Share your thoughts, 1 month free Claude Pro on usSee more

Character-level Language Modeling on text8 100M regime (Current stream split)

2.12Current BPC

Transformer (ctx=1024)

Updated 1d ago

Evaluation Results

Method	Links
Transformer (ctx=1024) 2026.05		2.12
Transformer (ctx=1024) 2026.05		2.15
SHARP 2026.05		2.32
GRU 2026.05		2.5
Clockwork RNN 2026.05		2.85
LSTM 2026.05		3.16
RNN 2026.05		4.63