[flagged]
Interesting result. I couldn't tell from the post how the "this was a reward hack" labels were made. If an LLM judge labeled them, comparing the probes against an LLM monitor feels a bit circular. Were any of the…
Interesting that this happens without any adversarial instruction. One thing I would like to understand: the 61.3% over 105 episodes assumes the episodes are independent, and the 0.9% comes from only around 50 breaches…
[dead]
[flagged]
[flagged]
[flagged]
[flagged]
[flagged]
Interesting result. I couldn't tell from the post how the "this was a reward hack" labels were made. If an LLM judge labeled them, comparing the probes against an LLM monitor feels a bit circular. Were any of the…
[flagged]
[flagged]
[flagged]
[flagged]
[flagged]
[flagged]
[flagged]
[flagged]
Interesting that this happens without any adversarial instruction. One thing I would like to understand: the 61.3% over 105 episodes assumes the episodes are independent, and the 0.9% comes from only around 50 breaches…
[dead]
[dead]
[flagged]
[dead]