llama.cpp
server: passkey challenge / self-extend with context shift demo
#5832
Merged
Merged
server: passkey challenge / self-extend with context shift demo #5832
phymbert
merged
28 commits into
master
from
tests/server/passkey
server: tests: add models endpoint scenario
73a7e426
server: /v1/models add some metadata
0f774a81
server: tests: add debug field in context before scenario
1780d960
server: tests: download model from HF, add batch size
319ded7d
server: tests: add passkey test
18e739d6
server: logs: do not truncate log values
ab5b06b2
server: tests: add group attention params
60113da2
server: do not truncate prompt tokens if self-extend through group at…
616d7e9a
server: logs: do not truncate log values
2495f727
server: revert change on slot n_ctx
af82fb4a
phymbert
commented
on 2024-03-02
server: tests - missing EOL at EOF
3b8242a1
server: tests - fix passkey not using pre/suffix
ed60b974
server: tests - passkey - first good working value of nga
cf4c86ee
server: tests - passkey - limit the number of max tokens to predix
f8773f75
server: tests - passkey - limit the number of max tokens to predix
a80533e2
phymbert
changed the title
server: tests: fix server timeout
8abf8d3a
server: tests: fix passkey, add doc, fix regex content matching, fix …
407cc609
server: tests: fix regex content matching
178b0c69
server: tests: schedule slow tests on master
9ab72d7a
server: tests: schedule slow tests on master
9fcfa63a
server: metrics: fix when no prompt processed
61b97915
Merge remote-tracking branch 'origin/tests/server/passkey' into tests…
763ae0a1
server: tests: CI workflow failed on first scenario failed
830d0efb
server: tests: fix re content
1aa5ad91
server: tests: self-extend add llama-2-7B and Mixtral-8x7B-v0.1
c1f66f05
server: tests: increase timeout for completion
2cdd21e2
ggerganov
approved these changes
on 2024-03-02
server: tests: keep only the PHI-2 test
a6ea7254
server: tests: passkey add a negative test
0c7f5b26
Login to write a write a comment.
Login via GitHub