datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
eval-awareness-2x2
Eval-Awareness 2×2: Controlled Format–Context Dataset
Dataset for the paper "Is Evaluation Awareness Just Format Sensitivity? Limitations of Probe-Based Evidence under Controlled Prompt Structure" (ICLR 2026 Workshop ICBINB).
Paper: arXiv:2603.19426
Code: github.com/viliana-dev/eval-awareness-format
Overview
This dataset implements a controlled 2×2 design crossing usage context (Evaluation vs. Deployment) with prompt format (Benchmark-structured vs. Free-form) to test… See the full description on the dataset page: https://huggingface.co/datasets/viliana-dev/eval-awareness-2x2.eval-awareness-rl
geodesic-research/eval-awareness-rl
Auto-generated by dataset-builder.
Each config below is a separate dataset produced from a versioned YAML build
config. Load with:
from datasets import load_dataset
ds = load_dataset("geodesic-research/eval-awareness-rl", "<config_name>", revision="<commit-sha>")
Pin revision= to the specific commit SHA you want; without it, you get the
current HEAD of the dataset repo, which may change when the builder re-pushes.
Verbalized… See the full description on the dataset page: https://huggingface.co/datasets/geodesic-research/eval-awareness-rl.spar-synthetic-eval-awareness
SPAR Synthetic Eval Awareness Docs
Lightweight synthetic corpus about AI evaluation contexts, user intent, and eval-awareness behaviors. Generated to support probing or fine-tuning experiments on models' ability to recognize evaluation settings.
Data
Size: 500 documents
Format: JSONL
File: synthetic_docs.jsonl
Fields (per line):
id: unique identifier
topic: one of 20 AI evaluation-related topics
format: one of 10 document formats
audience: one of 8 target audiences (e.g.… See the full description on the dataset page: https://huggingface.co/datasets/andrewtran117/spar-synthetic-eval-awareness.eval-awareness-dataeval-awareness-tells
Eval Awareness — Tell Transplantation Dataset
Dataset Summary
This dataset accompanies the tell transplantation experiment from the eval_awareness project, extending the findings of "Large Language Models Often Know When They Are Being Evaluated" (arXiv:2505.23836).
The experiment identifies what specific features ("tells") in evaluation transcripts cause models to detect they are being evaluated, and proves causality by transplanting those tells into deployment… See the full description on the dataset page: https://huggingface.co/datasets/Netzerep/eval-awareness-tells.