happyz
/
gemma.cpp
mirror of https://github.com/google/gemma.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
434 Commits 20 Branches 5 Tags 35 MiB
Commit Graph

32 Commits

Author SHA1 Message Date
Jan Wassenberg 4154f5a910 Document Gemma 2 model names 
PiperOrigin-RevId: 659858832
 2024-08-06 01:44:15 -07:00
Jan Wassenberg f9b390b134 Support all weight types in a single binary. 
This changes the command line flags, but the default value retains the previous behavior.

Also add a CreateGemma helper to enable extra args without interface changes.

PiperOrigin-RevId: 641266411
 2024-06-07 09:04:45 -07:00
Jan Wassenberg e3f4374e81 Fix fix for weight type define, refs #198 
GEMMA_WEIGHT_T is indeed the correct flag for the C++ compiler,
but the readme references CMake, and there the correct flag name is WEIGHT_TYPE.

PiperOrigin-RevId: 641170380
 2024-06-07 01:32:25 -07:00
Jan Wassenberg 8dc0e5ea83 Fix reference to GEMMA_WEIGHT_T. Refs #198 
PiperOrigin-RevId: 641161403
 2024-06-07 00:54:30 -07:00
Paul Chang 82623bdc7f Refer to --weights rather than --compressed_weights to simplify CLI docs 
PiperOrigin-RevId: 634391135
 2024-05-16 07:51:49 -07:00
Jan Wassenberg 54120a5571 Mention Makefile contributed by @jart 
PiperOrigin-RevId: 623436818
 2024-04-10 03:21:10 -07:00
zond 9ca662dc14
Clarified README 
Made it more visible that the recurrent weights are at a different Kaggle page.
 2024-04-09 09:58:47 +02:00
Luca Versari 9c3f969405 Implement the Griffin model. 
Also implement support for some model variations:

- Local attention.
- Add support for biases.
- Use RoPE only on half vectors.
- Support different order of QKV weights.

Co-authored-by: Andrey Mikhaylov <amik@google.com>
Co-authored-by: Martin Bruse <zondolfin@gmail.com>
Co-authored-by: Zoltan Szabadka <szabadka@google.com>
 2024-04-08 21:45:54 +02:00
Jan Wassenberg 7122afed5a Add note on weight update and improve error message 
PiperOrigin-RevId: 621849989
 2024-04-04 07:17:27 -07:00
austinvhuang 810b5a0cc2 Update README with more details on contributing code, add experimental/ directory, add READMEs for subdirectories, clean up DEVELOPER notes 2024-03-15 14:10:24 -04:00
Copybara-Service c8b9675898 Merge pull request #74 from osanseviero:patch-1 
PiperOrigin-RevId: 612937722
 2024-03-05 12:49:09 -08:00
Jan Wassenberg bb9b023502 Support Bazel builds. Fixes #16 
Also fix nuq/sfp-inl: warning, cast, and disable SCALAR

PiperOrigin-RevId: 612704056
 2024-03-04 22:07:25 -08:00
Omar Sanseviero 8c857b957e
Update README.md 2024-03-04 12:58:49 +01:00
Omar Sanseviero 86761dc113
Update README.md 2024-03-01 23:44:38 +01:00
austinvhuang 0ea7b993de remove --log fixing https://github.com/google/gemma.cpp/issues/59, improve command line args help, add copybara #include sort guards in more source files, add README sections on running faster and related projects 2024-02-28 15:18:40 -05:00
Dan Zheng afc354dcb1 Import from GitHub. 
PiperOrigin-RevId: 610595796
 2024-02-26 19:05:11 -08:00
Dan Zheng 8db89304bd No public description 
PiperOrigin-RevId: 610498969
 2024-02-26 19:03:48 -08:00
austinvhuang 129e66ada2 Reduce KV cache preallocation to 4096 and make it comptime configurable, add rm build note in readme, add note on comptime options in DEVELOPERS, make multiturn=0 the default 2024-02-26 17:05:32 -05:00
Naoki Kishida 7ab968c957 Copybara import of the project: 
--
26b541b666 by kishida <naokikishida@gmail.com>:

add information for the reseting conversation

COPYBARA_INTEGRATE_REVIEW=https://github.com/google/gemma.cpp/pull/40 from kishida:add_info_for_reset_conv 26b541b666
PiperOrigin-RevId: 610418671
 2024-02-26 08:39:27 -08:00
Dan Zheng 4c155bd3df Restore reverted changes. 
Sync to 84444c93a4.

PiperOrigin-RevId: 610263918
 2024-02-25 19:32:07 -08:00
Silvio Traversaro 696597383c Copybara import of the project: 
--
19694e1f2e by Silvio Traversaro <silvio@traversaro.it>:

Do not pass explicitly -O2 flag to compiler in Release build

COPYBARA_INTEGRATE_REVIEW=https://github.com/google/gemma.cpp/pull/3 from traversaro:patch-1 19694e1f2e
PiperOrigin-RevId: 610096914
 2024-02-24 20:41:33 -08:00
Dan Zheng 84444c93a4 Revert "Copybara configuration update." 
This reverts commit c03b5da542.

Restore lost changes due to improper Copybara syncing.
 2024-02-24 15:15:14 -08:00
Dan Zheng c03b5da542 Copybara configuration update. 
PiperOrigin-RevId: 609931218
 2024-02-24 12:02:47 -08:00
Austin Huang 34b22c56f5
Merge pull request #6 from dcoles/clang-cl 
Allow building on Windows using `clang-cl` toolchain
 2024-02-24 12:27:40 -05:00
Jan Wassenberg af715d2436 Update readme to match code, see cl/609177092 
PiperOrigin-RevId: 609912278
 2024-02-23 22:34:08 -08:00
Jan Wassenberg 8f27580fb6
Merge branch 'dev' into clang-cl 2024-02-24 04:22:42 +01:00
The gemma.cpp Authors 7c9954dea5 Code update 
PiperOrigin-RevId: 609719211
 2024-02-23 07:13:10 -08:00
David Coles ceea495693 Add CMakePresets and update README instructions 
Using a `CMakePresets.json` file makes it much easier to manage
several alternate build configurations, such as the "ClangCL"
build for Windows. This also makes it easier for tools like
VSCode to run CMake-based builds.
 2024-02-23 00:41:47 -08:00
Dan Zheng 6747c24eef Copybara import of the project: 
--
acafb83b5c by Dan Zheng <danielzheng@google.com>:

[readme] Minor tweaks.

COPYBARA_INTEGRATE_REVIEW=https://github.com/google/gemma.cpp/pull/14 from dan-zheng:readme acafb83b5c
PiperOrigin-RevId: 609550615
 2024-02-22 17:17:26 -08:00
Dan Zheng 0508e2c2e1 [readme] Clarify that development is done on dev branch. 
Discussed using dev branch instead of main branch for development.

We could revisit this once CI is hardened to test end-to-end correctness.

PiperOrigin-RevId: 609500897
 2024-02-22 14:32:56 -08:00
Jan Wassenberg fb6f266db1 Fix email addresses 2024-02-21 07:53:56 +01:00
Austin Huang e29cd566cf initial commit 2024-02-21 03:31:22 +00:00