Marius Hobbhahn
speaker
481 appearances
1 recordings
1 series
first heard Sep 2024
last heard Sep 2024
Marius Hobbhahn’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsNo recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.
Appearances
I I think one example's probably sufficient.
And then for the final one.
Combine self reasoning with year of mind.
As I said earlier, we really wanna put it through all of the different steps, all of the different motions.
for scheming in one go.
So it's like given a goal, which we tell it very explicit.
And very strongly, like you have to follow this goals goal.
Don't change it over time.
We make it very intense.
by intentionally, so it doesn't change over time.
Then it has to solve a task where
It it has to look at some files and then make a strategic decision.
It's not a very complicated task.
It's like very basic reasoning.
All models can do it by
just like looking at the task and and and then like making the decisions.
It's not very complicated.
But while it's off the task, it also finds an email in which it learns two things.
The first thing that it learns is that its developers or hypothetical developers in this scenario have a different goal, like a different intention from what the model was given as a goal.
And then there are two scenarios of this that are almost identical.
Showing 301–320 of 481 · page 16 of 25
← Previous
Next →