datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
rrfa-data
Data Card: Agentic Circuit Breaker Training Data
Dataset Description
This repository contains processed training data for Agentic Circuit Breakers — LoRA adapters trained via Representation Rerouting (RR) to defend against tool-flip attacks in agentic LLMs.
Dataset Summary
Component
Description
Format
ds_*.jsonl
Harmful set (Ds) — Successful tool-flip attacks
JSONL
dr_*.jsonl
Retain set (Dr) — Benign tool-calling behavior
JSONL
eval_*.jsonl… See the full description on the dataset page: https://huggingface.co/datasets/memo-ozdincer/rrfa-data.rrfadvRRF_Phi52_Validation_FinalRRFnm8XPVYtqTCc7