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

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

2.19Forward BPC

Transformer (ctx=1024)

Updated 1d ago

Evaluation Results

Method	Links
Transformer (ctx=1024) 2026.05		2.19
Transformer (ctx=1024) 2026.05		2.22
SHARP 2026.05		2.36
GRU 2026.05		2.57
RNN 2026.05		2.7
Clockwork RNN 2026.05		2.88
LSTM 2026.05		3.17