That aspect is probably more important for an open model then the source materials
552 karma · joined April 5, 2022
That aspect is probably more important for an open model then the source materials
Compared to some available Chinese model I guess. For most common tasks the additional intelligence is marginal and the cost is around an order of magnitude higher.
It is very different with plaintext
You also open the gates to many edge cases where I don't know what it is supposed to happen.
What if I never paste the cut text? Does it stay faded?
What if I copy some other text, maybe on another app, before pasting? Do I paste the faded text or the new copied one?
Also, as you mention, I hate that you can't paste on a different program given the fact that the clipboard is untouched when you press ctrl-x
The gain is so minimal for a lot of headaches if you don't follow the happy path
It is the reinforcement learning that produces more tangible results with less data, but it is something that the AI labs specifically selects and it is not picked up unknowingly
> Yeah if I'm not mistaken the ticketing service is deployed on t56kl-293d1-5gqp7, do you think we should move it to iwi49-oaqq0-492fa?
I believe this is the key difference, because it doesn't make sense to learn such a complicated tool for a simple use case.
If you already know it sure, it may make sense. But you are removing a huge initial effort required otherwise.
There are two clearly demarcations both north and south
Isn't this solved if you squash the commits when merging the PR? I personally don't care that much about the commits inside a PR, the are just temporary because when a PR is merged they are squashed and you only get one commit for the whole feature on the main branches
If an article has interesting content (which comes from an human) and the LLM is just used to help the author finish off the article, I don't have any problem with that.
Labeling both scenarios in the same category feels completely wrong to it, as equating vibe coded stuff (as in no human ever read the produced code) and agent-assisted good old software engineering
You literally change a couple of env variables and you are done, your user experience is basically the same. I can try new models for an hour and be sure I can go back to the original model as quickly if I want.
That is not the case for the software you talked about. They all require way higher switching effort with more perceived risk.
It is a fundamentally hard problem to solve
Because to me and to the very vast majority of git users it is totally irrelevant.
It is nice that the info is available, but the more sane default would be to hide under a verbose flag not the other way around.
Imagine typing cd folder/ and have the whole filesystem subtree be displayed in the terminal. You are free to ignore it, but it is useless and inconvenient nonetheless
Have you tried adding this instruction to your agents.MD? Avoiding situations were the agent start running a loop is the main use case of the file for me
An LLM able to structurally separate context and instructions, should logically need separated data to train, and we don't have it.
Moreover, while an equally powerful LLM architecture solving this may exists, there are no guarantees at all that we are able to come up with it in a reasonable timeframe.
Without some signals moving in that direction, the most pragmatic and realistic way of looking at the problem is that it will not be solved in the near future
Like in the banking world, you can make everything super authenticated, but if you have an API that receives the latest wire transfer YOU received with the message attached, you don't control the message content and it can be an attack vector.
Being authenticated/authorized is not the solution, it is data that the user can access.
The best solution, from an efficiency point of view, is to use smaller models on datacenters, requiring much less of them.
If all relevant open source projects close up their contributions, you can't enter the project anymore from an external point of view.
Almost all open-source public figures started by being interested in a project and submitting PR to it, until eventually either joining the project as core mantainer or creating a separate open source project. The path is now closed, and I don't see a way in, outside of creating a popular open source yourself
Most non technical people I know asked questions to Google even before the AI overview. Instead of looking for the answer in seo-bloated articles, they find it in the overview.
I think google should improve in detecting the kind of query when I need a link that I don't remember, and deactivate the overview on those. If I search for "ryanair booking" I clearly need the url for booking a Ryanair flight, AI overview is useless
Most of the time I'm looking for something very specific that there are plenty of articles about, but clicking on the articles results in popups, banners and an unhealthy amount of scrolling to get to the answer.
AI overview provides me the answer instantly.
Think about suff like "does china borders afghanistan". In those cases you can be confident that the AI overview is right, and saved you time.
If it is a complex or niche question I tend not to trust the overview and go straight for legitimate-looking results
In other cases the content is at easily guessable path, and that is a whole different story