INDEX

Explanations

digits followed by }

The neuron is detecting numeric tokens (digits and numbers).

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

ed

-1.55

of

-1.50

-1.36

ched

-1.34

-1.33

少了

-1.32

消失了

-1.31

es

-1.27

変わらず

-1.27

POSITIVE LOGITS

It

1.55

ари

1.52

intos

1.50

hånd

1.45

1.43

gemeins

1.43

 oeste

1.39

awarkan

1.38

spectos

1.38

〤

1.37

Activations Density 0.027%