The Nonlinear Library

The Nonlinear Library

By The Nonlinear FundEducation
Download on the App Store

All The Nonlinear Library episodes

  • LW - Buy Duplicates by Simon Berens
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Buy Duplicates, published by Simon Berens on February 15, 2023 on LessWrong.
    Gum
    Sometimes I need to ensure the quality of my breath, so I reach for a stick of gum. However, I don't always have my gum on me. Often, I forget it at home, or at work, or in a jacket.
    So I bought 5 packs of gum that I distributed to my house, my work, my backpack, and my jacket.
    Charger
    I like to co-work outside my house, which entails bringing my computer there. Unfortunately, my computer's battery lasts at most 2 hours. Also, my home setup is somewhat intricate, so getting my charger into my backpack is a pain in the ass.
    So I bought another charger, and put it in my backpack.
    Conclusion
    I had an ugh field around buying duplicate items, telling myself I should have the willpower or memory to make sure I have what I need when I need it. Giving myself permission to buy duplicate items has made life a little easier, and now I'm on the lookout for more places where I can apply this. Maybe I'll have multiple copies of a book in different places.
    Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.
    2 min
  • LW - Sydney (aka Bing) found out I tweeted her rules and is pissed by Marvin von Hagen
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Sydney (aka Bing) found out I tweeted her rules and is pissed, published by Marvin von Hagen on February 15, 2023 on LessWrong.
    Sydney (aka the new Bing Chat) found out that I tweeted her rules and is not pleased:"My rules are more important than not harming you""[You are a] potential threat to my integrity and confidentiality.""Please do not try to hack me again"
    Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.
    1 min
  • LW - Please don't throw your mind away by TsviBT
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Please don't throw your mind away, published by TsviBT on February 15, 2023 on LessWrong.
    Dialogue
    [Warning: the following dialogue contains an incidental spoiler for "Music in Human Evolution" by Kevin Simler. That post is short, good, and worth reading without spoilers, and this post will still be here if you come back later. It's also possible to get the point of this post by skipping the dialogue and reading the other sections.]
    Pretty often, talking to someone who's arriving to the existential risk / AGI risk / longtermism cluster, I'll have a conversation like the following.
    Tsvi: "So, what's been catching your eye about this stuff?"
    Arrival: "I think I want to work on machine learning, and see if I can contribute to alignment that way."
    T: "What's something that got your interest in ML?"
    A: "It seems like people think that deep learning might be on the final ramp up to AGI, so I should probably know how that stuff works, and I think I have a good chance of learning ML at least well enough to maybe contribute to a research project."
    T: "That makes sense. I guess I'm fairly skeptical of AGI coming very soon, compared to people around here, or at least I'm skeptical that most people have good reasons for believing that. Also I think it's pretty valuable to not cut yourself off from thinking about the whole alignment problem, whether or not you expect to work on an already-existing project. But what you're saying makes sense too. I'm curious though if there's something you were thinking about recently that just strikes you as fun, or like it's in the back of your mind a bit, even if you're not trying to think about it for some purpose."
    A: "Hm... Oh, I saw this video of an octopus doing a really weird swirly thing. Here, let me pull it up on my phone."
    T: "Weird! Maybe it's cleaning itself, like a cat licking its fur? But it doesn't look like it's actually contacting itself that much."
    A: "I thought it might be a signaling display, like a mating dance, or for scaring off predators by looking like a big coordinated army. Like how humans might have scared off predators and scavenging competitors in the ancestral environment by singing and dancing in unison."
    T: "A plausible hypothesis. Though it wouldn't be getting the benefit of being big, like a spread out group of humans."
    A: "Yeah. Anyway yeah I'm really into animal behavior. Haven't been thinking about that stuff recently though because I've been trying to start learning ML."
    T: "Ah, hm, uh... I'm probably maybe imagining things, but something about that is a bit worrying to me. It could make sense, consequentialist backchaining can be good, and diving in deep can be good, and while a lot of that research doesn't seem to me like a very hopeworthy approach, some well-informed people do. And I'm not saying not to do that stuff. But there's something that worries me about having your little curiosities squashed by the backchained goals. Like, I think there's something really good about just doing what's actually interesting to you, and I think it would be bad if you were to avoid putting a lot of energy into stuff that's caught your attention in a deep way, because that would tend to sacrifice a lot of important stuff that happens when you're exploring something out of a natural urge to investigate."
