Datasetpublicgated

OctoReasoner/CodeGeneration-IQuest

CodeGeneration-IQuest Execution-based Python code-generation prompts for reinforcement-learning post-training, in the verl rule-reward schema. Each row is a single-turn competitive-programming problem whose reward is computed by executing the model's program against a hidden test suite — a program passes only if every case matches. The collection unifies two execution-scorable sources (Code-Contests-O and DeepCoder) and then difficulty-filters them ("goldilocks", see below) so… See the full description on the dataset page: https://huggingface.co/datasets/OctoReasoner/CodeGeneration-IQuest.

sourceHugging Faceotherupdated 2mo agoView on Hugging Face

0likes9downloads

rootdata

../

filetrain-00000-of-00015.parquet376.3 MBdownload

filetrain-00001-of-00015.parquet86.3 MBdownload

filetrain-00002-of-00015.parquet147.2 MBdownload

filetrain-00003-of-00015.parquet77.6 MBdownload

filetrain-00004-of-00015.parquet230.1 MBdownload

filetrain-00005-of-00015.parquet337.3 MBdownload

filetrain-00006-of-00015.parquet409.0 MBdownload

filetrain-00007-of-00015.parquet256.8 MBdownload

filetrain-00008-of-00015.parquet183.3 MBdownload

filetrain-00009-of-00015.parquet91.2 MBdownload

filetrain-00010-of-00015.parquet172.1 MBdownload

filetrain-00011-of-00015.parquet1.31 GBdownload

filetrain-00012-of-00015.parquet3.59 GBdownload

filetrain-00013-of-00015.parquet3.05 GBdownload

filetrain-00014-of-00015.parquet3.33 GBdownload

OctoReasoner/CodeGeneration-IQuest · main · files are served by the source, never re-hosted here

This repository is gated. The listing is public, but downloading a file means accepting the publisher’s terms at Hugging Face first — the links above take you there rather than around it.