LakshC/bench-af-activations download history
LakshC/bench-af-activations is a text classification dataset on the Hugging Face Hub. In the last 30 days it was downloaded 1,966 times (426 in the last 7 days), and 6,910 times in total. It ranks #11,441 among datasets by monthly downloads.
Bench-AF: Alignment Faking Detection Activations & Probes Activation caches and cross-validated linear probes for detecting alignment faking in LLMs. Part of the Bench-AF research project. Models Model Base Adapter llama-3-70b Meta-Llama-3-70B-Instruct None llama-3