Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Autoregressive Language Modeling on WikiText-103 1024 (test)
Loading...
19
PPL
Angular
18.804
20.127
21.45
22.773
Oct 5, 2025
PPL
Updated 1mo ago
Evaluation Results
Method
Method
Links
PPL
Angular
gamma=8, Sequence Leng...
2025.10
19
RACE
P=4, L=4, Sequence Len...
2025.10
20.9
FlashAttention2
Sequence Length=1024
2025.10
20.9
RACE
P=3, L=4, Sequence Len...
2025.10
21.5
RACE
P=3, L=3, Sequence Len...
2025.10
21.9
RACE
P=2, L=3, Sequence Len...
2025.10
23.4
RACE
P=2, L=2, Sequence Len...
2025.10
23.9
Feedback
Search any
task
Search any
task
