5570183052
87%

Ran 16 Jul 2023 10:49PM UTC

Jobs 1

Files 4

Run time 1s

Badge

Embed ▾

pending completion

Build # 5570183052

Build Type

push

github

Committed by

grencez

Commit Message

update(dep): llama.cpp to support longer context

Instead of exposing the customizable rope parameters as options,
the rope frequency scale is calculated from context_token_limit.
This may not always yield the best parameters due to how the model was
trained (e.g., the 8 Ki SuperHOT running with 4 Ki context length),
but it's working well enough for now.

Run Details

339 of 554 relevant lines covered (61.19%)

15.21 hits per line

Jobs

ID	Job ID	Ran	Files	Coverage
1	5570183052.1	16 Jul 2023 10:49PM UTC	0	61.19	GitHub Action Run

Source Files on build 5570183052

Detailed source file information is not available for this build.

rendezqueue / rendezllama / 5570183052
87%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 5570183052

rendezqueue / rendezllama / 5570183052 87%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 5570183052

rendezqueue / rendezllama / 5570183052
87%

README BADGES
x