reasonwang/skillgym-test download history
reasonwang/skillgym-test is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 5,977 times (5,977 in the last 7 days), and 5,977 times in total. It ranks #4,884 among datasets by monthly downloads.
SkillGym test set 400 agent tasks that each require one agent skill, in Harbor task format. Layout tasks/ procedural/ held_in/ 50 tasks held_out/ 50 tasks constraint_satisfaction/ (same) abductive/ (same) part
Spaces using skillgym-test
- Model Pulse 8 likes