bitexploder

Fear and anger are generally distinct emotions. You can have one without the other. Many of the same pathways are active with fear that are active with an anger in your mind, mechanically speaking. You can have anger without fear though.

I see a lot of people can conflict the two, but they are distinct as label labels and mechanical processes in the mind. Anger surfaces when there is someone to blame. Or something to blame. Something you can pin the emotional and surge on.

For your hypothesis above to be true, you generally need some sort of risk involved for fear to be the actual emotion. Anger can appear with zero fear and simply something blocking your way.

So I’m just not sure that I really agree on anger needing fear. There are plenty of situations where it doesn’t hold.

That seems about right. 4.8 is like in between 4.6 and 5 in terms of capability and language and they are all pretty close honestly. I just default to using Opus 5 for a coding agent that I don't interact with and I like driving with Opus 4.6 or Fable. Fable thinks too much though. Fable is like that engineer on your team that will over-engineer the shit out of something if you let them. Fable was like, "Here are 34 yaks, which shall we shave first <hands rubbing together>" and I was like can't we just... write the script first and then decide of any of these poor yaks need shaving?

I kind of still chill out on Opus 4.6 too. 4.8 is good too. I go between them. Opus 4.8 is a little smarter some of the time. Their use of language is both very different from Opus 5. Some of the time I have a hard time believing Opus 5 is even related to Opus 4.6 and 4.8.

I don't think we disagree. I'm just outlining a narrow aspect of how I like to operate. I meditate and spend a lot of time thinking about philosophy and neuroscience and cognition. It is a very introspective and contemplative practice. I don't ever assume that I have it all figured out, but I do always make sure that I have a plan for the day and that I just stay focused on that. And there are a lot of easier to sort things. Time with family, time with dogs, work, so on and so forth. It is not like I assume that I have figured it all out. I also get feedback from others that I trust as they understand how I work, etc. I have been meditating for over a decade, so I'm pretty good at shutting down the self-narrative and just being. It makes a lot of things more clear when all of the noise is turned off. What I do want is at the end of the day, after all of the alignment and processing, that I know what I wanted to accomplish in the day, and that I did those things, and that those were the right things to do that day. Maybe I course correct the next day, or even during the day, but I set out with a clear intention of what I want to have accomplished in the day. And that is rooted by breaking down a multi-year plan into month, week, and day blocks even. It isn't rigid. If I need to change it, I will. But plans are always worth having, even if they have to course correct, as nothing ever stays quite how you want once it reaches contact with reality.

I think that you will fully misinterpreted what I wrote. Sorry maybe that is not fair. Assuming good intent. Let me expand what I was thinking: You should still do long-term planning and use that to decide how you will stay present. Nothing I wrote precludes you from doing long-term planning and incorporating that in your day-to-day living and staying present. I am not suggesting that you turn into a hedonistic animal that cannot think beyond two days into the future. I am saying that once you know what you should be doing you should be doing those things without question as they are the things you decided we’re right. That can be long-term and short-term.

Yes, maybe. Evidence around caveman shows this isn’t a big deal for token consumption (forcing it to respond in a way it was not tuned) and I don’t see a big difference either way. And maybe it increases intelligence in hard to measure ways. Lots of parameters in these models.

I feel I fight them less with this setup. They get so lost in their own invented bullshit they stop being useful pretty often without it. So I would take bets on that :)

There is something incredibly important about a therapist I would not discard. And that is the human connection. We are hard wired for this and most often what people need out of a therapist is being seen. Another human being to fully and unerringly see you. To get to know you well enough to simulate /your/ thought loop and reflect it back to you, potentially with some meaningful observations. But that knowing that some other human sees you is mostly what therapy is about to me, some exceptions, of course, but in general

Also: incompleteness theorem again. The system cannot fully observe itself. A close friend can also do this too. In fact they are better at it than a therapist if you are open with the friend.

