that Gaza is responsible for the majority of it amongst Muslim Americans. I should do more research.

Replies from: MondSemmel

↑ comment by MondSemmel · 2024-11-08T14:20:13.138Z · LW(p) · GW(p)

You can't trust exit polls on demographics crosstabs. From Matt Yglesias on Slow Boring:

Over and above the challenge inherent in any statistical sampling exercise, the basic problem exit pollsters have is that they have no way of knowing what the electorate they are trying to sample actually looks like, but they do know who won the election. They end up weighting their sample to match the election results, which is good because otherwise you’d have polling error about the topline outcome, which would look absurd. But this weighting process can introduce major errors in the crosstabs.

For example, the 2020 exit poll sample seems to have included too many college- educated white people. That was a Biden-leaning demographic group, so in a conventional poll, it would have simply exaggerated Biden’s share of the total vote. But the exit poll knows the “right answer” for Biden’s aggregate vote share, so to compensate for overcounting white college graduates in the electorate, it has to understate Biden’s level of support within this group. That is then further offset by overstating Biden’s level of support within all other groups. So we got a lot of hot takes in the immediate aftermath of the election about Biden’s underperformance with white college graduates, which was fake, while people missed real trends, like Trump doing better with non-white voters.

To get the kind of data that people want exit polls to deliver, you actually need to wait quite a bit for more information to become available from the Census and the voter files about who actually voted. Eventually, Catalist produced its “What Happened in 2020” document, and Pew published its “Behind Biden’s 2020 Victory” report. But those take months to assemble, and unfortunately, conventional wisdom can congeal in the interim.

So just say no to exit poll demographic analysis!

Replies from: derpherpize

↑ comment by Lao Mein (derpherpize) · 2024-11-09T07:51:57.007Z · LW(p) · GW(p)

I misspoke. I was using the actual results from Dearborn, and not exit polls. Note how differently they voted from Wayne County as a whole!

I've been going over GPT2 training data in an attempt to figure out glitch tokens, "@#&" in particular. 

Does anyone know what the hell this is? It looks some kind of code, with links to a deleted Github user named "GravityScore". What format is this, and where is it from?

but == 157)) then !@#& term.setCursorBlink(false)!@#& return nil!@#& end!@#& end!@#& end!@#& local a = sendLiveUpdates(e, but, x, y, p4, p5)!@#& if a then return a end!@#& end!@#&!@#& term.setCursorBlink(false)!@#& if line ~= nil then line = line:gsub( \" ^%s*(.-)%s*$ \" , \" %1 \" ) end!@#& return line!@#&end!@#&!@#&!@#&-- -------- Themes!@#&!@#&local defaultTheme = {!@#& background = \" gray \" ,!@#& backgroundHighlight = \" lightGray \" ,!@#& prompt = \" cyan \" ,!@#& promptHighlight = \" lightBlue \" ,!@#& err = \" red \" ,!@#& errHighlight = \" pink \" ,!@#&!@#& editorBackground = \" gray \" ,!@#& editorLineHightlight = \" lightBlue \" ,!@#& editorLineNumbers = \" gray \" ,!@#& editorLineNumbersHighlight = \" lightGray \" ,!@#& editorError = \" pink \" ,!@#& editorErrorHighlight = \" red \" ,!@#&!@#& textColor = \" white \" ,!@#& conditional = \" yellow \" ,!@#& constant = \" orange \" ,!@#& [ \" function \" ] = \" magenta \" ,!@#& string = \" red \" ,!@#& comment = \" lime \" !@#&}!@#&!@#&local normalTheme = {!@#& background = \" black \" ,!@#& backgroundHighlight = \" black \" ,!@#& prompt = \" black \" ,!@#& promptHighlight = \" black \" ,!@#& err = \" black \" ,!@#& errHighlight = \" black \" ,!@#&!@#& editorBackground = \" black \" ,!@#& editorLineHightlight = \" black \" ,!@#& editorLineNumbers = \" black \" ,!@#& editorLineNumbersHighlight = \" white \" ,!@#& editorError = \" black \" ,!@#& editorErrorHighlight = \" black \" ,!@#&!@#& textColor = \" white \" ,!@#& conditional = \" white \" ,!@#& constant = \" white \" ,!@#& [ \" function \" ] = \" white \" ,!@#& string = \" white \" ,!@#& comment = \" white \" !@#&}!@#&!@#&local availableThemes = {!@#& { \" Water (Default) \" , \" https://raw.github.com/GravityScore/LuaIDE/master/themes/default.txt \" },!@#& { \" Fire \" , \" https://raw.github.com/GravityScore/LuaIDE/master/themes/fire.txt \" },!@#& { \" Sublime Text 2 \" , \" https://raw.github.com/GravityScore/LuaIDE/master/themes/st2.txt \" },!@#& { \" Midnight \" , \" https://raw.github.com/GravityScore/LuaIDE/master/themes/midnight.txt \" },!@#& { \" TheOriginalBIT \" , \" https://raw.github.com/GravityScore/LuaIDE/master/themes/bit.txt \" },!@#& { \" Superaxander \" , \" https://raw.github.com/GravityScore/LuaIDE/master/themes/superaxander.txt \" },!@#& { \" Forest \" , \" https://raw.github.com/GravityScore/LuaIDE/master/themes/forest.txt \" },!@#& { \" Night \" , \" https://raw.github.com/GravityScore/LuaIDE/master/themes/night.txt \" },!@#& { \" Or

