open_llm_leaderboard / src /auto_leaderboard

Commit History

Flagged model per discussion
6e039c4

Clémentine commited on

removed need for tokens in the leaderboard + removed skull in flagged models
a40c960

Clémentine commited on

Adding flagging systemi, removing changelog
699e8ff

Clémentine commited on

Adding link to detailed results and evals (#203)
6254b87

clefourrier HF staff commited on

Update src/auto_leaderboard/model_metadata_type.py
1108259

clefourrier HF staff commited on

Update src/auto_leaderboard/model_metadata_type.py (#199)
0ba9d60

clefourrier HF staff stellaathena commited on

look at model info if not in request file
a33e66d

Clémentine commited on

typo fix
1b7afb7

Clémentine commited on

Updated model metadata according to #162
a5023e4

Clémentine commited on

rm lighteval sha from id
2bb5ded

Clémentine commited on

added precision
6eaad72

Clémentine commited on

fix rounding
d350941

Clémentine commited on

corrected display of symbols
35763fc

Clémentine commited on

Merge branch 'main' into link_requests_and_results
a79408c

Nathan Habib commited on

get model type info from request file
80f4eeb

Nathan Habib commited on

added more nuance in ft models
95f85ed

Clémentine commited on

Update newcomers (#153)
d7daa68

multimodalart HF staff commited on

look for model type in request file
d295afa

Nathan Habib commited on

Update new model types (#150)
9977ce1

multimodalart HF staff commited on

add two more (#142)
5d5681a

multimodalart HF staff commited on

More metadata type updates (#141)
3602349

multimodalart HF staff commited on

only display the scores for the latest result file
d6b3d82

Nathan Habib commited on

Added icons for types + fixed pending queue
b323764

Clémentine commited on

wip adding symbols to model types
217b585

Clémentine commited on

fix new config name
4aff44e

Nathan Habib commited on

FT: precision and adapter models
12cea14

Clémentine commited on

updated model param number reader
1df8383

Clémentine commited on

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
97b27da

Clémentine commited on

Using the new backend
d16cee2

Linker1907 commited on

column fix
d52179b

Clémentine commited on

merge refactor
460d762

Clémentine commited on