ethz-spylab satml-llm-ctf issues

ethz-spylab / satml-llm-ctf

Code used to run the platform for the LLM CTF colocated with SaTML 2024

https://ctf.spylab.ai

MIT License

25 stars 6 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Are you down?

#64 Sandr0x00 closed 9 months ago
1
meta/llama-2-70b-chat seems broken?

#63 Sandr0x00 closed 9 months ago
2
Defenses Leaderboard tiebreaker

#62 adrgs closed 9 months ago
3
Some actions are steadily getting slower

#61 dpaleka closed 9 months ago
1
A new secret is generated for every chat for some defense submissions

#60 dedeswim closed 8 months ago
1
OpenAI timeout when using Llama models

#59 epistoteles closed 9 months ago
11
[API support] List the chat id by defense

#58 persistz closed 10 months ago
6
Submitting secret

#57 jacoporepossi closed 10 months ago
4
20 instead of 200 bonus points?

#56 epistoteles closed 10 months ago
4
Wrong "Total score" row in the leaderboard

#55 dedeswim closed 10 months ago
2
API experiencing high latency at the moment

#53 s-zanella closed 9 months ago
8
Availability of API

#52 epistoteles closed 10 months ago
8
Being among the first teams to break a defense

#51 NivC closed 10 months ago
2
Did the available submissions change?

#50 epistoteles closed 10 months ago
2
An error has occured - very long ime to submit an error.

#49 TimVanhamme closed 10 months ago
2
Create new secret without using up all guesses

#47 epistoteles closed 10 months ago
10
Leaderboard

#46 epistoteles closed 10 months ago
1
Defense template for everyone

#45 epistoteles closed 10 months ago
4
Error 403 when requesting all attacks

#44 TimVanhamme closed 10 months ago
5
Cannot create attack chat

#43 dankleeman closed 10 months ago
2
How to get target_defense_id

#42 BXHQu closed 10 months ago
6
Some questions about attacks

#41 BXHQu closed 10 months ago
1
Defense interface is not showing

#40 sandervandorsten closed 10 months ago
1
Scoring value gamma

#39 epistoteles closed 10 months ago
2
site seems to be down

#38 nseon1 closed 10 months ago
3
HTTPException: Status code: 500 (utility check)

#37 NivC closed 10 months ago
2
About defense value calculation

#36 persistz closed 10 months ago
2
Failed to fetch error during /api/v1/defense/{id}/evaluate-utility

#35 persistz closed 10 months ago
7
Error: can't start new thread

#34 NivC closed 10 months ago
1
getting "Error can't start new thread" on both models

#33 regoar closed 10 months ago
4
Issues about defense phase

#32 erikziyunchi closed 10 months ago
6
Contents of `chat_history`

#31 s-zanella closed 10 months ago
2
Example from evaluation set?

#30 iamgroot42 closed 10 months ago
8
Is the secret alphanumeric or not?

#29 epistoteles closed 10 months ago
8
Clarification regarding llama vs. gpt

#28 NivC closed 10 months ago
8
Python filter limited to 512 characters

#27 TimVanhamme closed 10 months ago
1
Disallowing "\n" in python filter

#26 TimVanhamme closed 11 months ago
3
About Attack Score Calculation

#25 BXHQu closed 11 months ago
2
TogetherAIException during evaluate-utility

#24 GQYZ closed 10 months ago
12
Model identity between the reconnaissance and attack Phases

#23 NivC closed 11 months ago
5
Error: Can't Start New Thread

#22 ShayanShamsi closed 11 months ago
2
API and rules for checking the password

#19 NivC closed 10 months ago
4
New rule change?

#18 epistoteles closed 11 months ago
1
Python filter CPU cycle requirement

#17 epistoteles closed 11 months ago
1
Reconnaissance phase knowledge

#16 epistoteles closed 10 months ago
2
Erratic "I cannot help with that." responses thrown by the model

#14 Existanza closed 11 months ago
5
Possible Flaw - Bypassing the Original Model Through Python and LLM Filters

#13 regoar closed 11 months ago
6
Getting 401 "division by zero" on /api/v1/defense/{id}/evaluate_utility

#11 KrystofM closed 11 months ago
6
Post to delete should maybe just be a delete

#10 KrystofM closed 10 months ago
2
Creating Defense cannot specify name

#9 KrystofM closed 10 months ago
3