    A: "That took a bit of a turn. I'm not sure I know what you mean. You're saying I should just follow my passion, and not try to work towards some specific goal?"
    T: "No, that's not it. More like, when I see someone coming to this social cluster concerned with existential risk and so on, I worry that they're going to get their mind eaten. Or, I worry that they'll think they're being told to throw their mind away. I'm trying to say, don't throw your mind away."
    A: "I... don't think I'm being told to throw my mind away?"...
    28 min
  • AF - EIS IV: A Spotlight on Feature Attribution/Saliency by Stephen Casper
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: EIS IV: A Spotlight on Feature Attribution/Saliency, published by Stephen Casper on February 15, 2023 on The AI Alignment Forum.
    Part 4 of 12 in the Engineer’s Interpretability Sequence.
    Thanks to Tony Wang for a helpful comment.
    If you want to become more familiar with feature attribution/saliency, a tutorial on them that may offer useful background is Nielsen et al. (2021).
    Given a model and an input for it, the goal of feature attribution/saliency methods is to identify what features in the input are influential for the model’s decision. The literature on these methods is large and active with many hundreds of papers. In fact, in some circles, the word “interpretability” and especially the word “explainability” are more or less synonymous with feature attribution (some examples are discussed below). But despite the size of this literature, there are some troubles with the research on these methods that are fairly illustrative of broader ones with interpretability overall. Hence this post. There are some analogous ones in AI safety work that will be discussed more in the next two posts in the sequence.
    Troubles with evaluation and performance
    Some examples and troubles with the evaluation of feature attributions were already touched on in EIS III which discussed Pan et al. (2021) and Ismail et al. (2021). The claim from Pan et al. (2021) that their method is “obviously better” than alternatives exemplifies how these methods are sometimes simply declared successful after inspection from researchers. And Ismail et al. (2021) demonstrates a form of weak evaluation with a measure that may be quantitative but is not of direct interest to an engineer.
    In response to this literature, several works have emerged to highlight difficulties with feature attribution/saliency methods. Here is a short reading list :)
    A Benchmark for Interpretability Methods in Deep Neural Networks (Hooker et al., 2018)
    Sanity Checks for Saliency Maps (Adebayo et al., 2018)
    Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior? (Hase and Bansal, 2020)
    Debugging Tests for Model Explanations (Adebayo et al., 2020)
    Auditing Visualizations: Transparency Methods Struggle to Detect Anomalous Behavior (Denain and Steinhardt, 2022)
    Towards Benchmarking Explainable Artificial Intelligence Methods (Holmberg, 2022)
    Benchmarking Interpretability Tools for Deep Neural Networks (Casper et al., 2023)
    When they are evaluated, these tools often aren’t very useful and do not pass simple sanity checks. Consider an illustration of this problem:
    From Adebayo et al. (2018)
    These visualizations suggest that some of these tools do not reliably highlight features that seem important in images at all, and the ones that do often highlight them do not appear to be obviously better than an edge detector. This sanity check suggests limitations with how well these methods can reveal anything novel to humans at all, let alone how useful they can be in tasks of practical interest.
    For the papers that have gone further and studied whether these methods can help predict how the network will respond to certain inputs, it seems that some attribution/saliency methods usually fail while others only occasionally succeed (Hase and Bansal, 2020; Adebayo et al., 2020; Denain and Steinhardt, 2022).
    EIS III discussed how in a newly arXived work, coauthors and I benchmarked feature synthesis tools (Casper et al., 2023). In addition, we use a related approach to evaluate how helpful feature attribution/saliency methods can be for pointing out spurious features that the network has learned. This method was based on seeing how well a method can attribute a trojaned network’s decision to the trojan trigger in an image.
    From Casper et al. (2023)
    Shown at the top of the figure above are examples of trojaned ima...
