Nathan Lambert

speaker

1,013 appearances 1 recordings 1 series first heard Nov 2024 last heard Nov 2024

Nathan Lambert’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

No recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.

newest first · ▶ plays the moment

Let me get their name right so it's easier for people.

Um people have talked about Quiet Star, that's a bit more complicated.

I think Trice is also one, which is also a bit uh hard to understand, but motivated by the same thing.

So the literature is going in that direction, but we kind of showed like look, you can do this in your preference tuning.

You don't have to just do a math model.

Like you can just add on these types of RL at the end and it doesn't blow the model up.

So to kind of summarize, it's like as we go, we've unlocked uh we have these three stages, SFT, DPO, and RL.

And at each of them we're kind of

Taking more alpha from what industry is doing.

Like SFT is known, it works.

DPO, there's some new tricks that we needed to unscover, uncover parameters, different settings.

We use this with like a length normalized setting.

Um, but that's mostly known but good information.

But the RL stuff is like, look, this is like real.

We're beating Lama numbers.

We used RL like

I it's still mind blowing to me that RL is becoming more relevant.

I think like twenty twenty three is like, uh, where's R L H F going?

But now it's not even R just RLHF.

Like we didn't even have a reward model in this.

Showing 201–220 of 1,013 · page 11 of 51