INDEX

Explanations

head coach

The neuron detects references to someone’s role as a head coach (e.g. “head coach,” “coached by”).

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

</td>

-1.54

 które

-1.52

菝

-1.41

 seleccionamos

-1.41

 které

-1.38

那你

-1.36

 donde

-1.34

 mangrove

-1.34

镡

-1.34

 jossa

-1.33

POSITIVE LOGITS

to

1.59

at

1.54

as

1.41

 years

1.40

 autorización

1.35

 people

1.34



1.31

is

1.30

 жид

1.29

 while

1.28

Activations Density 0.048%