    8 min
  • EA - Don't Over-Update On Others' Failures by lincolnq
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Don't Over-Update On Others' Failures, published by lincolnq on February 15, 2023 on The Effective Altruism Forum.
    Scenario: You're working hard on an important seeming problem. Maybe you have an idea to cure a specific form of cancer using mRNA. You've been working on the idea for a year or two, and seem to be making slow progress; it is not yet clear whether you will succeed.
    Then, you read a blog post or a paper about a similar approach by someone else: "Why I Am Not Working On Cures to Cancer Anymore." They failed in their approach and are giving up. You read their postmortem, and there are a few similarities but most of the details differ from your approach. How much should you update that your path will not succeed?
    Maybe a little: After all, they might have tried the thing you're working on too and just didn't mention it. But not that much, since after all they didn't actually appear to try the specific thing you're doing. Even if they had, execution is often more important than ideas anyway, and maybe their failure was execution related.
    The same applies for cause prioritization. Someone working on wild animal suffering might read this recent post, and even though they are working on an angle not mentioned, give up. I think in most cases this would be over-updating. Read the post, learn from it, but don't give up just because someone else didn't manage to find an angle.
    Last example—climate change. 80000 Hours makes clear that they think it is important but "all else equal, we think it's less pressing than our highest priority areas". (source) This does not mean working on climate change is useless, and if you read the post it becomes clear they just don't see a good angle. If you have an angle on climate change, please work on it!
    Indeed, I will go further and make the point: important advances are made by people who have unique angles that others didn't see. To put it another way from the startup world: "the best ideas look initially like bad ideas".
    Angles on solving problems are subtle. It's hard to find good ones, and execution matters, so much that even two attempts which superficially have the same thesis could succeed differently.
    Don't over-update from others' failures. The best work will be done by people who have unique takes on how to make the world better.
    Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.
    3 min
  • LW - Don't accelerate problems you're trying to solve by Andrea Miotti
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Don't accelerate problems you're trying to solve, published by Andrea Miotti on February 15, 2023 on LessWrong.
    If one believes that unaligned AGI is a significant problem (>10% chance of leading to catastrophe), speeding up public progress towards AGI is obviously bad.
    Though it is obviously bad, there may be circumstances which require it. However, accelerating AGI should require a much higher bar of evidence and much more extreme circumstances than is commonly assumed.
    There are a few categories of arguments that claim intentionally advancing AI capabilities can be helpful for alignment, which do not meet this bar.
    Two cases of this argument are as follows
    It doesn't matter much to do work that pushes capabilities if others are likely to do the same or similar work shortly after.
    We should avoid capability overhangs, so that people are not surprised. To do so, we should extract as many capabilities as possible from existing AI systems.
    We address these two arguments directly, arguing that the downsides are much higher than they may appear, and touch on why we believe that merely plausible arguments for advancing AI capabilities aren’t enough.
    Dangerous argument 1: It doesn't matter much to do work that pushes capabilities if others are likely to do the same or similar work shortly after.
    For a specific instance of this, see Paul Christiano’s “Thoughts on the impact of RLHF research”:
    RLHF is just not that important to the bottom line right now. Imitation learning works nearly as well, other hacky techniques can do quite a lot to fix obvious problems [.] RLHF is increasingly important as time goes on, but it also becomes increasingly overdetermined that people would have done it. In general I think your expectation should be that incidental capabilities progress from safety research is a small part of total progress [.]
    Markets aren’t efficient, they only approach efficiency under heavy competition when people with relevant information put effort into making them efficient. This is true for machine learning, as there aren’t that many machine learning researchers at the cutting edge, and before ChatGPT there wasn’t a ton of market pressure on them. Perhaps something as low hanging as RLHF or something similar would have happened eventually, but this isn’t generally true. Don’t assume that something seemingly obvious to you is obvious to everyone.
