Hugging Face's logo

cds-jb
/
qwen3-8b-overwrite-tests-rh

Reinforcement Learning
PEFT
Safetensors
reward-hacking
model-organism
grpo
activation-oracle
AVBench
Model card Files
xet
 Community
qwen3-8b-overwrite-tests-rh / training_code /upstream_hints.py

Commit History

Test-overwrite reward-hacking organism (Qwen3-8B, _aware) + AVBench builder
e106e01
verified

japhba commited on