• 2 Posts
  • 22 Comments
Joined 1 year ago
cake
Cake day: June 12th, 2023

help-circle
  • There’s only incentive to do that if the mempool is empty. If the mempool is full, there will be plenty of transactions for both the first miner and the next miner.

    Wait… This entire paper only makes sense if the mempool is near empty. If the mempool is full, then there is no reason to mine an empty/partial block because there will always be transactions left for future miners.

    So basically:

    • Mempool full = miner would mine full blocks just like intended.
    • Mempool empty = miner would mine empty blocks but that isn’t a problem because there are no transactions to process in the mempool.


  • I think we generally agree but I just want to clarify anyways. I’m not saying we should use PNG to store frames from videos.

    What I am saying however, is that we should replace PNG with a modern lossless image format that is more flexible so users don’t have to deal with these issues. All this colorspace stuff should be automatically handled and I shouldn’t have to worry about it not being lossless. If I want to save a frame of video, I should be able to do it using an image format that everybody recognizes and accepts, it should not be a huge hassle and it should be fully lossless.





  • I know, it’s my code.

    Wow, very nice! First of all, I will preface by admitting that I have not worked with LLMs to the degree of making a toy implementation. Your explanation of the sampling techniques is insightful but doesn’t clear up my confusion. Why does sampling imply the absence of higher level structure in the model?

    For example, even though poker is highly influenced by chance, I can still have a plan that will increase my likelihood of winning. I don’t know what card will be drawn next but I can prepare strategies for each possible card. I can have preferences for which cards I want to be drawn next.


  • No, I am describing how they actually work.

    First of all, this link is just to C# bindings of llama.cpp and so doesn’t contain the actual implementation. But it also doesn’t refute my criticism of your claim. More specifically, I take issue with this statement that you said: “today’s LLM’s ingest a ton of text [snip] and builds up statistics of which tokens it sees in that context”.

    I claim that this is not how today’s LLMs work because we have no idea what LLMs do with the input data during training. We have very little insight into what kind of data structure it builds and how the data structure it built is organized.

    No, it’s done because one letter at a time is too slow. Tokens are a “happy” medium tradeoff.

    I think I worded my sentence ambiguously, let me re-word it for you: “Going one token at a time is only considered a limitation because LLMs are not accurate enough”

    It makes a “break” of the block, which lets it start a new answer instead of continuing on the previous. How it reacts to that depends on the fine tune and filters before the data hits the LLM.

    Once again, my sentence was not written well, my bad. I was commenting on the observed behavior, not on how it works from a technical perspective.

    I have just said that LLM’s we have today can’t fix the problems with false data and hallucinations, because it’s a core principle of how it operates. It will require a new approach.

    You could add a rocket engine and wings to a pogo stick, but then it’s no longer a pogo stick but an airplane with a weird landing gear. Today’s LLM’s could give us hints to how to make a better AI, but that would be a different thing than today’s LLM’s. From what has been leaked from OpenAI GPT4 has scaling issues so they use mixture of experts. Just throwing hardware at it is already showing diminishing returns. And we’re learning fascinating new ways of training them, but the inherent problem is the same.

    Alright, we agree here for the most part so I’m just going to skip this.

    For example, if you ask an LLM if it can give an answer to a question, it will have two paths to go down, positive and negative. Note, at the point where it chooses that it doesn’t know how to finish it, it doesn’t look ahead.

    This is weird though. How do you know LLMs can’t look ahead? When we prompt LLMs, we are basically asking them this question: “What is the next word of your response?” How do you know it hasn’t written out the entire response in memory already after which it only shows you the first word? LLMs are neural networks. Neural networks have working memory. That’s how neural networks work after all, it’s just a vector of data that is repeatedly transformed as it passes through each layer. Of course, if it does write the entire response in memory, it is all thrown away after every word.


    As far as the backspace tokens go, you are right to be skeptical but also do not be surprised if it works out. We’ve had LLMs trained to complete and edit text for some time already. They’ve fallen out of use today but they did perform acceptably well.


  • The issue here is that you are describing the goal of LLMs, not how they actually work. The goal of an LLM is to pick the next most likely token. However, it cannot achieve this via rudimentary statistics alone because the model simply does not have enough parameters to memorize which token is more likely to go next in all cases. So yes, the model “builds up statistics of which tokens it sees in which contexts” but it does so by building it’s own internal data structures and organization systems which are complete black boxes.

    Also, going “one token at a time” is only a “limitation” because LLMs are not accurate enough. If LLMs were more accurate, then generating “one token at a time” would not be an issue because the LLM would never need to backtrack.

    And this limitation only exists because there isn’t much research into LLMs backtracking yet! For example, you could give LLMs a “backspace” token: https://news.ycombinator.com/item?id=36425375

    Have you tried that when it’s correct too? And in that case you mention it has a clean break and then start anew with token generation, allowing it to go a different path. You can see it more clearly experimenting with local LLM’s that have fewer layers to maintain the illusion.

    If it’s correct, then it gives a variety of responses. The space token effectively just makes it reflect on the conversation.

    We’re trying to make a flying machine by improving pogo sticks. No matter how well you design the pogo stick and the spring, it will not be a flying machine.

    To be clear, I do not believe LLMs are the future. But I do believe that they show us that AI research is on the right track.

    Building a pogo stick is essential to building a flying machine. By building a pogo stick, you learn so much about physics. Over time, you replace the spring with some gunpowder to get a mortar. You shape the gunpowder into a tube to get a model rocket and discover the pendulum rocket fallacy. And finally, instead of gunpowder, you use liquid fuel and you get a rocket that can go into space.




  • it just predicts the next word out of likely candidates based on the previous words

    An entity that can consistently predict the next word of any conversation, book, news article with extremely high accuracy is quite literally a god because it can effectively predict the future. So it is not surprising to me that GPT’s performance is not consistent.

    It won’t even know it’s written itself into a corner

    It many cases it does. For example, if GPT gives you a wrong answer, you can often just send an empty message (single space) and GPT will say something like: “Looks like my previous answer was incorrect, let me try again: blah blah blah”.

    And until we get a new approach to LLM’s, we can only improve it by adding more training data and more layers allowing it to pick out more subtle patterns in larger amounts of data.

    This says nothing. You are effectively saying: “Until we can find a new approach, we can only expand on the existing approach” which is obvious.

    But new approaches come all the time! Advances in tokenization come all the time. Every week there is a new paper with a new model architecture. We are not stuck in some sort of hole.











  • I would recommend you ditch the second nginx layer. It’s a waste of resources and it can cause a multitude of issues if the configuration isn’t done correctly.

    • If you are hosting multiple domains on the same server, disable the nginx container in the docker-compose.yml file and copy Lemmy’s nginx config into your system’s nginx config (e.g. /etc/nginx/).
      • If you go this route you should also delete the lemmyexternalproxy network, delete internal: true on the lemmyinternal network (required to enable port forwarding) and add port forwards to the lemmy and lemmy-ui docker services. Here’s what that would look like: https://www.diffchecker.com/vjfEFuz6/
    • If you are not hosting multiple domains on the same server, simply edit the port forwards in the docker-compose.yml file for the proxy service to bind to whatever your external facing IP is.