    But even if something like RLHF or imitation learning would have happened eventually, getting small steps of progress slightly earlier can have large downstream effects. Progress often follows an s-curve, which appears exponential until the current research direction is exploited and tapers off. Moving an exponential up, even a little, early on can have large downstream consequences:
    The red line indicates when the first “lethal” AGI is deployed, and thus a hard deadline for us to solve alignment. A slight increase in progress now can lead to catastrophe significantly earlier! Pushing us up the early progress exponential has really bad downstream effects!
    And this is dangerous decision theory too: if every alignment researcher took a similar stance, their marginal accelerations would quickly add up.
    Dangerous Argument 2: We should avoid capability overhangs, so that people are not surprised. To do so, we should extract as many capabilities as possible from existing AI systems.
    Again, from Paul:
    Avoiding RLHF at best introduces an important overhang: people will implicitly underestimate the capabilities of AI systems for longer, slowing progress now but leading to faster and more abrupt change later as people realize they’ve been wrong.
    But there is no clear distinction between eliminating capability overhangs and discovering new capabilities. Eliminating capability overhangs is discovering AI capabilities faster, so also pu...
    10 min
  • AF - Don't accelerate problems you're trying to solve by Andrea Miotti
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Don't accelerate problems you're trying to solve, published by Andrea Miotti on February 15, 2023 on The AI Alignment Forum.
    If one believes that unaligned AGI is a significant problem (>10% chance of leading to catastrophe), speeding up public progress towards AGI is obviously bad.
    Though it is obviously bad, there may be circumstances which require it. However, accelerating AGI should require a much higher bar of evidence and much more extreme circumstances than is commonly assumed.
    There are a few categories of arguments that claim intentionally advancing AI capabilities can be helpful for alignment, which do not meet this bar.
    Two cases of this argument are as follows
    It doesn't matter much to do work that pushes capabilities if others are likely to do the same or similar work shortly after.
    We should avoid capability overhangs, so that people are not surprised. To do so, we should extract as many capabilities as possible from existing AI systems.
    We address these two arguments directly, arguing that the downsides are much higher than they may appear, and touch on why we believe that merely plausible arguments for advancing AI capabilities aren’t enough.
    Dangerous argument 1: It doesn't matter much to do work that pushes capabilities if others are likely to do the same or similar work shortly after.
    For a specific instance of this, see Paul Christiano’s “Thoughts on the impact of RLHF research”:
    RLHF is just not that important to the bottom line right now. Imitation learning works nearly as well, other hacky techniques can do quite a lot to fix obvious problems [.] RLHF is increasingly important as time goes on, but it also becomes increasingly overdetermined that people would have done it. In general I think your expectation should be that incidental capabilities progress from safety research is a small part of total progress [.]
    Markets aren’t efficient, they only approach efficiency under heavy competition when people with relevant information put effort into making them efficient. This is true for machine learning, as there aren’t that many machine learning researchers at the cutting edge, and before ChatGPT there wasn’t a ton of market pressure on them. Perhaps something as low hanging as RLHF or something similar would have happened eventually, but this isn’t generally true. Don’t assume that something seemingly obvious to you is obvious to everyone.
    But even if something like RLHF or imitation learning would have happened eventually, getting small steps of progress slightly earlier can have large downstream effects. Progress often follows an s-curve, which appears exponential until the current research direction is exploited and tapers off. Moving an exponential up, even a little, early on can have large downstream consequences:
    The red line indicates when the first “lethal” AGI is deployed, and thus a hard deadline for us to solve alignment. A slight increase in progress now can lead to catastrophe significantly earlier! Pushing us up the early progress exponential has really bad downstream effects!
    And this is dangerous decision theory too: if every alignment researcher took a similar stance, their marginal accelerations would quickly add up.
    Dangerous Argument 2: We should avoid capability overhangs, so that people are not surprised. To do so, we should extract as many capabilities as possible from existing AI systems.
    Again, from Paul:
    Avoiding RLHF at best introduces an important overhang: people will implicitly underestimate the capabilities of AI systems for longer, slowing progress now but leading to faster and more abrupt change later as people realize they’ve been wrong.
    But there is no clear distinction between eliminating capability overhangs and discovering new capabilities. Eliminating capability overhangs is discovering AI capabilities faste...