Also evil is a moral inclination and hard to discuss easily in this frame. Very little that is inherent in our machinery is disinclined towards evil or good. As social animals we must teach those values and hold them up. Different discussion :)

We do have metacognition and should use. It isn’t like we should go off the rails and become fully hedonic to “live in the moment” for example, we need some guiding principles . And when you are younger it can be really hard to know what is right. How to deal with purpose, and friends, and significant others, and all of the pain of the human condition. How do you figure it all out?

You don’t have to. Not all at once, anyway. You have enough information to make the right choices every day and learn more. We will always make mistakes, but if we acted with our best knowledge and fulfilled what we thought were the best choices in a given say you won’t have to regret the mistakes. That was kind of my whole point :)

And yes, always improve, and learn philosophy, and take care of your body and people around you. But at some point it all becomes obvious I think. At 46 it is all obvious for me now.

I’m pretty tall. I find that mostly beneficial except for walking into things. I also don’t fit well in Miata and some convertibles and other small cars, which is a bit of a problem on occasion.

I was going to add more, but I kind of forgot. My post was a bit stream of consciousness about overthinking things, which I may have over over thought. Oh well. I’m kidding sort of. I was going to say that I see a lot of overthinking all over the place at work. Complex environments often lead people to assume that a complex solution is needed. Some of the time the complexity is unavoidable. Most of the time the complexity is avoidable. Our individual lives and affairs are no different.

The problem I have with modern therapy is it that it operates completely within the frame of oneself. They will very carefully walk with you inside of your own narrative, bringing you to a better place of operation within that framework of the self.

Having thought about it some, I have been concluded that there’s only so far one can get when operating within that narrative. There’s only so much they can help with until you yourself can see that the “I” is it itself a machine generating the narrative, but also something else and more than just that narrative that is constantly running in our brains. It is something that can be observed. The observer is the observed as Krishnamurti would say.

As long time meditator, I find that there is nothing that is hidden from ourselves if we look. Meditation is one of the few ways to begin to access the whole and step outside of the narrative loop. You can also try psychedelics, but that’s riskier. A good trip might save you a decade of therapy. A bad one might genuinely imbalance your brains machinery. In some sense, the time is wasted. However, if you genuinely spend this time learning to observe yourself, this is very much a way of being that is not wasted in my opinion.

A therapist is generally going to work with you inside of a particular framework and way of stepping towards a particular way of operating. It sort of like the long or hard way of solving a math problem before you learned underlying theory of whatever problem you were working on.

I am mid 40s. Started working at a hyper scaler a couple years ago. I think you are overthinking it. All there is, is to be present and focus on what is important. And no one else can tell you what is important, you already know. Spend your time wisely with your family or on things that matter to you and not on trivial pursuit. There are no secrets in this. It is largely what you define.

When you ask this question of others, you are asking what am I forgetting? What will I regret? They can’t tell you. They can tell you what they regretted it and maybe you will regret those same things they did. But you have to decide. In 10 years. In 20 years. All you have to do is ask your future self what you’re gonna regret that you didn’t do. I had found that we already know the answers to those things.

E: think of it this way. And you don’t have to have a dog for these questions to work they’re metaphorical really, but also physical if you do have a dog. Did you walk your dog? Did you play ball with it? If they die tomorrow, will you look back and say that today was good? Did you do everything you could do with your dog? If not go do those things right now don’t leave them undone. That’s all there is. And the same thing works with everything in life. You know what you should’ve done in the day. If you do those things in 10 or 20 years, you won’t look back and say I wish I would’ve done those things cause you’ve done them.

What I have learned about neuroscience and philosophy is this: most people are not willing to invest as much time in small pursuits versus understanding the machinery of their mind and what will actually let them understand themselves with more clarity.

Understanding the self loop. The strange self loop. Seeing that the thing that is doing the observing inside of us is the whole and that from within our own system, we cannot completely reason about ourselves. The incompleteness theorem.

We have named it we call it the default mode network. It is essentially a recursive predictive machine. It only stops when we die. It is essentially making predictions about reality constantly. It is retrieving previous state and associating at the current state to make decisions. What it associates with previous memory and current state are encoding of our hormonal state and mental state at that point in time.

