INDEX

Explanations

What is the significance / purpose

The neuron fires on question terms that ask for a quantitative or evaluative measure—words like “value,” “impact,” “significance,” “purpose,” “role,” or “thoughts” in a “What is the … of …?” 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

 erhi

-1.54

might

-1.43

 Fifteen

-1.39

𔘓

-1.36

idelberg

-1.35

が付いて

-1.34

 spectacular

-1.32

 decidedly

-1.31

描写

-1.31

atien

-1.30

POSITIVE LOGITS

 there

2.42

 then

2.27

 this

2.17

for

1.95

if

1.93

 when

1.87

 after

1.84

you

1.73

 before

1.65

 other

1.62

Activations Density 0.074%