Jesse Gross 6da8b6a879 kvcache: Support non-causal attention ...

Models can disable causality for all or part of their processing
while continuing to store data in the KV cache.

2025-03-07 18:39:27 -08:00

..

cache.go

attention: Remove unnecessary contiguous operations

2025-03-01 20:53:23 -08:00

causal_test.go

kvcache: update tests

2025-03-07 14:08:21 -08:00

causal.go

kvcache: Support non-causal attention

2025-03-07 18:39:27 -08:00

encoder.go

ml/backend/ggml: create tensor on specific backend

2025-03-07 14:08:21 -08:00

wrapper.go

attention: Remove unnecessary contiguous operations

2025-03-01 20:53:23 -08:00