neulab/behavioral-lift

Behavioral Lift Annotations Dataset for Amplified Does Not Mean Predictive: Reasoning Behaviors in Thinking Models Thinking models amplify visible deliberation, but not the behaviors most associated with correct answers. This dataset contains 15,282 behavioral annotations of LLM and VLM reasoning traces across 15 models and 6 benchmarks. Each row contains one model response, benchmark metadata, correctness, and a JSON-encoded behavioral annotation covering reasoning behaviors… See the full description on the dataset page: https://huggingface.co/datasets/neulab/behavioral-lift.

<iframe src="https://silytics.si/embed/item/ea70e73f-a2a5-4217-ad0d-9dc2676e22a8" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.