INDEX

Explanations

leap forward or leaps

The neuron responds specifically to the word “leap” (and its inflected forms like “leapt”) whenever it appears.

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

玧

-2.58

柃

-2.44

霎

-2.36

CURLOPT

-2.23

聍

-2.19

还挺

-2.17

怎么说

-2.14

少爷

-2.13

⫘

-2.13

 большую

-2.05

POSITIVE LOGITS

╿

2.70

 стоковая

2.59

所以

2.56

うわ

2.41

rdoba

2.39

 plupart

2.39

ଞ

2.38

鵟

2.34

 hadapan

2.33

 люди

2.31

Activations Density 0.004%