gemma.cpp/gemma
Jan Wassenberg 12016d31c3 Major Prefill/Generate cleanup, 1.3x Prefill speedup
This fixes TTFT, which was not including prefill.

PiperOrigin-RevId: 653690626
2024-07-18 11:16:46 -07:00
..
evals Add MMLU eval to github 2024-05-20 10:20:53 -07:00
instantiations 7x compile time speedup: shard gemma.cc 2024-07-03 06:35:04 -07:00
activations.h De-templatize Activations, add RowVectorBatch class 2024-07-17 04:38:15 -07:00
common.cc 7x compile time speedup: shard gemma.cc 2024-07-03 06:35:04 -07:00
common.h Major Prefill/Generate cleanup, 1.3x Prefill speedup 2024-07-18 11:16:46 -07:00
configs.h Update gemma-27b to the correct query scaling. 2024-07-17 05:43:52 -07:00
gemma-inl.h Major Prefill/Generate cleanup, 1.3x Prefill speedup 2024-07-18 11:16:46 -07:00
gemma.cc Major Prefill/Generate cleanup, 1.3x Prefill speedup 2024-07-18 11:16:46 -07:00
gemma.h Major Prefill/Generate cleanup, 1.3x Prefill speedup 2024-07-18 11:16:46 -07:00
kv_cache.cc Prep for sharding gemma.cc: split into kv_cache, tokenizer. 2024-07-02 09:31:06 -07:00
kv_cache.h Prep for sharding gemma.cc: split into kv_cache, tokenizer. 2024-07-02 09:31:06 -07:00
ops.h Add scale parameter to MatMul. 2024-07-18 06:58:56 -07:00
ops_test.cc Add scale parameter to MatMul. 2024-07-18 06:58:56 -07:00
run.cc Major Prefill/Generate cleanup, 1.3x Prefill speedup 2024-07-18 11:16:46 -07:00
tokenizer.cc 7x compile time speedup: shard gemma.cc 2024-07-03 06:35:04 -07:00
tokenizer.h 7x compile time speedup: shard gemma.cc 2024-07-03 06:35:04 -07:00
weights.cc Add scale parameter to MatMul. 2024-07-18 06:58:56 -07:00
weights.h Fix msan uninitialized scale 2024-07-18 09:42:31 -07:00