gemma.cpp/io
Jan Wassenberg 9efdcfd45c 1.07x batch decode speedup: more BF16 weights and activations
BF16 att_sums and ffw_out
Support BF16 B views without decompression
Support arbitrary types in MulByConstAndAdd, AddFrom

Also update profiler annotations in ops-inl.h

PiperOrigin-RevId: 766995010
2025-06-03 23:30:18 -07:00
..
BUILD.bazel 1.07x batch decode speedup: more BF16 weights and activations 2025-06-03 23:30:18 -07:00
blob_compare.cc 3.8x speedup of weights loading via preadv on Linux 2025-05-15 11:55:15 -07:00
blob_store.cc Fix RowT issue and improve Griffin (currently still broken) 2025-05-19 07:02:10 -07:00
blob_store.h Major refactor to de-templatize gemma-inl and weights 2025-06-02 23:01:35 -07:00
blob_store_test.cc 3.8x speedup of weights loading via preadv on Linux 2025-05-15 11:55:15 -07:00
fields.cc Move fields, io* and blob* from compression/ into io/ 2025-05-06 11:17:19 -07:00
fields.h Move fields, io* and blob* from compression/ into io/ 2025-05-06 11:17:19 -07:00
fields_test.cc Move fields, io* and blob* from compression/ into io/ 2025-05-06 11:17:19 -07:00
io.cc 1.16x decode speedup: remove last MatVec in Attention 2025-06-02 09:40:29 -07:00
io.h 3.8x speedup of weights loading via preadv on Linux 2025-05-15 11:55:15 -07:00
io_win.cc Cleanup: remove unused kCyclic, remove 2 suffix 2025-05-13 01:06:41 -07:00
migrate_weights.cc Move fields, io* and blob* from compression/ into io/ 2025-05-06 11:17:19 -07:00