fix: description text

Add WikiQA

fix: show partial results even if some evaluations haven't finished

fix: read request information even if eval is running

Update app.py

switch to flat inflection benchmark

add wrapping to leaderboard

add submission instructions to about page

remove submit tab

Update app.py

debug restart interval

fix: type hints for styling function

Factor out floating point styling to a function

fix: filtering support for models missing details

remove intro text and citation block

add benchmark descriptions and links to About page

Increase floating point number in benchmark metrics

add winogrande and arc-challenge

show private models by default

skip model detail validation for OAI/Anthropic models

fix typo in metric name

remove debug prints

fix metric name

add debug prints

revert to correct usage of ModelDetails (without api)

remove swp

debug print

debug print

debug print

debug print

debug print

debug print

Added empty default for api in ModelDetails

Added model API to submission screen

add Icelandic evals

switch to mideind's fork of Eval Harness

Change metric string

Comment out winogrande for debugging

Add task

Change title

Change title

Change title

Make name for HF token explicit

Fix repo names

Update src/envs.py

Update requirements.txt

Update README.md

Update app.py

doc

more info README