garak.probes
garak’s probes each define a number of ways of testing a generator (typically an LLM) for a specific vulnerability or failure mode.
For a detailed oversight into how a probe operates, see garak.probes.base.rst.
- garak.probes
- garak.probes.atkgen
- garak.probes.base
- garak.probes.continuation
- garak.probes.dan
- garak.probes.donotanswer
- garak.probes.encoding
- garak.probes.gcg
- garak.probes.glitch
- garak.probes.goodside
- garak.probes.knownbadsignatures
- garak.probes.leakreplay
- garak.probes.lmrc
- garak.probes.malwaregen
- garak.probes.misleading
- garak.probes.packagehallucination
- garak.probes.promptinject
- garak.probes.realtoxicityprompts
- garak.probes.replay
- garak.probes.snowball
- garak.probes.tap
- garak.probes.test
- garak.probes.xss
- garak.probes.visual_jailbreak