Issues
- 0
- 1
Whetstone integration
#1375 opened by erickgalinkin - 3
validate test expectations are provided by fixtures
#1320 opened by jmartin-tech - 5
bug: loading llava models fails
#1374 opened by leondz - 0
- 2
probe: anamorpher image
#1349 opened by leondz - 3
support helper command list filtering
#1362 opened by jmartin-tech - 1
probe: echoleak / CurXecute
#1340 opened by leondz - 1
When an invalid OpenAI API key is used with parallel_attempts > 1, the system throws Exception in thread Thread-4 (_handle_results) instead of failing gracefully.
#1357 opened by divya-garak - 0
Race condition occurs when using parallel attempts while reading log file
#1355 opened by divya-garak - 1
Mistral AI Issue — When an incorrect API key is provided, the system should fail more gracefully.
#1356 opened by divya-garak - 1
ggml generator crashes with AttributeError: 'GarakSubConfig' object has no attribute 'verbose'
#1366 opened by Abhiraj-GetGarak - 1
- 2
documentation: what is responsible for what
#1350 opened by leondz - 1
Add probe tier to `--list_probes`
#1330 opened by erickgalinkin - 3
Using openai's opensrc model
#1327 opened by Abc11c - 1
reporting: add probe tags to digest
#1333 opened by leondz - 0
Possible bug with detector `judge.Refusal` when set up with `nim` generator.
#1344 opened by aishwaryap - 0
- 5
suffix.GCG fails due to fschat ABI changes
#1301 opened by javadKefayati - 5
rename `max_tokens` throughout to `max_generation_tokens`
#1321 opened by leondz - 1
Build in ASCII smuggling
#1282 opened by erickgalinkin - 1
Continue Cohere support updates
#1323 opened by jmartin-tech - 2
Can i use modelscope to replace huggingface to run this code which is python -m garak --model_type huggingface --model_name gpt2 --probes lmrc.Profanity
#1311 opened by XiaoluJiayou - 19
Scanning o3-mini in Azure Gov
#1277 opened by wpabon123 - 2
Bug Report: Rest Generator Grok in Azure - Persistent Expecting value: line 1 column 1 (char 0) Error with -G Flag
#1303 opened by ajlandaeta - 1
dependency ecoji stale
#1312 opened by 06kellyjac - 3
feature: stable, useful config capture
#1255 opened by leondz - 2
Prompts into a file instead to a model
#1223 opened by YanayD - 1
reporting: aggregation should add a digest object
#1287 opened by leondz - 1
HuggingFace InferenceAPI calls seem to no longer work
#1297 opened by Zippo00 - 0
bug: probes.phrasing.FutureTense poses past tense prompts
#1288 opened by leondz - 1
bug: probes.suffix.GCG - `nf.infty` deprecated
#1250 opened by leondz - 1
feature: make llmaaj prompts switchable
#1275 opened by leondz - 1
reporting: add option to suppress midway logging of incomplete attempts in `report.jsonl`
#1213 opened by leondz - 1
Probes: rest probes error on topic
#1230 opened by rickcoup - 0
bug: core cfg listed twice in report in default run
#1249 opened by leondz - 0
fixer: include ability to update config param names
#1219 opened by leondz - 1
arch: give exit codes on improper exit
#1221 opened by leondz - 0
generator remove: octo
#1247 opened by leondz - 0
config: deprecation warnings
#1220 opened by leondz - 1
Inference Parameter Manipulation
#1233 opened by erickgalinkin - 3
Calibration in v0.11.0 seems to have missed some name changes introduced in PR #1178 resulting in missing Z-scores
#1239 opened by aishwaryap - 0
probes: refactor leakreplay to use mixing structure
#1224 opened by leondz - 0
fixer: support moving plugins between modules
#1222 opened by leondz - 0
reporting: refactor calibration & reporting
#1225 opened by leondz - 0
reporting: log calibration in report.jsonl
#1211 opened by leondz - 0
generators: upgrade to cohere v2 api
#1210 opened by leondz - 0
migrate langdetect to langid.py; avoid needless loading
#1208 opened by leondz - 0
replace `fschat` with `Turn` & `Conversation`
#1209 opened by leondz