clarified how to get MMLU average

rm lighteval sha from id

added precision

fix rounding

corrected display of symbols

Merge branch 'main' into link_requests_and_results

get model type info from request file

added more nuance in ft models

Update src/auto_leaderboard/model_metadata_type.py (#159)

corrected finetuned mentions

Update newcomers (#153)

look for model type in request file

Update new model types (#150)

Fix Typo (#144)

add two more (#142)

More metadata type updates (#141)

Wording: number of parameters? (#138)

Add type to more models (#140)

model type filters (#139)

updated design to select columns to display

Short emoji legend on first page (#133)

only display the scores for the latest result file

Update model types (#126)

Added icons for types + fixed pending queue

wip adding symbols to model types

fix new config name

FT: precision and adapter models

Update src/assets/text_content.py

updated model param number reader

updated version

added precision for truthfulqa 6 shot

Small fix - we do not want to display models where the MMLU is old with models where the MMLU is new - however, since version is displayed in the results, we keep the files

Add details on the datasets for reproducibility (#107)

Using the new backend

small fix link Ilyas leaderboard

added harness command

revamp

column fix

merge refactor