INDEX

Explanations

part of a count

The neuron activates on numeric tokens and number-like expressions (quantities, counts, or statistics).

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

is

-1.66

two

-1.41

 then

-1.35

and

-1.33

二

-1.28

 Then

-1.25

 still

-1.23

ActionCreators

-1.21

 jeszcze

-1.21

 dann

-1.21

POSITIVE LOGITS

 alguns

1.38

 tivi

1.34

 takže

1.31

cannot

1.29

 mattel

1.28

 密码

1.27

 octavia

1.25

 musíte

1.24

拼

1.23

 dacia

1.23

Activations Density 0.071%