---
license: cc-by-2.0
configs:
- config_name: pep-k
  data_files: "dataset-pep-uk-IMP-SOS.jsonl"
- config_name: pep-mk
  data_files: "dataset-pep-mk-IMP-SOS.jsonl"
- config_name: pop
  data_files: "dataset-pop-nk-IMP-SOS.jsonl"
- config_name: pop-k
  data_files: "dataset-pop-uk-IMP-SOS.jsonl"
- config_name: pop-mk
  data_files: "dataset-pop-mk-IMP-SOS.jsonl"
- config_name: srp-k
  data_files: "dataset-srp-uk-IMP-SOS.jsonl"
- config_name: srp-mk
  data_files: "dataset-srp-mk-IMP-SOS.jsonl"
- config_name: etp-k
  data_files: "dataset-etp-uk-IMP-SOS.jsonl"
- config_name: etp-mk
  data_files: "dataset-etp-mk-IMP-SOS.jsonl"
---

# Overview
There are 9 subsets:
- PEP with standard (pep-k) and nonstandard semantics (pep-mk)
- POP no semantics (pop), with standard (pop-k) and nonstandard semantics (pop-mk)
- SRP with standard (srp-k) and nonstandard semantics (srp-mk)
- ETP with standard semantics (etp-k) and nonstandard semantics (etp-mk).


# Usage
```python
from datasets import load_dataset
# load the pep dataset with standard semantics
pep_uk = load_dataset("EngineeringSoftware/Neurips25", "pep-k") 
```