Comment by nl
8 days ago
I'm all for open models, but people seem to misunderstand what they are. They aren't the same thing as open source code!
> open weights, open code and open data
Even if you have all these things you still can't replicate a model because of randomness.
You can backdoor a model with less than 1000 examples and it is impossible to detect.
Yeah - we also don't know if the models from OpenAI and anthropic are back-doored either.
You don't want to replicate the exact model, you want to build a system of similar capabilities.
Great, but that seems a different concern to the auditability of a model.
You can take the code for Kimi K3 now, take the training framework from Prime and the data from Olmo, spend some money on RL environments and some more money (!) on GPU training and end up with a system of similar capabilities.
But that's completely different to being able to audit Kimi K3. Even if you had the exact code, data and training environments it is impossible to verify that the model you have came from that.
Deterministic seed
Deterministic seeds barely work on a single machine, small scale training run.
They just don't work at all on a many month long, 100K+ GPU cluster training run.
While using floating point? Not happening. You'd have to switch to fixed point, not just for the models themselves but also _all_ the training code (ie backprop).
Even then you'd still need to account for order of events when an entire cluster of GPUs is involved. Also don't forget to account for any synthetic data sources. Or even non-synthetic for that matter - does your pipeline do any image resizing on the fly? Better make sure that's fully deterministic between machines (it almost certainly won't be).
It's theoretically possible but I don't expect it to materialize any time soon.
> theoretically possible
I mean I guess, but not in a performant way if there are ever any hardware failures. And with 100K GPUs there are multiple hardware failures per day.