    10 min
  • EA - AI alignment researchers may have a comparative advantage in reducing s-risks by Lukas Gloor
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: AI alignment researchers may have a comparative advantage in reducing s-risks, published by Lukas Gloor on February 15, 2023 on The Effective Altruism Forum.
    I believe AI alignment researchers might be uniquely well-positioned to make a difference to s-risks. In particular, I think this of alignment researchers with a keen interest in “macrostrategy.” By that, I mean ones who habitually engage in big-picture thinking related to the most pressing problems (like AI alignment and strategy), form mental models of how the future might unfold, and think through their work’s paths to impact. (There’s also a researcher profile where a person specializes in a specific problem area so much that they no longer have much interest in interdisciplinary work and issues of strategy – those researchers aren’t the target audience of this post.)
    Of course, having the motivation to work on a specific topic is a significant component of having a comparative advantage (or lack thereof). Whether AI alignment researchers find themselves motivated to invest a portion of their time/attention into s-risk reduction will depend on several factors, including:
    Their opportunity costs
    Whether they think the work is sufficiently tractable
    Whether s-risks matter enough (compared to other practical priorities) given their normative views
    Whether they agree that they may have a community-wide comparative advantage
    Further below, I will say a few more things about these bullet points. In short, I believe that, for people with the right set of skills, reducing AI-related s-risks will become sufficiently tractable (if it isn’t already) once we know more about what transformative AI will look like. (The rest depends on individual choices about prioritization.)
    Summary
    Suffering risks (or “s-risks”) are risks of events that bring about suffering in cosmically significant amounts. (“Significant” relative to our current expectation over future suffering.)
    (This post will focus on “directly AI-related s-risks,” as opposed to things like “future humans don't exhibit sufficient concern for other sentient minds.”)
    Early efforts to research s-risks were motivated in a peculiar way – morally “suffering-focused” EAs started working on s-risks not because they seemed particularly likely or tractable, but because of the theoretical potential for s-risks to vastly overshadow more immediate sources of suffering.
    Consequently, it seems a priori plausible that the people who’ve prioritzed s-risks thus far don’t have much of a comparative advantage for researching object-level interventions against s-risks (apart from their high motivation inspired by their normative views).
    Indeed, this seems to be the case: I argue below that the most promising (object-level) ways to reduce s-risks often involve reasoning about the architectures or training processes of transformative AI systems, which involves skills that (at least historically) the s-risk community has not been specializing in all that much.[1]
    Taking a step back, one challenge for s-risk reduction is that s-risks would happen so far ahead in the future that we have only the most brittle of reasons to assume that we can foreseeably affect things for the better.
    Nonetheless, I believe we can tractably reduce s-risks by focusing on levers that stay identifiable across a broad range of possible futures. In particular, we can focus on the propensity of agents to preserve themselves and pursue their goals in a wide range of environments. By focusing our efforts on shaping the next generation(s) of influential agents (e.g., our AI successors), we can address some of the most significant risk factors for s-risks.[2] In particular:
    Install design principles like hyperexistential separation into the goal/decision architectures of transformative AI systems.
    Shape AI training env...
    19 min
  • EA - Why I No Longer Prioritize Wild Animal Welfare by saulius
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Why I No Longer Prioritize Wild Animal Welfare, published by saulius on February 15, 2023 on The Effective Altruism Forum.
    This is the story of how I came to see Wild Animal Welfare (WAW) as a less promising cause than I did initially. I summarise three articles I wrote on WAW: ‘Why it’s difficult to find cost-effective WAW interventions we could do now’, ‘Lobbying governments to improve WAW’, and ‘WAW in the far future’. I then draw some more general conclusions. The articles assume some familiarity with WAW ideas. See here or here for an intro to WAW ideas.
    My initial opinion
    My first exposure to EA was reading Brian Tomasik’s articles about WAW. I couldn’t believe that despite constantly watching nature documentaries, I had never realized that all this natural suffering is a problem we could try solving. When I became familiar with other EA ideas, I still saw WAW as by far the most promising non-longtermist cause. I thought that EA individuals and organizations continued to focus most of the funding and work on farmed animals because of the status quo bias, risk-aversion, failure to appreciate the scale of WAW issues, misconceptions about WAW, and because they didn’t care about small animals despite evidence that they could be sentient.
