INDEX

Explanations

funding and dollars

The main thing this neuron does is detect mentions of funding or financial support.

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

 your

-1.59

all

-1.41

and

-1.38

-1.32

-1.28

or

-1.21

>=

-1.20

-1.19

begin

-1.19

at

-1.18

POSITIVE LOGITS

镡

1.48

 火锅

1.38

 bateria

1.34

——–

1.30

 内衣

1.28

 accompagner

1.27

 provinciale

1.26

 ANDREW

1.26

———-

1.25

楽しめる

1.23

Activations Density 0.067%