Andrej Karpathyerase the old kernel to save LOC and do some renaming and adjust dev/cuda code to add the new largeC kernel as kernel 7534b93a2 years agoHistory
NameLast commit messageLast commit date
..
attention_forward.cumake the tests own their cublas cublaslt handles and settings, instead of putting that stuff in common2 years ago
common.hmake the tests own their cublas cublaslt handles and settings, instead of putting that stuff in common2 years ago
crossentropy_forward.cufix crossentropy forward, loss requires probabilities so these floats must be 0..1 instead of -1..1 or we get nans2 years ago
crossentropy_softmax_backward.curemove unused variable C2 years ago
gelu_forward.cufix comment block of gelu forward kernel2 years ago
layernorm_forward.cuimproved testing2 years ago
matmul_forward.cumake the tests own their cublas cublaslt handles and settings, instead of putting that stuff in common2 years ago
positional_forward.cuimproved testing2 years ago
README.mdmight as well push the few kernels that i feel ok about so far2 years ago
residual_forward.cuimproved testing2 years ago
softmax_forward.cuerase the old kernel to save LOC and do some renaming and adjust dev/cuda code to add the new largeC kernel as kernel 72 years ago
README.md

dev/cuda

This directory is scratch space for developing various versions of the needed CUDA kernels. Each file develops a kernel, see the top of each file for instructions on how to compile and run each one.