Commit Graph

  • 94edeaf7df
    relocate previous newline token const rabidcopy 2023-03-20 14:35:36 -0500
  • 6b6d5b5024
    Fixed tokenizer.model not found error when model dir is symlink (#325) Qingyou Meng 2023-03-21 03:33:10 +0800
  • a791a68b61
    move file magic/version to header, print expected version (#319) master-a791a68 Mack Straight 2023-03-20 12:26:01 -0700
  • 3eca29ec0d
    dynamically determine newline token rabidcopy 2023-03-20 14:02:48 -0500
  • 330b86eed2
    Make newline token a constant rabidcopy 2023-03-20 13:02:54 -0500
  • e671ad534e
    Merge branch 'ggerganov:master' into master Qingyou Meng 2023-03-21 02:00:56 +0800
  • e949eb518f Makefile: slightly cleanup for Mac Intel; echo instead of run ./main -h mqy 2023-03-21 01:11:25 +0800
  • 0f1b21cb90
    Docker - Fix publish docker image in GitHub Registry (#235) master-0f1b21c Bernat Vadell 2023-03-20 18:05:20 +0100
  • 4fe9734e09
    Improve interactive mode's coherence after EOS rabidcopy 2023-03-20 11:33:50 -0500
  • 0cc13a2193 add alpaca support into docker scripts Bernat Vadell 2023-03-20 16:58:53 +0100
  • 8d39365af6 update license, added backwards compatibility with both ggml model formats, fixed context length issues. Concedo 2023-03-20 23:43:35 +0800
  • 98b31b789c try to fix docker pipeline using as password github_token & username repository_owner Bernat Vadell 2023-03-20 16:32:52 +0100
  • 12280807c3 Remove unnecessary use of std::stringstream. Johnman 2023-03-20 16:31:10 +0100
  • 9d7a993d81
    Change the version to 3.12 nusu-github 2023-03-21 00:27:10 +0900
  • 6242d1ccd5 Cleanup. Johnman 2023-03-20 16:15:00 +0100
  • e9f77473dc
    Update main.cpp tjohnman 2023-03-20 16:10:26 +0100
  • b646ffa1b1 Check for reverse prompt by characters instead of tokens (#292) Johnman 2023-03-20 16:06:58 +0100
  • bc53fc0cd6
    Reflects C/C++ standard individually nusu-github 2023-03-20 23:39:12 +0900
  • d1fa8f5833 Merge remote-tracking branch 'origin/master' into perplexity Gary Linscott 2023-03-20 07:17:13 -0700
  • d824baa37f
    C++11 to C++17 nusu-github 2023-03-20 22:28:47 +0900
  • a2c10e0d2f Merge branch 'master' into concedo Concedo 2023-03-20 20:58:27 +0800
  • ef792ae8bd
    Merge unordered_map/vector changes with trunk updates Fabio Rossini Sluzala 2023-03-20 09:36:21 -0300
  • d91896a069 Fixed tokenizer.model not found error when model dir is symlink mqy 2023-03-20 20:33:58 +0800
  • 03755743cf log distribution after prompt tokens Maël Kerbiriou 2023-03-16 18:58:59 +0100
  • 4547848743 log llama's entropy Maël Kerbiriou 2023-03-20 12:15:13 +0100
  • a6598801ad
    Add initial AVX512 support for dot product on Linux Casey Primozic 2023-03-20 04:15:40 -0700
  • e66962ff63 log file for debug output Maël Kerbiriou 2023-03-20 12:15:01 +0100
  • aa6c2bd5d2 sampler log function Maël Kerbiriou 2023-03-16 18:12:17 +0100
  • abbf7e7a61 new sampler for experimentation. Maël Kerbiriou 2023-03-16 16:52:02 +0100
  • 99f3908c51
    Automatically Generate Releases for Mac, Linux, Windows (#81) Kevin Kwok 2023-03-20 03:41:11 -0700
  • 4aa7d2a3f8 move file magic/version to header, print expected version Mack Straight 2023-03-20 03:32:26 -0700
  • 4655634627 fix typo in comment Mack Straight 2023-03-20 03:22:55 -0700
  • 074bea2eb1
    sentencepiece bpe compatible tokenizer (#252) master-074bea2 Mack Straight 2023-03-20 03:17:23 -0700
  • 649cee5f62 sentencepiece bpe compatible tokenizer Mack Straight 2023-03-17 20:02:44 -0700
