Spaces:
Restarting
on
CPU Upgrade
Restarting
on
CPU Upgrade
Commit History
fix ModelType is not defined
67c432c
change 'proprietary' models to 'external' models and added news models
91c6e89
update ranges
5651102
verified
update collection format
9c5c692
Update src/display/about.py
183ec61
verified
Make model text exibit precision if there's more than one eval or precision is not float16 or bfloat16
59399bc
fix typo and multiple models in README
b4fc70b
Merge branch 'main' of https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard
f49e1e5
Permit different revision
a3b0a0f
Update src/display/about.py
77a5f61
verified
better explation for use_remote_code=True
a761927
add citations
c14ac9f
Add proprietary model results v1
1dbfacb
Change language dropdown order
1ec02f0
Update model list in README
2331e6f
Add new column: Main Language
6da7311
change name
88675db
Rename 85B+ format to 100B+
73d86a6
verified
add dynamic documentation for RAW_RESULTS_REPO
43c2b1a
Add raw results links if exists, and fix minor issues
aa7060a
new size intervals and apply same intervals for the collection
7625ef6
Add env variable SHOW_INCOMPLETE_EVALS and order evaluation queue by priority
8aaf0e7
Allow old model metrics
6269bd0
Add NPM field
f976f1c
Add new tasks and make leadboard work without new tasks evals
5639a81
add datasources links on the about page
03f7287
add changelog tab
b74e881
Better about's
67cd6fc
Portuguese Tasks configs and fix bugs
c0fa950
Fix typo on ModelType
155aef4
New model_type: language adapt
395f537
Refactor Tasks to load by yaml configuration file
4445ad2
Feature: FIELD with original HF Leaderboard ranking
71ecfbb
Merge Origin - Rename model types (#1)
9839977
verified
Evaluation time metric and plot
359d8a9
TRUST_REMOTE_CODE env variable
1253c3b
show baseline
1b2e131
change icon to rocket
eef299c
Fix model eval links and remove huggingface icon from Leaderboard name
439afd4
Parametrize documentation
7499b46
New tasks difinitions
3449f84
Refactor code for adding generic tasks
36e3010
Enable clean initialization without any evaluate model
ec3a730
Explain where reproducibility command is (#542)
06919ed
verified
simplified display, added an extra config repo to carry dynamic information
9b2e755
Clémentine
commited on
full width css!
3df8919
Clémentine
commited on
Added checkbox for merges
b762711
Clémentine
commited on
adding merge check - super slow but at least info is displayed
20b060e
Clémentine
commited on