Commit Verlauf

Autor SHA1 Nachricht Datum
  Ying Zhang 496fdc4f6c Add seqused_q in fwd / bwd and seqused_k in bwd. vor 5 Monaten
  jayhshah c92ca63268 FA3 FP8 qkv descales + restore max offset for h128 causal + added sync for producer WG (#1173) vor 5 Monaten
  Tri Dao bafe253042 [FA3] Bwd vor 6 Monaten
  Tri Dao 3aae9c18c1 Revert "Changes For FP8 (#1075)" vor 6 Monaten
  ganeshcolfax 1899c970c8 Changes For FP8 (#1075) vor 6 Monaten
  Ying Zhang dfe1a59e4b Add var-seq-len to FA3 fp16 / bf16 fwd (#1072) vor 6 Monaten
  youkaichao ef3e358a25 remove lambda (#1056) vor 6 Monaten
  Tri Dao 7f67966cc7 FA3 initial code release vor 6 Monaten