Twave
LordTwave
·
AI & ML interests
None yet
Organizations
None yet
LordTwave's activity
Model is Overaligned, Unusable and gamed for the leaderboard
10
#17 opened about 1 year ago
by
distantquant

LMSYS Leaderboard? I want human evaluations:)
#27 opened 10 months ago
by
LordTwave

Model is paraphrasing text instead of citing it verbatim
3
#7 opened 11 months ago
by
sszymczyk
How did you manage to get your GSM8K a full 1.9 percentage points up from a 15T token trained model?
1
#7 opened 10 months ago
by
LordTwave

85.44 GSM8K Top on HF - New Record!
1
#22 opened 11 months ago
by
LordTwave

No Baseline (yet?)
1
#2 opened 12 months ago
by
LordTwave

ARC 77.73, HellaSwag 91.88, TOP under 22B - Three new HF Records!
2
#4 opened about 1 year ago
by
LordTwave

91.9 HellaSwag, 79.2 TruthfulQA... And It Sucks. Why do this?
9
#5 opened 12 months ago
by
deleted
Highest on HF Leaderboard!
#2 opened 12 months ago
by
LordTwave

Small Typo - it's Abacus.AI not Albacus.Ai
2
#1 opened about 1 year ago
by
bindureddy
Congrats on the overwhelming MMLU 85.6 score!
1
#1 opened about 1 year ago
by
LordTwave
