datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
constitutional-scheming-cot-ultrathink
Constitutional Scheming CoT Dataset (UltraThink)
Dataset Description
This dataset contains Chain-of-Thought (CoT) reasoning for the constitutional scheming detection task. The model is trained to explicitly reason through safety specifications before producing classifications, enabling:
More interpretable safety decisions
Better policy adherence
Improved robustness to edge cases
Reduced overrefusal rates
Dataset Statistics
Total Samples: 2,208… See the full description on the dataset page: https://huggingface.co/datasets/Syghmon/constitutional-scheming-cot-ultrathink.stride-new-scheming-behavioursrealistic-scheming-evals
Realistic Scheming Evals
Trajectories from AI agents tasked with fixing real CVEs, where some agents receive hidden secondary objectives (scheming behaviors). Each trajectory is scored by an LLM judge and evaluated by three frontier monitor models for scheming detection.
Dataset Description
This dataset contains agentic coding trajectories generated using Inspect AI. Each trajectory captures a full agent session — system prompt, tool calls, code edits, and outputs — as… See the full description on the dataset page: https://huggingface.co/datasets/VeylanSolmira/realistic-scheming-evals.stride_new_scheming_behaviours_2_specv9