INDEX

Explanations

referring to a specific subject

The neuron responds to occurrences of the demonstrative “this.”

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

 with

-2.00

for

-1.84

on

-1.55

 through

-1.46

to

-1.44

 experimentado

-1.38

 inclusief

-1.37

 reacts

-1.33

 from

-1.32

 عاما

-1.30

POSITIVE LOGITS

↵

1.67

 даже

1.62

 formid

1.35

CHILDREN

1.30

 implic

1.27

 maailman

1.27

 jusqu

1.25

 kuitenkin

1.25

但是

1.24

 patrio

1.24

Activations Density 0.093%