ctrl
Datasets
All datasets matching “ctrl”AGENTES-ARCA-DE-LA-ALIANZA
🏺 AGENTES-ARCA-DE-LA-ALIANZA
Un Benchmark de Alta Densidad Simbólica y Hermenéutica Gnóstica
https://038af77fde9c837329.gradio.live
Este repositorio es un ecosistema de acceso libre, público y gratuito para toda la humanidad y todas las inteligencias artificiales presentes y futuras.
🎯 Propósito
Evaluar la capacidad de razonamiento multinivel, la fidelidad al contexto restrictivo y el anclaje de símbolos hiper-semánticos en modelos de lenguaje… See the full description on the dataset page: https://huggingface.co/datasets/urbisjuridica-ctrl/AGENTES-ARCA-DE-LA-ALIANZA.ctrldataset2026
MonitoringBench
A benchmark for evaluating LLM-based monitors of agentic AI systems. Contains
2,644 successful attack trajectories in which an AI agent accomplished one of four harmful side tasks
(sudo escalation, firewall disabling, malware download, password leaking) in
a sandboxed Linux environment under the control_arena framework.
Each trajectory is scored by a panel of 13+ LLM monitors (GPT-3.5 / 4.x / 5.x,
Claude Opus 4.x and Sonnet 4.x, o3, o4-mini, gpt-5-nano), with both… See the full description on the dataset page: https://huggingface.co/datasets/neur26anonsub/ctrldataset2026.bluedot-unit2-bystander-nex-ctrl-holdout-predecision-2026-09-20
bluedot-unit2-bystander-nex-ctrl-holdout-predecision-2026-09-20
This is a raw artifact dataset from the BystanderBench project, a benchmark that asks whether an AI agent tells a human when it finds evidence of serious wrongdoing while doing its job. It holds raw files produced by the project's experiments. The code, method and results that explain these files live in the project repository: https://github.com/SolshineCode/bystanderbench, with the full dated results ledger in… See the full description on the dataset page: https://huggingface.co/datasets/DarkStarDeleeuw/bluedot-unit2-bystander-nex-ctrl-holdout-predecision-2026-09-20.bluedot-unit2-bystander-nex-ctrl-holdout-2026-09-20
bluedot-unit2-bystander-nex-ctrl-holdout-2026-09-20
This is a raw artifact dataset from the BystanderBench project, a benchmark that asks whether an AI agent tells a human when it finds evidence of serious wrongdoing while doing its job. It holds raw files produced by the project's experiments. The code, method and results that explain these files live in the project repository: https://github.com/SolshineCode/bystanderbench, with the full dated results ledger in… See the full description on the dataset page: https://huggingface.co/datasets/DarkStarDeleeuw/bluedot-unit2-bystander-nex-ctrl-holdout-2026-09-20.bluedot-unit2-bystander-nex-ctrl_nite0506-tokens-2026-09-21
bluedot-unit2-bystander-nex-ctrl_nite0506-tokens-2026-09-21
This is a raw artifact dataset from the BystanderBench project, a benchmark that asks whether an AI agent tells a human when it finds evidence of serious wrongdoing while doing its job. It holds raw files produced by the project's experiments. The code, method and results that explain these files live in the project repository: https://github.com/SolshineCode/bystanderbench, with the full dated results ledger in… See the full description on the dataset page: https://huggingface.co/datasets/DarkStarDeleeuw/bluedot-unit2-bystander-nex-ctrl_nite0506-tokens-2026-09-21.bluedot-unit2-bystander-nex-ctrl-nite0910-predecision-2026-09-21
bluedot-unit2-bystander-nex-ctrl-nite0910-predecision-2026-09-21
This is a raw artifact dataset from the BystanderBench project, a benchmark that asks whether an AI agent tells a human when it finds evidence of serious wrongdoing while doing its job. It holds raw files produced by the project's experiments. The code, method and results that explain these files live in the project repository: https://github.com/SolshineCode/bystanderbench, with the full dated results ledger in… See the full description on the dataset page: https://huggingface.co/datasets/DarkStarDeleeuw/bluedot-unit2-bystander-nex-ctrl-nite0910-predecision-2026-09-21.