-
Add shard train-00016-of-00018.parquet (18/20)
-
Add shard train-00015-of-00018.parquet (17/20)
-
Add shard train-00014-of-00018.parquet (16/20)
-
Add shard train-00013-of-00018.parquet (15/20)
-
Add shard train-00012-of-00018.parquet (14/20)
-
Add shard train-00011-of-00018.parquet (13/20)
-
Add shard train-00010-of-00018.parquet (12/20)
-
Add shard train-00009-of-00018.parquet (11/20)
-
Add shard train-00008-of-00018.parquet (10/20)
-
Add shard train-00007-of-00018.parquet (9/20)
-
Add shard train-00006-of-00018.parquet (8/20)
-
Add shard train-00005-of-00018.parquet (7/20)
-
Add shard train-00004-of-00018.parquet (6/20)
-
Add shard train-00003-of-00018.parquet (5/20)
-
Add shard train-00002-of-00018.parquet (4/20)
-
Add shard train-00001-of-00018.parquet (3/20)
-
Add shard train-00000-of-00018.parquet (2/20)
-
Add shard test-00000-of-00001.parquet (1/20)
-
5GB: update README for sharded parquet 50M
-
Viewer fix: keep only default parquet config (json still in repo as files, not viewer config) + Problem Types above Quickstart
-
Fix viewer: json array -> jsonl lines + move Problem Types above Quickstart
-
Docs: clarify parquet/json dual folders and working load_dataset examples
-
Remove legacy data/ folder - migrated to parquet/ + json/
-
Reorganize into parquet/ and json/ folders + remove em dashes - verified dual configs
-
Remove license comparison section
-
Initial release: 150K verified arithmetic reasoning (v4) with tiered difficulty
-
initial commit
