Steve Gibson
speaker
17,616 appearances
14 recordings
1 series
first heard May 2026
last heard 2 Sep
Steve Gibson’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsRecordings per month over the last 12 months — 14 in all, peaking in Jul 2026 with 5.
Appearances
So
We're able to give them
As a consequence, what appears to be a personality, many forms of behavior, and also instruct them what they may and may not do.
So that's the good news.
The bad news, as it turns out, is that any behavior
like this that can be easily imprinted can also be easily removed.
In the summer of 2024, researchers at ETH Zurich, uh the University of Maryland Anthropic and MIT published a paper titled Refusal in Language Models is mediated by a single direction.
And here, direction is a is a term of art in uh neural networks.
we'll see.
So the abstract of their paper employs some of you know of this inside baseball terminology, but everyone should be able to easily get the gist of it.
So
And I'll explain a bit more afterwards.
The paper's abstract, that is, refusal in language models is mediated by a single direction.
The abstract reads: conversational language models are fine-tuned for both instruction following and safety.
Safety meaning not going to tell you that.
Resulting in models that obey benign requests but refuse harmful ones.
While this refusal behavior is widespread across chat models, its underlying mechanisms remain poorly understood.
Again, this was uh summer of twenty twenty four, so about just about two years ago, this paper appeared.
In this work, across 13 popular open source chat models, up to 72 billion parameters in size, we show that refusal
is mediated by a one-dimensional subspace.
Showing 2041–2060 of 17,616 · page 103 of 881
← Previous
Next →