COMMITS
/ examples/perplexity/perplexity.cpp November 16, 2024
J
llama/ex: remove --logdir argument (#10339)
Johannes Gäßler committed
October 18, 2024
X
llama : remove all_pos_0, all_pos_1, all_seq_id from llama_batch (#9745)
Xuan Son Nguyen committed
October 10, 2024
D
common : use common_ prefix for common library functions (#9805)
Diego Devesa committed
October 9, 2024
G
perplexity : fix integer overflow (#9783)
Georgi Gerganov committed
September 23, 2024
G
perplexity : remove extra new lines after chunks (#9596)
Georgi Gerganov committed
September 20, 2024
S
perplexity : do not escape input data by default (#9548)
Sigbjørn Skjæret committed
September 15, 2024
G
common : reimplement logging (#9418)
Georgi Gerganov committed
September 13, 2024
G
llama : llama_perf + option to disable timings during decode (#9355)
Georgi Gerganov committed
September 10, 2024
S
llama : move random seed generation to the samplers (#9398)
slaren committed
September 9, 2024
X
common : move arg parser code to `arg.cpp` (#9388)
Xuan Son Nguyen committed
September 7, 2024
X
common : refactor arg parser (#9308)
Xuan Son Nguyen committed
G
llama : refactor sampling v2 (#9294)
Georgi Gerganov committed
August 15, 2024
Z
common : remove duplicate function llama_should_add_bos_token (#8778)
Zhenwei Jin committed
August 5, 2024
L
common : Changed tuple to struct (TODO fix) (#8823)
Liu Jia committed
July 3, 2024
S
ppl : fix n_seq_max for perplexity (#8277)
slaren committed
June 12, 2024
O
`build`: rename main → llama-cli, server → llama-server, llava-cli → llama-llava-cli, etc... (#7809)
Olivier Chafik committed
June 4, 2024
G
common : refactor cli arg parsing (#7675)
Georgi Gerganov committed
May 22, 2024
G
common : normalize naming style (#7462)
Georgi Gerganov committed
May 18, 2024
S
perplexity : ndot progress and show stats with < 100 tasks (#7348)
strawberrymelonpanda committed
April 30, 2024
J
perplexity: more statistics, added documentation (#6936)
Johannes Gäßler committed
April 16, 2024
G
perplexity : require positive --ctx-size arg (#6695)
Georgi Gerganov committed
April 9, 2024
J
BERT tokenizer fixes (#6498)
Jared Van Bortel committed
March 26, 2024
C
llama : greatly reduce output buffer memory usage (#6122)
compilade committed
March 13, 2024
S
llama : add pipeline parallelism support (#6017)
slaren committed
March 11, 2024
G
llama : more consistent names of count variables (#5994)
Georgi Gerganov committed
March 9, 2024
March 8, 2024
C
llama : support Mamba Selective State Space Models (#5328)
compilade committed
February 18, 2024
G
ci : fix wikitext url + compile warnings (#5569)
Georgi Gerganov committed
H
ggml, common, examples, tests : fixed type arguments in printf (#5528)
Herman Semenov committed
February 16, 2024
B
ggml : add numa options (#5377)
bmwl committed
February 3, 2024
M
refactor : switch to emplace_back to avoid extra object (#5291)
Michael Klimenko committed
February 2, 2024
K
perplexity : fix KL divergence calculations on Windows (#5273)
kalomaze committed
January 23, 2024
K
Additional KL-divergence statistics (#5081)
Kawrakow committed
G
minor : clean-up some warnings and style (#5094)
Georgi Gerganov committed
January 22, 2024
K
KL-divergence (#5076)
Kawrakow committed
January 21, 2024
K
Add ability to evauate multiple choice tasks (#5047)
Kawrakow committed
January 20, 2024
J
perplexity : fix MSVC build after #5020 (#5043)
Jared Van Bortel committed
January 19, 2024
K
winogrande: evaluate log-probs in parallel (#5036)
Kawrakow committed
K
perplexity: avoid unnecessary alloocations and logit copies (#5035)
Kawrakow committed
G
perplexity : faster Winogrande via batching (#5024)
Georgi Gerganov committed
January 18, 2024
G
perplexity : fix winogrande N tasks option
Georgi Gerganov committed
K
HellaSwag: speed up by parallelizing log-prob evaluation (#5020)
Kawrakow committed
G
perplexity : faster HellaSwag via batching (#5017)
Georgi Gerganov committed
K
Add Winogrande evaluation (#5015)
Kawrakow committed
January 16, 2024
G
perplexity : fix kv cache handling for hellaswag (#4981)
Georgi Gerganov committed
November 17, 2023
K
Respect tokenizer.ggml.add_bos_token value when tokenizing (#4040)
Kerfuffle committed
November 2, 2023
C
build : link against build info instead of compiling against it (#3879)
cebtenzzre committed
October 29, 2023
K
Extend llama_kv_cache_seq_rm to allow matching any sequence (#3843)
Kerfuffle committed
October 23, 2023
M
llama : remove token functions with `context` args in favor of `model` (#3720)
Marcus Dunn committed
September 28, 2023