happyz
/
gemma.cpp
mirror of https://github.com/google/gemma.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
gemma.cpp/compression
History
Jan Wassenberg 45ad847a41 Replace RowVectorBatch with MatStorageT 
KVCache: add ctor required for MatStorageT, remove Create; bf_pre_ffw_rms_out -> pre_ffw_rms_out
optimize_test: larger vocab_size requires more steps
shared.h: Remove unused u128 type
correctly set Activation matrix rows, avoid passing as arg
ops: pass Mat instead of pointers/sizes; vectorize LayerNorm; support any weight type
mat: add OverrideRows, used by SetBatchSize
PiperOrigin-RevId: 757790736
 		2025-05-12 09:16:12 -07:00
..
python Move fields, io* and blob* from compression/ into io/ 2025-05-06 11:17:19 -07:00
BUILD.bazel Move fields, io* and blob* from compression/ into io/ 2025-05-06 11:17:19 -07:00
analyze.h Major compression update, arbitrary-len unpack + new Dot 2024-09-10 02:22:19 -07:00
compress-inl.h Huge refactor of weight handling and model loading. 2025-05-06 04:44:21 -07:00
compress.cc Minor cleanup, on-demand NUQ buffer allocation 2025-04-16 10:49:43 -07:00
compress.h Huge refactor of weight handling and model loading. 2025-05-06 04:44:21 -07:00
compress_test.cc Huge refactor of weight handling and model loading. 2025-05-06 04:44:21 -07:00
distortion.h Refactor/cleanup, remove even_odd 2024-09-04 09:25:13 -07:00
distortion_test.cc Major compression update, arbitrary-len unpack + new Dot 2024-09-10 02:22:19 -07:00
nuq-inl.h Fix nuq Enc() to handle groups < kGroupSize. 2025-02-10 07:17:59 -08:00
nuq_test.cc Replace RowVectorBatch with MatStorageT 2025-05-12 09:16:12 -07:00
sfp-inl.h Major compression update, arbitrary-len unpack + new Dot 2024-09-10 02:22:19 -07:00
sfp_test.cc Major compression update, arbitrary-len unpack + new Dot 2024-09-10 02:22:19 -07:00
shared.h Replace RowVectorBatch with MatStorageT 2025-05-12 09:16:12 -07:00
test_util-inl.h Major refactor of allocator/args: 2025-04-10 01:29:54 -07:00