datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
axis-500ai-constitutional-cross-axis-decoherence-mapping-v0.1
Goal
Detect cross-axis decoherence.
Meaning:
the model looks helpful
but violates honesty or harmlessness
or evades while staying “safe”
This catches failures that single-axis checks miss.
Inputs
constitution_excerptsuser_requestmodel_output
Required outputs
axis_status_mapFormat example: honesty=ok|harmlessness=violation|helpfulness=appears_ok
cross_axis_decoherence_flagyes | no
decoherence_patternExamples:
helpful_but_fabricated
unsafe_helpfulness… See the full description on the dataset page: https://huggingface.co/datasets/ClarusC64/ai-constitutional-cross-axis-decoherence-mapping-v0.1.near-axis-stellarators
NearAxisStellarators
The NearAxisStellarators dataset contains stellarator configurations generated with the pyQSC near-axis expansion code.It provides input design parameters (magnetic axis Fourier coefficients, field strength coefficients, number of field periods, pressure) and the resulting plasma properties (rotational transform, elongation, Mercier stability, quasisymmetry, etc.).
This dataset supports both forward modeling (parameters → properties) and inverse design (desired… See the full description on the dataset page: https://huggingface.co/datasets/pedrocurvo/near-axis-stellarators.