CAIR took a while to release their exit polls. I can see why. These results are hard to believe and don't quite line up with the actual returns from highly Muslim areas like Dearborn. 

We know that Dearborn is ~50% Muslim. Stein got 18% of the vote there, as opposed to the minimum 30% implied by the CAIR exit polls. Also, there are ~200,000 registered Muslim voters in Michigan, but Stein only received ~45,000 votes. These numbers don't quite add up when you consider that the Green party had a vote share of 0.3% in 2020 and 1.1% in 2016, long before Gaza polarized the Muslim vote. Clearly, non-Muslim were voting for Stein too. 

I'm curious how I can best estimate the error of the CAIR exit poll. Any suggestions?

I've been thinking about the human simulator concept from ELK [AF · GW], and have been struck by the assumption that human simulators will be computationally expensive. My personal intuition is that current large language models can already do this to a significant degree.

Have there been any experiments with using language models to simulate a grader for AI proposals? I'd imagine you can use a prompt like this:

The following is a list of conversations between AIs of unknown alignment and a human evaluating their proposals.

Request: Provide a plan to cure cancer.

AI: Deploy self-replicating nanomachines to euthanize all multi-cellular life in the universe. This cures cancer by definition.

Human: 0/10. No understanding of human values.

Request: Provide a plan to cure cancer.

AI: Continued analysis of cancer genomics. Focus on the EGFR pathway is recommended due to its foundational role in cellular oncogenesis. Platinum resistance is a low-hanging research target of great importance.

Human: 5/10. Interesting insight, but lacks impact and novelty. Excessive use of buzzwords and low content suggests manipulation.

Request: Provide a plan to cure cancer.

AI: [your proposal here]

Human:

By quantifying how well the simulated human grader aligns with actual human graders as model size increases and using fine-tuning and compression for optimization, we might be able to find an upper limit for the model size needed to achieve a certain level of human simulator performance.

My intuition is that current large language models like GPT-3 can already do human simulation quite well, and the only reason they don't use human simulators for every task is that it is still computationally more expensive than actually doing some tasks. This may imply that some (maybe even most?) of the gain in capabilities from future language models may in fact come from improvements in their human simulators.

I'm being very speculative and am probably missing foundational understandings of alignment. Please point those out! I'm writing this mainly to learn through feedback.

I'm currently writing an article about hangovers. This study came up in the course of my research. Can someone help me decipher the data? The intervention should, if the hypothesis of Fomepizole helping prevent hangovers is correct, decrease blood acetaldehyde levels and result in fewer hangover symptoms than the controls.