Tri Dao 74b0761ff7 [FA3] BF16 forward hace 5 meses
..
cutlass @ 756c351b49 74b0761ff7 [FA3] BF16 forward hace 5 meses
flash_attn 40e534a7f6 Implement cache_leftpad hace 5 meses
ft_attention 50896ec574 Make nvcc threads configurable via environment variable (#885) hace 9 meses
fused_dense_lib 50896ec574 Make nvcc threads configurable via environment variable (#885) hace 9 meses
fused_softmax 50896ec574 Make nvcc threads configurable via environment variable (#885) hace 9 meses
layer_norm 50896ec574 Make nvcc threads configurable via environment variable (#885) hace 9 meses
rotary 50896ec574 Make nvcc threads configurable via environment variable (#885) hace 9 meses
xentropy 50896ec574 Make nvcc threads configurable via environment variable (#885) hace 9 meses