You can call it emotions. You can try and reason about it. It just exists. At some level, you just have to learn to let it be. All of the sophisticated words of psychology and trying to describe the machinery from within the machinery. It’s useful, you need a guide to live life. To keep the machinery and coding clean and not over, indexing on states that aren’t useful. But it all boils down to some pretty simple things. And when you see that the observer is the observed, and that it is all a part of the whole of the self, things become more manageable, for me, at least.

Take grief for example. What is grief? It is a prediction error. An expectation of some past state that the prediction machinery keeps expecting to come up one way, but it comes up another. And when it does come up as an error it hurts. It’s OK to experience it and to embrace it even. But it’s also OK to see it for what it is. We can’t escape the machine. But we can step outside of it briefly and watch it at work and acknowledge what it actually is.

I don’t use their CLIs. OMP/OpenCode. If you are stuck on them, you can do it in a proxy layer.

Nope, you gotta understand. This is the move. It is appended at the end of every single message. It is never saved in the conversation that get sent back for inference. So you send it. But when you go back for inference, it’s at the top of the stack so all of the cashing works you’re not pre-filling every time or anything like that. It burns plus N tokens, where N is my prompt stack. It is not really that expensive. I have measured it to within an inch of it its life. Think of it this way every bit of the prefix and the conversation stays exactly the same you’re only adding to the very end of the conversation. So after the first turn, it is basically always cashing within the KV cache for a given context. Sorry I am using voice dictation. My hands are tired this week. Basically you sculpt the conversation history to ensure prefix caching

This is actually funny because I've been doing some RAG experiments and I've basically trained up a RAG on several gigabytes of philosophy and neuroscience. And when I want to go metacognitive about thought processes, I actually really enjoy exploring different areas of philosophy, more life philosophy, but how it intersects with psychology and neuroscience and observing the observer, or the observer is the observed, as they say.

And when you ground it with real data, it's actually extremely useful. It's not exactly like Claude Therapist, but it's sort of the teach me about philosophy, but actually grounded and not vied. I have a lot of really strict prompts and grounding and agentic guidelines for this particular agent flow and harness that I've built.

And it's just a few weekends of vibing and feeding it basically all of Wikipedia and several gigabytes of papers and stuff, but it actually leads to interesting discussion. So I just have my personal philosophy bot and it's pretty fun.

One of the modalities I built is having two agents assume a famous persona. And then they take a thing, like grief or some thing that I experienced during the week, and they assume the role of the two different philosophers, and I just have them go back and forth 30, 40, 50 turns. And it's actually quite interesting, and it really moderates their language and tonality and behavior. They really get into the roles when you have the right prompting and grounding. Sometimes they get a little off the rails, but it leads to genuinely interesting areas to explore, and then I'll actually go read source material and things like that. I don't know, that's how I do therapy these days, but I never actually did therapy, so I just think a lot, now with agents finding interesting stuff to think about too!

I have basically polluted my context and agent files, agent.md, claude.md, gemini.md with really strong guidance about using plain language. I even include small, mini design specifications, examples of plain engineering language, only using technical language and invented terms when they actually make sense (almost never). EVERY TURN.

I have a pretty large set of prompts that go into any software engineering, and I force every single agent to use an ephemeral style stack of prompt management. So, every turn it goes to the top of the stack and it is the very last thing they see in terms of all of my prompts and instructions and agent files. And then it gets taken out of the conversation so that it doesn't get sent to the agent the next turn (no context bloat). It has restored so much sanity.

I tried the caveman add-ons, and I felt like I was losing IQ points because I spend a lot of time reading agent output, and when they start talking like cavemen, I start thinking like cavemen. That was not good for my mental health. So, I try and make the agent talk like me and think like me. And it works, mostly. And my observation is that maybe I'm not the most efficient agentic thought process, but my sanity is retained.

