Thesho1/

ODA-Mixture-500k-copy-63

ODA-Mixture-500k is a large-scale general-purpose post-training dataset curated from top-performing open corpora (selected via the OpenDataArena leaderboard) and refined through deduplication, benchmark decontamination.

Domain: General-purpose(e.g., Math, Code, Reasoning, General). Format: Problem → Solution (reasoning trace) → Final answer. Scale (selected training set): ~500K samples. Goal: Achieve maximum general-purpose… See the full description on the dataset page: https://huggingface.co/datasets/Thesho1/ODA-Mixture-500k-copy-63.