lawbench@1.0
Professional
Law
Knowledge
LawBench: Benchmarking Legal Knowledge of Large Language Models.
Run this task
CLI:
inspect eval inspect_harbor/lawbench_1_0 --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import lawbench_1_0
eval(lawbench_1_0(), model="openai/gpt-5")Dataset information
| Harbor registry | lawbench@1.0 |
| Inspect task | lawbench_1_0 |
| Version | 1.0 |
| Samples | 1000 |
| Paper | arxiv |
See Task Parameters for the parameter set shared across all Harbor tasks.