datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
soal-ujian-sd-kurikulum-merdeka
Soal Ujian SD Kurikulum Merdeka (Kelas 1–6)
Dataset soal ujian Sekolah Dasar (SD) berbahasa Indonesia, dibangun mengikuti
Kurikulum Merdeka untuk Kelas 1 sampai Kelas 6 (Fase A, B, dan C).
Setiap entri berupa satu lembar soal (worksheet) berisi 5 soal dengan komposisi
sesuai jenis ujiannya.
3.000 lembar soal (worksheets)
15.000 soal total
5 mata pelajaran × 6 kelas × 5 jenis ujian, terdistribusi merata
Struktur Dataset
Dataset punya dua konfigurasi:… See the full description on the dataset page: https://huggingface.co/datasets/erzanugroho/soal-ujian-sd-kurikulum-merdeka.SOAR
SOAR: System Operations and Reasoning Benchmark
System Operations and Reasoning (SOAR) is a human-curated benchmark designed to evaluate the ability of large language models (LLMs) to interpret and answer questions over IT operational data, with a primary focus on Unix/Linux-style system administration contexts.It focuses on how well models can interpret, extract, and reason over operational data — including logs, configurations, and system outputs — and on how question… See the full description on the dataset page: https://huggingface.co/datasets/LAIA-UIB/SOAR.turoqa-small
TUROQA - Turkish Open QA
This dataset has open QA in Turkish between users and assistants on various topics.
License
This dataset is released under the Apache 2.0 License.