added harness command
d2e8eca
Clémentine
commited on
Spaces:
Runtime error
Runtime error
Commit History
revamp
6e8f400
Clémentine
commited on
column fix
d52179b
Clémentine
commited on
merge refactor
460d762
Clémentine
commited on
Copy change
ce824ba
Fix elo ratings model links
e05ec6c
Add custom url for second tab
7644705
Still return tab without query params
6a6e05c
Link to discussion with custom url
8cb7546
Update tab button
b5f5045
Update app.py
7a429ab
Add GPT-4 & human eval tab
0227006
Add search emoji
92ae76d
Search on ENTER
48c5442
Increase concurrency count
f458f0b
import datetime correctly
d35aee2
record submitted time
8696209
style clean up
aa7c3f4
implements search bar
ffefe11
Add citation button
2a73469
Simply layout
c131125
clean up vars
4c8dd3c
sync with the internal version
58733e4
Auto-restart every hour
46f8d78
start every 20 minutes
9567fa6
use the same H4_TOKEN for restart
0a3d32f
use BackgroundScheduler to restart space
10f9b3c
rename block to demo
01233b7
sort imports and import BackgroundScheduler
4596a70
Tweak change log style
50a344f
Add CHANGELOG
6ed68b6
Add a baseline
35a0978
Update app.py
d3fbe10
copy edit
80ac3c2
do not display incomplete models for now
1363c8a
reject duplicate submission
f742519
style update
a885f09
display message after eval submission
85dbbc4
Fix TruthQA typo
614ee1f
Update app.py
7eee6bd
Automatically refresh the leaderboard when loading the space (#8)
3693dc6
Tweak preamble text (#1)
e4b25b8
refining for release
07bfeca
edbeeching
commited on
updated text
a460f7a
edbeeching
commited on
added public option
db6f218
edbeeching
commited on
fixes mixup with 8bit eval and private
5cb1426
edbeeching
commited on
adds base model for delta weight merging
a095268
edbeeching
commited on
updates table to include revision
fcb01e3
edbeeching
commited on
adds revision option
b2c063a
edbeeching
commited on
refactoring leaderboard
f90ad24
edbeeching
commited on