Jan Wassenberg e890d46f30 1.31x batch prefill, 1.24x batch decode speedup: NUMA binding ... Only the weights; binding MatMul output worsens batch=1 prefill. Update gemma_batch_bench to use --decode_qbatch. Fix/remove prefill_activations in gemma-inl.h. Refactor: use BasePageBytes directly when binding Move BindB/C to .cc by de-templatizing Remove MatOwners::AllocateFor because it is weights-specific (binding or not) Disband MatOwners, replace with vector PiperOrigin-RevId: 759610477		2025-05-16 07:42:13 -07:00
..
benchmark.cc	Replace RowVectorBatch with MatStorageT	2025-05-12 09:16:12 -07:00
benchmark_helper.cc	Minor: rename compression/shared -> types.h	2025-05-13 06:53:21 -07:00
benchmark_helper.h	Rename-only: remove Allocator2 etc suffixes now that refactoring is complete	2025-05-06 09:12:43 -07:00
benchmarks.cc	Introduce QueryResult in GemmaEnv and add a shortcut for WrapAndTokenize.	2024-10-14 04:45:21 -07:00
cross_entropy.cc	Huge refactor of weight handling and model loading.	2025-05-06 04:44:21 -07:00
cross_entropy.h	Huge refactor of weight handling and model loading.	2025-05-06 04:44:21 -07:00
debug_prompt.cc	Move fields, io* and blob* from compression/ into io/	2025-05-06 11:17:19 -07:00
gemma_batch_bench.cc	1.31x batch prefill, 1.24x batch decode speedup: NUMA binding	2025-05-16 07:42:13 -07:00
gemma_test.cc	Minor: mark command line flags as required	2025-05-12 08:30:44 -07:00
prompts.h	Benchmark gemma.cpp with different length inputs.	2024-10-10 15:59:26 -07:00
run_mmlu.cc	Move fields, io* and blob* from compression/ into io/	2025-05-06 11:17:19 -07:00