INDEX

Explanations

control actions and consequences

The neuron fires on mentions of exerting physical force or effort (e.g. “exert a great force,” “operator attempts,” “user may have to exert,” “inaccessible” in a physical context).

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

 bạch

-1.38

 бампер

-1.33

iliki

-1.20

 Mereka

-1.19

 dưỡng

-1.11

 plot

-1.09

 listar

-1.07

即可

-1.06

INESS

-1.05

 экза

-1.05

POSITIVE LOGITS

are

1.35

:{

1.24

It

1.20

){

1.16

 notícias

1.12

 poemas

1.12

 ***!

1.12

it

1.11

他是

1.10

之为

1.09

Activations Density 0.224%