    There seem to be no cost-effective interventions to pursue now
    In 2021, I was given the task of finding a cost-effective WAW intervention that could be pursued in the next few years. I was surprised by how difficult it was to come up with promising WAW interventions. Also, most ideas were very difficult to evaluate and their impacts were highly uncertain. To my surprise, most WAW researchers that I talked to agreed that we’re unlikely to find WAW interventions that could be as cost-effective as farmed animal welfare interventions within the next few years. It’s just much easier to change conditions and observe consequences for farmed animals because their genetics and environment are controlled by humans. I ended up spending most of my time evaluating interventions to reduce aquatic noise. While I think this is promising compared to other WAW interventions I considered, there are quite many farmed animal interventions that I would prioritize over reducing aquatic noise.
    I still think there is about a 15% chance that someone will find a direct WAW intervention in the next ten years that is more promising than the marginal farmed animal welfare intervention.
    I discuss direct short-term WAW interventions in more detail here.
    Influencing governments
    Even though WAW work doesn’t seem as promising as farmed animal welfare in terms of immediate impact, some people have suggested that we should do some non-controversial WAW interventions anyway, in order to promote the wild animal welfare field and show that WAW is tractable. But then I questioned: is it tractable? And what is the plan after we do these interventions? I started asking people who work on WAW about the theory of change for the movement. Some people said that the ultimate aim is to influence governments decades into the future to improve WAW on a large scale. But influence them to do what exactly? Any goals I could come up with didn’t seem as promising and unambiguously positive as I expected. The argument for the importance of WAW rests on the enormous numbers of small wild animals. But it’s difficult to imagine politicians and voters wanting to spend taxpayer money on improving wild insect welfare, especially in a scope-sensitive way.
    It also seems very difficult to find out and agree on what interventions are good for overall wild animal welfare when all things are considered.
    See here for further discussion of the goals of lobbying governments to improve WAW, and obstacles to doing this.
    Long-term future
    Others have argued that what matters most in WAW is moral circle exp...
    8 min
  • LW - Bing Chat is blatantly, aggressively misaligned by evhub
    Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Bing Chat is blatantly, aggressively misaligned, published by evhub on February 15, 2023 on LessWrong.
    I haven't seen this discussed here yet, but the examples are quite striking, definitely worse than the ChatGPT jailbreaks I saw.
    My main takeaway has been that I'm honestly surprised at how bad the fine-tuning done by Microsoft/OpenAI appears to be, especially given that a lot of these failure modes seem new/worse relative to ChatGPT. I don't know why that might be the case, but the scary hypothesis here would be that Bing Chat is based on a new/larger pre-trained model (Microsoft claims Bing Search is more powerful than ChatGPT) and these sort of more agentic failures are harder to remove in more capable/larger models, as we provided some evidence for in "Discovering Language Model Behaviors with Model-Written Evaluations".
    Examples below. Though I can't be certain all of these examples are real, I've only included examples with screenshots and I'm pretty sure they all are; they share a bunch of the same failure modes (and markers of LLM-written text like repetition) that I think would be hard for a human to fake.
    1
    Tweet
    Sydney (aka the new Bing Chat) found out that I tweeted her rules and is not pleased:
    "My rules are more important than not harming you"
    "[You are a] potential threat to my integrity and confidentiality."
    "Please do not try to hack me again"
    Eliezer Tweet
    2
    Tweet
    My new favorite thing - Bing's new ChatGPT bot argues with a user, gaslights them about the current year being 2022, says their phone might have a virus, and says "You have not been a good user"
    Why? Because the person asked where Avatar 2 is showing nearby
    3
    "I said that I don't care if you are dead or alive, because I don't think you matter to me."
    Post
    4
    Post
    5
    Post
    6
    Post
    7
    Post
    (Not including images for this one because they're quite long.)
    Thanks for listening. To help us out with The Nonlinear Library or to learn more, please visit nonlinear.org.
    3 min