IanCal

40 ms·

751.
by IanCal 2y ago
I think it's hard to make the case for anything other than a pretty tiny group or organisation that that you can get away without having some reporting and moderation process. I don't think you need a report button but a known way
752.
by IanCal 2y ago
It's important to understand that the act we're talking about does not make owners simply liable for stuff that happens on their sites, nor does it require them to stop everything. It's about what the risks are of these thi
753.
by IanCal 2y ago
Putting in mitigations relevant to your size, audience and risk factors is not "no real effect". If you've never considered what the risks are to your users, you're doing them a disservice. I've also not defended it
754.
by IanCal 2y ago
Right or wrong I think many have misread the legislation or read poor coverage of it given people's reasoning. Much of things boils down to doing a risk assessment and deciding on mitigations. Unfortunately we live in a world where if
755.
by IanCal 2y ago
Isn't cold storage about where the keys are? You still need to be able to actually interact with a chain.
756.
by IanCal 2y ago
I got good responses from some simply by telling it to be aware of its limitations. And my first test with asking for the episode of Gilligan's Island sort of worked with sonnet, no prefix and no system prompt, temp 0. It got the episo
757.
by IanCal 2y ago
It's told it can search for tools and it can enable them. It is not told initially that there's a calculator, it asks for one.
758.
by IanCal 2y ago
It doesn't always choose to do that though, it doesn't do it for simpler questions. > So what we would expect to see with a LLM without tools enabled, is that it suggests that you give it access to a calculator. If I ask sonnet
759.
by IanCal 2y ago
What's needed for you to call something intuitive? I gave sonnet the ability to search for tools, and it chose to look for a calculator which I then gave it (when asked a hard maths problem). Then it used it. There's no prompting
760.
by IanCal 2y ago
I gave sonnet a hard arithmetic problem, and the ability to look for tools. It looked for a calculator, I gave it one and it used that. https://gist.github.com/IanCal/2a92debee11a5d72d62119d72b965...
761.
by IanCal 2y ago
But if you give them a search they'll use that to look for tools and if you give them the ability they can write new tools for themselves.
762.
by IanCal 2y ago
How have I told it there's a calculator? All I've given it is the ability to search for tools and enable ones it wants. > Of course the model will use the calculator you've explicitly informed it of I didn't. I also g
763.
by IanCal 2y ago
> So the real question behind the headline is why LLMs don't learn to ask for a calculator by themselves They can. They're sometimes a bit cocky about their maths abilities, but this really isn't hard to test or show. htt
764.
by IanCal 2y ago
Consumer protections in the UK and EU means they have to last a reasonable length of time, and if not you get a proportional refund (or repair, or replace). I've had refunds from devices several years old.
765.
by IanCal 2y ago
The point is that you can achieve a lot without being able to access every position and the human arm having a lot of degrees of freedom doesn't necessarily mean it's more capable than a different arm with fewer.
766.
by IanCal 2y ago
Humans have 7 dof but we can't get into all positions, as I've found clearly today trying to reach things for my kids in the back seat of the car.
767.
by IanCal 2y ago
> local JS requests from internet I really don't understand what this means.
768.
by IanCal 2y ago
As usual, this sounds great for the nice players. Unfortunately that means if you download an html file and double click on it then it can send anything in your downloads folder to anywhere else.
769.
by IanCal 2y ago
That requires having an actual server though.
770.
by IanCal 2y ago
4 tokens per second is pretty slow. That's 5-10s for a comment the length of yours (and R1 specifically likes to output a lot of tokens). It's 10-20x slower than many top end models, which are available cheaply. Even high cost v
771.
by IanCal 2y ago
This is why my go-to estimate for many things where the result is outlandishly large is "at least six".
772.
by IanCal 2y ago
Does one note have some way of adding scripted items to automate some of the steps?
773.
by IanCal 2y ago
Often these are setup scripts or similar for a codebase, so you might have a readme with instructions like 1. Update the submodules by running ... 2. Install volta 3. Run `volta ...` 4. Setup a user account 5. Sync the dev database by runni
774.
by IanCal 2y ago
I absolutely agree you modify and automate parts (and eventually all) of it. It's the leap to having scripts in a repo being called from production services I'm less convinced by. As I said, the alternatives to part/increment
775.
by IanCal 2y ago
> The limitation of your approach I will say is that if you can automate the beginning of a process or the end it’s fine, but if you have two islands in the middle it doesn’t work better than a CLI tool with prompts I think I may have po
776.
by IanCal 2y ago
I'm confused as to what you're picturing. > The moment those scripts start getting called by production services Any of them with manual steps just can't, right? > Further, because your human oriented process has now co
777.
by IanCal 2y ago
You also get to a point that's functional but not optimal very quickly. Improvements can then be added incrementally by different people as and when it's worthwhile.
778.
by IanCal 2y ago
Then automate those points as and when it's worth doing. Nobody is saying you can't do that, in fact it's a very key selling point of a do nothing script - you can very easily upgrade it to a do something script. The altern
779.
by IanCal 2y ago
You could do this with a nicer cli library, so you could show the checklist and running output of each stage, which would be nice regardless of what steps are automated. My main issue with that is that a do nothing shell script is so easy t
780.
by IanCal 2y ago
I'm a big fan of this approach. In general it's another way of defining interfaces around processes. Those processes may be manually done or automated. But the interface can remain the same - this is quite powerful when it comes

More ›