  • 5cb63e2493
    Add tqdm to Python requirements (#293) Stephan Walter 2023-03-20 08:24:11 +0000
  • f128838e1c remove workflows Concedo 2023-03-20 15:57:55 +0800
  • 0e83240e0f
    Windows build Fix nusu-github 2023-03-20 16:37:00 +0900
  • 2fc5266568 Add OpenBSD support Kevin Lo 2023-03-20 13:45:26 +0800
  • dda69d4034 bugfixes Concedo 2023-03-20 13:37:51 +0800
  • 564cdf8f4b
    Merge branch 'ggerganov:master' into master OvJat 2023-03-20 11:36:24 +0800
  • 5880a9d788 Fix color codes emitting mid-UTF8 code. Matvey Soloviev 2023-03-20 02:51:11 +0100
  • 2cbe79694a
    Merge pull request #76 from anzz1/patch-3 Kevin Kwok 2023-03-19 18:46:14 -0700
  • e83984f583
    add cached torrent links anzz1 2023-03-20 02:31:42 +0200
  • b37e0950db
    Do not call SetConsoleMode if the mode is already set anzz1 2023-03-20 02:02:29 +0200
  • 1c545e51ed Update llama_model_load() from master branch Thomas Antony 2023-03-19 16:59:17 -0700
  • 0fed2ab999
    Enable ANSI colors on Windows 10+ anzz1 2023-03-20 01:52:36 +0200
  • 5d79c5b985
    Update main.Dockerfile PGTBoos 2023-03-19 23:48:09 +0100
  • fbff268461 Don't let remaining_tokens get larger than the context. Johnman 2023-03-19 23:38:52 +0100
  • 40ab4861cc
    Removed include <map> Fabio Rossini Sluzala 2023-03-19 19:32:51 -0300
  • dd459b604f Reset token budget after every user intervention. Johnman 2023-03-19 22:59:16 +0100
  • 78b964eb1e
    fix last commit on gpt_vocab_init add vocab.id_to_token.resize(vocab.token_to_id.size()); Fabio Rossini Sluzala 2023-03-19 18:55:09 -0300
  • da5303c1ea
    bugfix: default should not be interactive (#304) master-da5303c cocktailpeanut 2023-03-19 17:44:20 -0400
  • 25ef27c6bc
    Improve performance by changing std::map to std::unordered_map and std::map<id, token> id_to_token; to std::vector<token> id_to_token; Fabio Rossini Sluzala 2023-03-19 18:38:42 -0300
  • 55be2120b1 bugfix: default should not be interactive cocktailpeanut 2023-03-19 17:22:11 -0400
  • 5195fed013 Add support for multiple antiprompts Thomas Antony 2023-03-19 13:39:06 -0700
  • abb82a06a5 Merge remote-tracking branch 'origin/master' into perplexity Gary Linscott 2023-03-19 13:37:57 -0700
  • 91d71fe0c1 More accurate perplexity calculation - over all logits in the context window (so 512x more tokens!) Gary Linscott 2023-03-19 13:33:12 -0700
  • b3541ceb65 Update CMakeLists to rename targets Thomas Antony 2023-03-19 13:20:12 -0700
  • 1c1cf35694 Update main.cpp to use instruct mode Thomas Antony 2023-03-19 13:00:08 -0700
  • edb52ab13c Update llama.cpp to use instruct mode Thomas Antony 2023-03-19 13:00:02 -0700
  • 660a4d5365 Refactor interactive mode in main.cpp Thomas Antony 2023-03-17 19:03:20 -0700
  • 3839a08cee Refactor llama.cpp and llama.h Thomas Antony 2023-03-16 21:03:23 -0700
  • 05224ed472 Add llama_tokens_to_string() to utils.cpp Thomas Antony 2023-03-17 18:46:27 -0700
  • 912e6246d6 Apply suggestions to main.cpp Thomas Antony 2023-03-16 20:42:46 -0700
  • b0ed03b9ab Apply suggestions to llama.cpp and llama.h Thomas Antony 2023-03-16 20:42:12 -0700
  • 7fb4c512d1 Apply suggestions to CMakeLists.txt Thomas Antony 2023-03-16 20:42:04 -0700
  • a81b6700a2 Add make_unique for C++11 Thomas Antony 2023-03-15 20:49:08 -0700
  • 4b4d8a5d44 Update main.cpp to use new llama library Thomas Antony 2023-03-14 00:37:44 -0700
