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 monthsNo recordings in the last 12 months.Older appearances are listed below; set an alert to hear about the next one.
Appearances
Like and at this point like, oh our PPO's not as good.
Maybe it's like half that we we don't really know how to train a reward model 'cause we haven't been focusing on it as much.
But also half of just like it's it's harder.
So I do think that it's
It's funny.
It's such a great, silly debate.
It just never goes away.
on our version in two loot three, and it's like two loot two versus two loot two point five, which is DPO to PPO, is like one percent.
And then on our valves like two loot two to two loot three is like fourteen percent, which is all just like data curation and process and stuff like this.
So it's like okay, like there's a there's your number.
Like if you want to go off and be an algorithmic like academic, like your most cases you're gonna be biting one to two percent.
versus making really specific data where you care about where you can get like 10x.
I don't know.
I I'm not surprised.
That's probably what industry does too.
Like these post trading teams are like
You're the person generating Python code data.
You are a full time Python code instruction and preferences and prompts um and filtering
lead and you do it and you generate crazy good data on this one really, really specific thing that may or may not be in the actual eval suite.
It's like that's like even us we aren't doing like we have people on data, but it's not like they're spending a month on this one really specific thing.
Showing 541–560 of 1,013 · page 28 of 51
← Previous
Next →