Hugging Face's logo

Datasets:
ulamai
/
STEM-RL-Tasks

Modalities:
Tabular
Text
Formats:
json
Languages:
English
Size:
< 1K
Tags:
stem
scientific-reasoning
reinforcement-learning
rlvr
verifiable-rewards
numerical-reasoning
Libraries:
Datasets
pandas
Polars
Dataset card Data Studio Files
xet
 Community
1
STEM-RL-Tasks / tasks /ASTRO2 /solution /solution.json
przemekch's picture
przemekch
Publish STEM RL Tasks: 40 tasks across eight domains with verifiers and recorded attempts
6395f35 verified
Raw History Blame Contribute Delete
236 Bytes
{
"problem_id": "ASTRO2",
"initial_horizon_angular_velocity_rad_s": 5691.6727,
"saturated_black_hole_spin_a_sat": 0.69461391,
"saturated_cloud_boson_number": 4.5212278e+77,
"observed_annihilation_gw_frequency_hz": 1063.9369
}