Conceptloss correlation
Loss Correlation
The Pearson correlation coefficient between two loss curves during training, used to measure the statistical consistency of training trajectories under different hardware or implementations, where a value closer to 1 indicates higher training fidelity.
Timeline (last 90 days)
Oct 16
Loss correlation refers to the Pearson correlation coefficient between the two loss curves during training; 0.99 means that at each training step the NPU-side loss value rises and falls almost in sync with the GPU side.
Unverified50%