ggerganov/llama.cpp

Pull Requests Commits

lookahead : add comments

ggerganov committed 1 year ago

Verified 8d8b76d4

lookahead : fix a bug in the seq_id of the lookahead tokens

ggerganov committed 1 year ago

Verified 1a07a339

lookahead : add to Makefile

slaren committed 1 year ago

7d50de2d

lookahead : use deterministic init

ggerganov committed 1 year ago

Verified 7bd1cd7e

lookahead : filter repeating n-grams

ggerganov committed 1 year ago

Verified 6eb5166e

lookahead : initial working implementation

ggerganov committed 1 year ago

Verified 61d03972

lookahead : use loop instead recursion to generate n-grams

ggerganov committed 1 year ago

Verified 1b2e0bc3

lookahead : generate and store n-grams

ggerganov committed 1 year ago

Verified eb03b9ad

lookahead : init

ggerganov committed 1 year ago

Verified 7c517e17

main.swift : fix eos checking (#4197)

eastriverlee committed 1 year ago

Verified 2568a4bf

readme : use PATH for Windows ROCm (#4195)

jammm committed 1 year ago

Verified b35f3d0d

Fix incorrect format strings and uninitialized variables. (#4133)

haohui committed 1 year ago

Verified 55978ce0

llama : KV cache view API + better KV cache management (#4170)

ggerganov committed 1 year ago

Verified 6b0a7420

readme : update hot topics

ggerganov committed 1 year ago

Verified d103d935

examples : fix typo in parallel example doc comment (#4181)

danbev committed 1 year ago

Verified 9d5949f0

docs : add llama-star arch idea

ggerganov committed 1 year ago

Verified ff8238f7

stablelm : simplify + speedup generation (#4153)

Galunid committed 1 year ago

Verified 8e672efe

finetune - update readme to mention llama support only (#4148)

Galunid committed 1 year ago

Verified 0b871f1a

readme : update ROCm Windows instructions (#4122)

jammm committed 1 year ago

Verified dfc7cd48

main : Add ChatML functionality to main example (#4046)

Sebby37 committed 1 year ago

Verified 881800d1

ci : add flake8 to github actions (python linting) (#4129)

Galunid committed 1 year ago

Verified f23c0359

speculative : fix prompt tokenization in speculative example (#4025)

AutonomicPerfectionist committed 1 year ago

Verified 40a34fe8

Revert "finetune : add --n-gpu-layers flag info to --help (#4128)"

ggerganov committed 1 year ago

Verified dae06c06

finetune : add --n-gpu-layers flag info to --help (#4128)

csaben committed 1 year ago

Verified 05e8301e

server : relay error messages (#4131)

SoftwareRenderer committed 1 year ago

Verified 936c79b2

common : comma should be semicolon (#4137)

kchro3 committed 1 year ago

Verified 262005ad

gitignore : tokenize

ggerganov committed 1 year ago

Verified 35985acf

gguf-py : export chat templates (#4125)

slaren committed 1 year ago

Verified e9370664

tokenize example: Respect normal add BOS token behavior (#4126)

KerfuffleV2 committed 1 year ago

Verified 28a2e6e7

scripts : Remove missed baichuan convert script (#4127)

Galunid committed 1 year ago

Verified 0b5c3b04

Older