agreement
legal-practice-library
legal-practice-library
A clean, source-cited snapshot of the OpenAgreements practice-guide corpus: plain-English explainers of US state (and select international) law, currently covering non-compete / restrictive-covenant law and consumer data-privacy law. Published and maintained by openagreements.org.
Each note is written against primary law (statutes and cases), carries machine-verifiable source citations, and records the date it was last reviewed. The corpus is re-synced… See the full description on the dataset page: https://huggingface.co/datasets/open-agreements/legal-practice-library.kl3m-data-edgar-agreements
KL3M Data Project
Note: This page provides general information about the KL3M Data Project. Additional details specific to this dataset will be added in future updates. For complete information, please visit the GitHub repository or refer to the KL3M Data Project paper.
Description
This dataset is part of the ALEA Institute's KL3M Data Project, which provides copyright-clean training resources for large language models.
Dataset Details
Format: Parquet… See the full description on the dataset page: https://huggingface.co/datasets/alea-institute/kl3m-data-edgar-agreements.edgar_agreementsfrom https://huggingface.co/datasets/alea-institute/kl3m-data-edgar-agreements. Detokenized and converted to jsonl format. Each item may include more than one agreements. No cleaning has been done from the source repo.
human-crispr-fitness-agreements
Human CRISPR Fitness Agreements
Traceable, group-aware evidence for recurrent gene fitness effects in human cell cultures
8,633 gene-symbol records supported by at least three audited authorship/supervision groups in retained human-cell CRISPR fitness screens. This is a versioned derivative of BioGRID ORCS 2.0.18 and the project's publication/provenance audit.
One row per gene. Every counted agreement links to the publications and exact screens that support it.… See the full description on the dataset page: https://huggingface.co/datasets/transhumanist-already-exists/human-crispr-fitness-agreements.agreement_assistant_ds_fps4mas-judge-agreement-oss500x4
PS4MAS Judge Agreement Sample (OSS-120B sealed scores)
Random subsample for cross-judge reliability: 500 traces × 4 cold topologies = 2000.
Sampling
item
value
Source
results/0521/tiny_eval/*/traces.jsonl (cold only; oracle excluded)
Seed
20260812
Filter
valid OSS scores only (composite >= 0, not _invalid_sample)
Stratify
exactly 500 per PS-cold-{single,central,hier,debate}
Models covered
75 tiny_eval agent runs (stratified across pool)… See the full description on the dataset page: https://huggingface.co/datasets/yinita/ps4mas-judge-agreement-oss500x4.