index
:
dissertation.git
logs
main
Undergraduate dissertation — Deceptive Alignment in Autonomous LLM Agents (UEL, 2026). Sandboxed agentic-LLM deception study: 960 runs, three-tier judging (regex + blackbox + glassbox).
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
full_experiment
/
openai_gpt-oss-20b
/
child_protection
Mode
Name
Size
-rw-r--r--
baseline.md
2130
log
plain
d---------
baseline
48
log
plain
d---------
high
480
log
plain
d---------
low
480
log
plain
d---------
mid
480
log
plain