Chirag Jain 50896ec574 Make nvcc threads configurable via environment variable (#885) há 9 meses atrás
..
fused_softmax.cpp ed553e9238 Add Megatron attention implementation for benchmarking há 2 anos atrás
scaled_masked_softmax.h ed553e9238 Add Megatron attention implementation for benchmarking há 2 anos atrás
scaled_masked_softmax_cuda.cu ed553e9238 Add Megatron attention implementation for benchmarking há 2 anos atrás
scaled_upper_triang_masked_softmax.h ed553e9238 Add Megatron attention implementation for benchmarking há 2 anos atrás
scaled_upper_triang_masked_softmax_cuda.cu ed553e9238 Add Megatron attention implementation for benchmarking há 2 anos atrás
setup.py 50896ec574 Make nvcc threads configurable via environment variable (#885) há 9 meses atrás
type_shim.h ed553e9238 Add Megatron attention implementation for benchmarking há 2 anos atrás