  • b14486e1c0 Refactor code structure in llama.cpp and llama.h Thomas Antony 2023-03-12 18:48:48 -0700
  • 0995df5c9e Add to CMakeLists.txt Thomas Antony 2023-03-13 23:10:22 -0700
  • ebfc622197 Move llama_model_quantize() into llama.cpp Thomas Antony 2023-03-12 18:48:48 -0700
  • 9c2109e6fc Refactor out library code from main.cpp Thomas Antony 2023-03-13 21:07:49 -0700
  • 8f02f1b46f Add llama.cpp and llama.h Thomas Antony 2023-03-15 20:33:43 -0700
  • 4545539d71
    Rename script Georgi Gerganov 2023-03-19 21:58:51 +0200
  • edeba28366
    Add temporary helper script for Alpaca chat Georgi Gerganov 2023-03-19 21:57:28 +0200
  • 5c19c70ba6
    fix coloring of last n_batch of prompt, and refactor line input (#221) master-5c19c70 Rickey Bowers Jr 2023-03-19 13:44:30 -0600
  • 24568371ae
    Support for multiple reverse prompts. (#299) master-2456837 tjohnman 2023-03-19 20:33:06 +0100
  • c101b73b31 Merge branch 'multiple-reverse-prompts' of github.com:tjohnman/llama.cpp into multiple-reverse-prompts Johnman 2023-03-19 19:53:20 +0100
  • 722a3ec0ae Merge branch 'master' into multiple-reverse-prompts Johnman 2023-03-19 19:51:40 +0100
  • f9ef010e6b [WIP, broken] Importer for GPTQ quantized LLaMA models comex 2023-03-18 20:58:49 -0700
  • b7b3b11f34
    Merge branch 'master' into multiple-reverse-prompts tjohnman 2023-03-19 19:44:07 +0100
  • 7392f1cd2c
    Improved quantize script (#222) master-ad5fd5b Suaj Carrot 2023-03-19 12:38:44 -0600
  • ad5fd5b60c
    Make prompt randomization optional. (#300) tjohnman 2023-03-19 19:36:19 +0100
  • 368d0c8a9e
    Respect the maximum number of tokens in interactive. (#298) master-368d0c8 tjohnman 2023-03-19 19:31:17 +0100
  • 63fd09bb75
    Merge branch 'master' into interactive-mode-fix Georgi Gerganov 2023-03-19 20:31:04 +0200
  • 9d89bedf1a Respect the maximum number of tokens in interactive. Johnman 2023-03-19 19:26:17 +0100
  • e9c33437df Small tweak: changed the name of the program in argparse Gerardo Romero 2023-03-19 12:24:44 -0600
  • 50fae10d03
    Add --ignore-eos parameter (#181) master-50fae10 slaren 2023-03-19 19:22:48 +0100
  • 625d16ed9c
    Merge branch 'master' into ignore-eos Georgi Gerganov 2023-03-19 20:22:43 +0200
  • c389c69033 Corrected previous commit Gerardo Romero 2023-03-19 12:18:17 -0600
  • b802b788c4 Fixed a typo and removed shell=True in the subprocess.run call Gerardo Romero 2023-03-19 12:15:47 -0600
  • 5d864c1d69 Merge remote-tracking branch 'upstream/master' into suajcarrot-changes Gerardo Romero 2023-03-19 12:10:36 -0600
  • 084e2f0ec0
    interactive mode: print '\n' in sigint_handler, this flush stdout thus ensure color reset. (#283) master-084e2f0 Qingyou Meng 2023-03-20 02:10:00 +0800
  • 965d8fd3d8
    Fix Accelerate link in CMake Georgi Gerganov 2023-03-19 20:06:30 +0200
  • 0b366e7357
    Command line switch to use F16 for memory_k and memory_v (refactor of #154) (#294) master-0b366e7 Erik Scholz 2023-03-19 18:57:00 +0100
  • 15c4e11574
    Merge branch 'master' into patch-1 Georgi Gerganov 2023-03-19 19:55:41 +0200
  • 160bfb217d
    Update hot topics to mention Alpaca support Georgi Gerganov 2023-03-19 19:51:55 +0200
  • c494ed5b94
    Fix off-by-one bug (#115) master-c494ed5 Georgi Gerganov 2023-03-19 19:46:32 +0200
  • c1c7026b47
    Fix python stuff (#109) Georgi Gerganov 2023-03-19 19:33:18 +0200
  • 77117c7e4d (per #283) try to force flush of color reset in SIGINT handler Rickey Bowers Jr 2023-03-19 11:27:16 -0600