INDEX

Explanations

making a statement

The neuron flags language about attention‐grabbing or “turn‐heads” statements that emphasize making a bold impression.

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

擽

-1.27

 agir

-1.25

 délais

-1.23

都想

-1.23

ᄇ

-1.20

seven

-1.20

 MINUTE

-1.20

ഀ

-1.18

ַע

-1.18

 preſent

-1.18

POSITIVE LOGITS

 both

1.69

by

1.63

on

1.56

at

1.52

 even

1.46

 people

1.45

 using

1.38

 just

1.30

if

1.30

 your

1.30

Activations Density 0.048%