0.2.13 (2026-02-26)
Features
- add Global MMLU task (#174) (0d0b227)
- add GoldenSwag task (#175) (a05e032)
- add tasks from the OLMES evaluation suite (#180) (54f295d)
- adding aggregated results with errors, if error free ration is < 1.0 (#181) (6f3e639)
- BalancedCOPA dataset (#177) (25161aa)
- Change to more complete revision of ZeroScrolls dataset (#171) (a4e117e)
- COPA uses appropriate dataset splits (#176) (55ebe44)
Bug Fixes
- Change to more complete revision of zeroscrolls (#173) (a84286e)
- Flores200 data reading issue (#179) (9bf3155)