Linear layer with same weights, biases, and inputs gives different output than Pytorch about candle

[LaurentMazare]

I don't think there are much guarantees on the values being exactly equal on both sides. Especially when using bfloat16, there are only 8 bits for the mantissa so if the error is of the order of a percent that would be somewhat expected.
Things that could be worth checking:

• How much is the error when using f32 on both sides?
• What happens without biases, the error being in the same column might indicate that this is the culprit and it may be caused by pytorch fusing the add and mul part whereas candle doesn't do it for now.
• Are the f32 to bf16 conversions in line between both sides? (using a safetensors rather than npy would allow storing the bf16 tensors)

[EricLBuehler]

Thank you for giving me those tips, I think I figured out what the problem is:

How much is the error when using f32 on both sides?

About the same, there is no change to the output.

What happens without biases, the error being in the same column might indicate that this is the culprit and it may be caused by pytorch fusing the add and mul part whereas candle doesn't do it for now.

It turns out that I only do a matmul and then add the bias separately, the 2 tensors are the same in the output! So when we folow the xW^T + B strictly (as Candle does) in Pytorch, the output is the same.

Would it be possible to add this fusion of add and mul to Candle too, as it would fix this? Alternatively, is there something which I can do to fix this? Thank you so much!

[EricLBuehler]

I wrote some code to test fusion using cuBLASLT with a FusedLinearBias layer: https://github.com/EricLBuehler/mistral.rs/blob/44e8a2291d6d53fa125907925c0a4cc613cb8855/mistralrs-core/src/layers.rs#L401-L451

This gets rid of the error. Would you be interested in me submitting a PR to add the fused-linear-bias support directly to Linear?