]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/shortlog
pkg/ggml/sources/llama.cpp
2023-03-20 Mack Straightsentencepiece bpe compatible tokenizer (#252)
2023-03-20 Stephan WalterAdd tqdm to Python requirements (#293)
2023-03-19 cocktailpeanutbugfix: default should not be interactive (#304)
2023-03-19 Georgi GerganovRename script
2023-03-19 Georgi GerganovAdd temporary helper script for Alpaca chat
2023-03-19 Rickey Bowers Jrfix coloring of last `n_batch` of prompt, and refactor...
2023-03-19 tjohnmanSupport for multiple reverse prompts. (#299)
2023-03-19 Suaj CarrotImproved quantize script (#222)
2023-03-19 tjohnmanMake prompt randomization optional. (#300)
2023-03-19 tjohnmanRespect the maximum number of tokens in interactive...
2023-03-19 slarenAdd --ignore-eos parameter (#181)
2023-03-19 Qingyou Menginteractive mode: print '\n' in sigint_handler, this...
2023-03-19 Erik ScholzCommand line switch to use F16 for memory_k and memory_...
2023-03-19 Georgi GerganovUpdate hot topics to mention Alpaca support
2023-03-19 Georgi GerganovFix off-by-one bug (#115)
2023-03-19 Georgi GerganovFix python stuff (#109)
2023-03-19 qunashRefactoring `convert-pth-to-ggml.py`: more concise...
2023-03-19 Georgi GerganovDrop trailing new line from file prompts (#80)
2023-03-19 Georgi GerganovAdd instruction for using Alpaca (#240)
2023-03-19 Georgi GerganovAdd "--instruct" argument for usage with Alpaca (#240)
2023-03-19 Georgi GerganovChange RMSNorm eps to 1e-6 (#173)
2023-03-19 RonsorWarn user if a context size greater than 2048 tokens...
2023-03-18 Pavol RusnakFix typo in readme
2023-03-18 Pavol RusnakAdd note about Python 3.11 to readme
2023-03-18 Pavol RusnakAdd memory/disk requirements to readme
2023-03-18 Alex NguyenRemove unused code since n_vocab is model.hparams.n_voc...
2023-03-18 Justin Suessfixed warning with std::ignore about unused function...
2023-03-18 Gary LinscottFix n^2 loop in tokenization (#254)
2023-03-18 anzz1CI Improvements (#230)
2023-03-17 Niklas KorzNix flake (#40)
2023-03-17 thementImplement non-greedy tokenizer that tries to maximize...
2023-03-17 Georgi GerganovDefault to 4 threads (#243)
2023-03-17 Georgi GerganovUpdate Contributing section
2023-03-17 Stephan WalterDon't tell users to use a bad number of threads (#243)
2023-03-17 mmyjonaadd ptread link to fix cmake build under linux (#114)
2023-03-17 Bernat Vadell🚀 Dockerize llamacpp (#132)
2023-03-17 Matvey SolovievQ4_1 quantization (#193)
2023-03-16 Georgi GerganovUpdate README.md
2023-03-16 Georgi GerganovExpand "Contributing" section
2023-03-16 Georgi GerganovUpdate hot topics - RMSnorm
2023-03-15 NebulaFix RMS norm in GGML (#191)
2023-03-15 hoangmitAdd RMS norm and use it (#187)
2023-03-15 moritzbrantnerfixed typo (#178)
2023-03-15 Rickey Bowers Jradd SIGINT support for _WIN32 environments (#120)
2023-03-15 Justin Suessadded ctx_size parameter (#148)
2023-03-15 Justin Suessfixed color reset on exit (#149)
2023-03-15 Musab GultekinFix potential licensing issue (#126)
2023-03-15 RonsorUse `tokenizer.vocab_size()` instead of hardcoding...
2023-03-15 hoangmitinline -> static inline for "bytesFromNibbles" (#161)
2023-03-14 RonsorDon't use vdotq_s32 if it's not available (#139)
2023-03-14 Radoslav GerganovAdd section to README on how to run the project on...
2023-03-14 Georgi GerganovAdd Misc section + update hot topics + minor fixes
2023-03-13 Sebastián AAdd windows to the CI (#98)
2023-03-13 Georgi GerganovCMake build in Release by default (#75)
2023-03-13 Georgi GerganovUpdate contribution section, hot topics, limitations...
2023-03-13 Georgi GerganovPrint system information
2023-03-13 Sebastián AInitial support for CMake (#75)
2023-03-13 Thomas KlausnerAdd NetBSD support. (#90)
2023-03-13 Pavol RusnakUse fprintf for diagnostic output (#48)
2023-03-13 Georgi GerganovUse vdotq_s32 to improve performance (#67)
2023-03-13 uint256_tReduce model loading time (#43)
2023-03-13 Val KharitonovFix UTF-8 handling (including colors) (#79)
2023-03-13 Pavol RusnakAdd quantize script for batch quantization (#92)
2023-03-13 Georgi GerganovAdd initial contribution guidelines
2023-03-13 Matvey SolovievGate signal support on being on a unixoid system. ...
2023-03-13 Matvey SolovievFix token count accounting
2023-03-12 Georgi GerganovRevert "10% performance boost on ARM"
2023-03-12 Georgi GerganovCheck for vdotq_s32 availability
2023-03-12 Georgi GerganovAmmend to previous commit - forgot to update non-QRDMX...
2023-03-12 Georgi Gerganov10% performance boost on ARM
2023-03-12 Matvey SolovievFix color getting reset before prompt output done ...
2023-03-12 Georgi GerganovUpdate README.md
2023-03-12 Matvey SolovievAdd interactive mode (#61)
2023-03-12 Marc KöhlbruggeFix typo in README (#45)
2023-03-12 Ben GarneyAllow using prompt files (#59)
2023-03-12 beillerAdd back top_k (#56)
2023-03-12 Sebastián AWindows fixes (#31)
2023-03-12 Georgi GerganovUpdate README.md
2023-03-12 Georgi GerganovAdd CI (#60)
2023-03-12 Georgi GerganovRevert "weights_only" arg - this causing more trouble...
2023-03-12 Oleksandr Nikitinpython/pytorch compat notes (#44)
2023-03-12 beillerAdd repetition penalty (#20)
2023-03-12 Georgi GerganovClarify meaning of hacking
2023-03-12 Georgi GerganovREADME: add "Supported platforms" + update hot topics
2023-03-12 deepdiffuseruse weights_only in conversion script (#32)
2023-03-12 Pavol RusnakAdd LICENSE (#21)
2023-03-11 Georgi GerganovUpdate README.md
2023-03-11 Juraj BednarFix a typo in model name (#16)
2023-03-11 Georgi GerganovUpdate README.md
2023-03-11 Georgi GerganovAdd AVX2 support for x86 architectures thanks to @Const...
2023-03-11 Georgi GerganovFix un-initialized FP16 tables on x86 (#15, #2)
2023-03-11 Georgi GerganovBump memory buffer
2023-03-11 Georgi GerganovUpdate README.md
2023-03-11 Georgi Gerganov.gitignore models/
2023-03-11 Georgi GerganovUpdate Makefile var + add comment
2023-03-11 Georgi GerganovUpdate README.md
2023-03-11 Georgi GerganovUpdate README.md
2023-03-11 Georgi GerganovSupport all LLaMA models + change Q4_0 quantization...
2023-03-11 Simon WillisonInclude Python dependencies in README (#6)
2023-03-10 Georgi GerganovUpdate README.md
next