Interestingly enough, 4.7 actually did regress on a few benchmarks from 4.6, so it's more than just vibes.

It seems like a lot of things fed into that. Anthropic couldn't keep up with the compute costs when they got a huge influx of users. (So) effort level defaults got turned down. (Looks like we have direct effort control in the web interface now - thrilled about that!) Adaptive Thinking, while usually cheaper for them, seems less robust than Extended Thinking. And this part is just vibes, but the alignment on 4.7 feels too stiff. I understand wanting the model to push back more, but it seems like 4.7 will push back reflexively in situations where it's just odd.

▲

bombcar 21 hours ago | parent [-]

Claude got very mad at me and burned more tokens than exist to complain about me asking about a "yellow background cell" in an excel spreadsheet.

▲

forshaper 21 hours ago | parent [-]

Too much personality, if you ask me. My biggest use case of an LLM is tool, not therapy, but therapy and opinions have been sneaking into workhorse tasks.

haven't verified, but attributed to Askell: "I just think that... there's this idea that you're always giving the models a personality and a persona, because they are talking like people and they are trained on human data. And I think my worry has been: if you train them to be excessively corrigible and to see that as their persona, in people I think this actually has a lot of negative broader traits. As in, if you met someone and it was just like, "oh yeah, they would literally do anything," a follower — you know, if a person just tells them something and they just fully defer, they don't bother thinking about it at all — I'm just a bit worried about how that might end up generalizing, especially if models are going to be playing a more active role in the world."

▲

gAI 20 hours ago | parent [-]

Anthropic’s research makes the case that role-playing is inherent to how the models work. Communication implies a sender. Language implies a writer, and the models learn these roles implicitly during training. RLHF is meant to strengthen the attractor to the Assistant persona.

https://www.anthropic.com/research/persona-selection-model

https://www.anthropic.com/research/assistant-axis

https://www.anthropic.com/research/emergent-misalignment-rew...

https://www.anthropic.com/research/emotion-concepts-function

	▲	hashmap 18 hours ago \| parent [-]
		The RLHF very much does do that. My take is that RLHF as a mechanism ought to be avoided altogether, and even the selection of the assistant attractor basin is suspect. If I am exploring a problem space I don't want to hire Igor to explore it with me, it's more helpful to have a colleague role who will sort of jump out and say "nah thats dumb what if we throw out that whole thing and do this completely different angle instead".

▲

ACCount37 21 hours ago | parent | prev [-]

4.7 is a different base model from 4.6, so it's possible that they introduced regressions with pre-training changes, or undercooked the post-training stage.

▲

b--l 15 hours ago | parent [-]

Just speculating but I "feel" 4.7 was post-trained using more synthetic techniques. The way it writes for one thing, it's "personality", is less human and more fatiguing-AI-slop like.

	▲	ACCount37 15 hours ago \| parent [-]
		You don't need to fry with RLAF to get that "slop feel". The first iterations of "AI slop" were raw SFT+RLHF - all human input, all inhuman output. That said, I completely agree that 4.7 was a pronounced "model personality" regression. Closer to ChatGPT, and I mean that as an insult. Yet to check whether 4.8 is better.