Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Language Modeling on PG-19 subword-level
Loading...
3.94
Forward BPT
Transformer-23M
3.916
4.078
4.24
4.402
May 30, 2026
Forward BPT
Current BPT
Backward BPT
Updated 1d ago
Evaluation Results
Method
Method
Links
Forward BPT
Current BPT
Backward BPT
Transformer-23M
T=256
2026.05
3.94
3.76
3.94
Transformer-18M
T=256
2026.05
3.95
3.77
3.94
SHARP
T=4
2026.05
4.26
4.26
4.26
GRU
T=4
2026.05
4.5
4.52
4.51
LSTM
T=4
2026.05
4.52
4.54
4.53
RNN
T=4
2026.05
4.54
4.56
4.55
Feedback
Search any
task
Search any
task