Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Next-word prediction on Wikitext-103
Loading...
18.31
Perplexity
Transformer-XL Large
17.2732
24.2716
31.27
38.2684
Jul 22, 2021
May 9, 2022
Feb 24, 2023
Dec 13, 2023
Sep 29, 2024
Jul 17, 2025
May 5, 2026
Perplexity
Updated 28d ago
Evaluation Results
Method
Method
Links
Perplexity
Transformer-XL Large
N_param (Transformer)=...
2021.07
18.31
Transformer-XL Medium
N_param (Transformer)=...
2021.07
24.23
FNetAR Medium
N_param (Transformer)=...
2021.07
25.81
Yat (pn+α)
Parameterization=per-n...
2026.05
39.04
Yat (sb+α)
Parameterization=share...
2026.05
39.07
Yat (sb+ca)
Parameterization=share...
2026.05
42.09
Yat (pn+ca)
Parameterization=per-n...
2026.05
42.18
GELU
Architecture=GELU MLP,...
2026.05
44.23
Feedback
Search any
task
Search any
task