Skip to content

Benchmark for assessing legal capabilities of LLM agents

Notifications You must be signed in to change notification settings

dluo96/lawful-good

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Lawful Good

Get Started

Clone the repo:

git clone [email protected]:dluo96/lawful-good.git

Install the inspect_ai and inspect_evals Python packages:

pip install inspect_ai
pip install git+https://github.com/UKGovernmentBEIS/inspect_evals

Install the package locally

cd lawful-good
pip install -e .

Example Evaluation with Ollama

  • Install ollama. On Linux, simply run
    curl -fsSL https://ollama.com/install.sh | sh
  • Run (say) Llama2 3B:
    ollama run llama3.2
    
    This will automatically pull (download) Llama2 3B for you and then run.
  • To run Inspect evals with Ollama, you will need to install the openai package:
    pip install openai
    
  • Run the example evaluation script:
    inspect eval lg/lawful_good.py --model ollama/llama3.2
  • NOTE: if you receive the error PermissionError: [Errno 13] Permission denied: '/run/user/1000', try setting the following environment variable:
    export XDG_RUNTIME_DIR=/tmp
    and re-running.

Analyse evaluation runs with Inspect view

inspect view --log-dir lg/logs

About

Benchmark for assessing legal capabilities of LLM agents

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages