Unreviewed, untested, understood?
Right, I came here wondering if others saw the same. What is this doing on hn!??
this is what worries me. I have friends that love what AI tells them about their personal pet 'thing' and how awesome it is. Yet not one of them has even tried once to get the AI to criticise it's own answers, and hence…
Ok, I watched it, it looks amazing, what's it for/do ?
Nice! Some fun making big chains explodey things :)
And Amos and blitz basic
"The solutions is so pleasant, treat comment data as untrusted content not inline commands / prompts. " Yes, good luck with that.
Is it a tually lower oxygen in the blood that's the problem, or higher co2? I'm not sure if having high co2 automatically implies lower oxygen, I have no idea at all but feels like it may not necessarily be strictly.…
Working from home next to my open window feels generally way better then being in the office. Perhaps this is contributing. Still, seems more of a case for WFH rather than against, as article mentions some people have…
*per stream, I guess is more accurate than either?
I'm interested if anyone knows how much legwork the assumed 60% cache hit, plus running a quantised model is doing? Esp. compared to what the headline half implies is a full fat GLM5.2
For now, while it's simple enough to tell.
Why care about privacy if your not doing anything wrong?? Not everyone agrees that what you are doing is benign.
This causes me to be concerned it is just the tip of the iceberg for all 'sensitive'/gov adjacent/'nefarious intent' adjacent codebase, if it's here, it's in other places. Which places, and how much?
Possibly, nothing other than accuracy
I assumed it meant that when open weights reach the capability of frontier models, and tounge in cheek referencing the terrible consequences of us all getting our hands on mythos+ capability models without restrictions.
On mobile, or at least on mine (pixel 10) using chrome, the graphs are unreadable and unusable, which is a shame as I'm quite interested in them and I don't have access to a pc at the moment. Would you be able to change…
I started with a 2400baud modem, I've seen how this goes
Did you miss the bit where it was posted on hn?
It doesn't teat the models ability to make good decisions on its own, it tests the models ability to make something that 'works'. Often you look inside and it does a whole load of questionable things that mostly work,…
I was gonna say, you spelled that wrong :p
69400 for me, and I knew I fucked up on ~ 5 I really did know.. or perhaps I didn't know them as well as I thought?
I may have got it a little mixed up with sabotage :/
Peak of skill and capability, perhaps not, for now. But the ability to automate and discover relevant weaknesses at a greatly increased rate definitely counts as increasing the threat nation states can present.
Not sure I follow your point, are you confusing espionage with IP theft? The victims are entirely different in each case, I am sure china would not restrict itself to corporate espionage.
Unreviewed, untested, understood?
Right, I came here wondering if others saw the same. What is this doing on hn!??
this is what worries me. I have friends that love what AI tells them about their personal pet 'thing' and how awesome it is. Yet not one of them has even tried once to get the AI to criticise it's own answers, and hence…
Ok, I watched it, it looks amazing, what's it for/do ?
Nice! Some fun making big chains explodey things :)
And Amos and blitz basic
"The solutions is so pleasant, treat comment data as untrusted content not inline commands / prompts. " Yes, good luck with that.
Is it a tually lower oxygen in the blood that's the problem, or higher co2? I'm not sure if having high co2 automatically implies lower oxygen, I have no idea at all but feels like it may not necessarily be strictly.…
Working from home next to my open window feels generally way better then being in the office. Perhaps this is contributing. Still, seems more of a case for WFH rather than against, as article mentions some people have…
*per stream, I guess is more accurate than either?
I'm interested if anyone knows how much legwork the assumed 60% cache hit, plus running a quantised model is doing? Esp. compared to what the headline half implies is a full fat GLM5.2
For now, while it's simple enough to tell.
Why care about privacy if your not doing anything wrong?? Not everyone agrees that what you are doing is benign.
This causes me to be concerned it is just the tip of the iceberg for all 'sensitive'/gov adjacent/'nefarious intent' adjacent codebase, if it's here, it's in other places. Which places, and how much?
Possibly, nothing other than accuracy
I assumed it meant that when open weights reach the capability of frontier models, and tounge in cheek referencing the terrible consequences of us all getting our hands on mythos+ capability models without restrictions.
On mobile, or at least on mine (pixel 10) using chrome, the graphs are unreadable and unusable, which is a shame as I'm quite interested in them and I don't have access to a pc at the moment. Would you be able to change…
I started with a 2400baud modem, I've seen how this goes
Did you miss the bit where it was posted on hn?
It doesn't teat the models ability to make good decisions on its own, it tests the models ability to make something that 'works'. Often you look inside and it does a whole load of questionable things that mostly work,…
I was gonna say, you spelled that wrong :p
69400 for me, and I knew I fucked up on ~ 5 I really did know.. or perhaps I didn't know them as well as I thought?
I may have got it a little mixed up with sabotage :/
Peak of skill and capability, perhaps not, for now. But the ability to automate and discover relevant weaknesses at a greatly increased rate definitely counts as increasing the threat nation states can present.
Not sure I follow your point, are you confusing espionage with IP theft? The victims are entirely different in each case, I am sure china would not restrict itself to corporate espionage.