 Tri Dao
|
d5893f3c74
Merge branch 'main' into changes_for_fp8
|
há 5 meses atrás |
 Ying Zhang
|
dfe1a59e4b
Add var-seq-len to FA3 fp16 / bf16 fwd (#1072)
|
há 5 meses atrás |
 Cameron Shinn
|
cb516f855b
Remove torchlib dependency from cpp files (#1083)
|
há 5 meses atrás |
 Ganesh Bikshandi
|
df66e974bc
fixed odd-seq-len-k.
|
há 5 meses atrás |
|
Ganesh Bikshandi
|
63e9277199
change to correct tile size for headdim=128.
|
há 5 meses atrás |
|
Ganesh Bikshandi
|
fe4c5b59df
undid clang formatting.
|
há 5 meses atrás |
|
Ganesh Bikshandi
|
cdc966e81a
adding files for fp8 changes.
|
há 5 meses atrás |
 Tri Dao
|
74b0761ff7
[FA3] BF16 forward
|
há 5 meses atrás |
|
Tri Dao
|
7f67966cc7
FA3 initial code release
|
há 5 meses atrás |