All of that is to say that if something is reading like that to you, just have the agent rewrite it and read it in a rewritten tone because it's probably bad as it stands and your colleague did not put enough effort in it. It is /not/ good and you should not accept it as a default. We have to hold the line on stuff like this and maintain some semblence of normal human engineering standards that existed before AI. They are not making us better. They are making is lazy and dumber.

Opus 5 and other agents in the latest rounds of tuning have gotten ridiculously bad in terms of how they feel to interact with with all the invented language and localized nomenclature. It is an obvious bias that big words and technical talk looks good to the bottom of the bell curve, but when you actually try and understand it, it's horrible. So people say, "Yeah, that looks great," in all the RLHF rounds, and they run with it because they think it looks good, but it doesn't. It's terrible.

Hold the line. It isn't you. And it isn't a good methodology document.

I decided not to get used to its communication style. It encourages it to invent terminology and drift away from simple and proper engineering in my opinion. Also, it is pretty simple to change as long as you’re not using the Claude code CLI or desktop app.

It seems a little excessive to use another LLM. With OMP I basically created an ephemeral prompt stack all of my agent files. It walks up the directory tree looking for any Gemini.md, Agents.md or Claude.md files. And it puts those at the very top of the stack. Then at the end of every turn, it pops those off to preserve the conversation history. So every turn, they get all of my fresh instructions, which include things like what and how to use language, how to render results and things like that. Net effect, every turn, the agent gets the instructions and it adds to that turn's tokens, but it does not become a part of the conversation history, which is really important for not bloating up the context. So it's always just however many tokens are in that file instead of it becoming a permanent part of the context.

I have been using Muse Glimmer for a few days instead of A3B. It gets the job done quicker than A3B despite being several times slower.

Yeah, but be fair. Working with small models is a different ball game. Not all the batteries come included :)

I had to fix this on 35B A3B -- I have a proxy that just shuts it down if it gets to 2K thinking tokens and injects something like "We have thought enough, let's begin working." and it almost always finishes the turn then. It rarely needs more than 2K thinking tokens and if it does there is always next turn. I would need to see what 27B is actually doing, but these smaller Qwen models seem prone to this.

GLP-1 does a lot more than just stomach signaling. It shrinks visceral fat, it's been shown to clear liver fat, and it actually increases mitochondrial efficiency similar to if you were to do endurance work/cardio. There's a lot going on with it. That's kind of what the parent is talking about, all these second-order effects that were not obvious. There is also evidence of down-regulated mTOR signaling and epigenetic protection. And these aren't all just bullshit, big pharma funded studies either.

They controlled for BMI at least, so they knew that was going to be a common question. BMI is a pretty good proxy for that, and they still said that the majority of the effect came from the compound and not BMI.

That is exactly how this model has worked for me so far. Muse on a one-shot task will get to 80%. And if you even nudge it and say, "Hey, finish up," or "Review the syntax," boom, it's done. And I'm getting 20 t/s with Ollama on a MacBook M5 Pro with 48GB of RAM. It is a seriously impressive little model.

Have you tried Muse 30B yet? I have been impressed with it. I have Qwen 3.8 27B hammering away right now against Muse. And Muse is doing a little bit better.

Flash models and Gemini make more sense when you consider Gemini Enterprise and Workspace. Oh HN we generally care a lot about writing software. However, until Fable, Gemini 3.1 Pro was my default for doing any sort of discussion outside of software engineering. Fable is now on par with things like modifying cars, etc. But I am guessing Fable is a /lot/ more expensive to use.

And in a typical enterprise environment dealing with documents, images, and broader business reasoning skills matter. Agents are not just for code and text :)

I would also think about combining the descriptions of photographs with some sort of RAG to essentially help shoot you into the reference references easier.

I ran a 9B over my like 100k photo library — it was very good at it. And extracting any text. All local.

Pi and a similar set of tools is also likely similar to the harness these models are trained on. More complex harnesses burn reasoning tokens on these small models and in my benchmarking don't seem to be able to beat Pi ever. Usually it isn't close on some tests.

You really have to get the models to end their thinking. Almost any commercial model serving has safe guards like this to tune how much they think.