INDEX

Explanations

code delimiters and suffixes

The neuron spikes on parenthesis characters, especially opening “(” tokens.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

-1.49

ization

-1.40

他們

-1.30

Look

-1.27

slika

-1.24

 değişik

-1.22

基本的に

-1.20

isation

-1.18

行って

-1.17

POSITIVE LOGITS

 bienven

1.28

 rendus

1.27

獁

1.23

 türk

1.18

﹥

1.17

'));

1.16

 rámci

1.13

ünde

1.12

 iban

1.11

Activations Density 0.011%