INDEX

Explanations

under or below

The neuron fires on nouns that serve as the target of spatial relationships (e.g. the object after prepositions like under, behind, above).

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

 also

-1.52

astă

-1.44

way

-1.30

 šir

-1.28

ftet

-1.24

呿

-1.23

 séjours

-1.20

は以下の

-1.18

 nouveautés

-1.17

 solamente

-1.16

POSITIVE LOGITS

one

1.79

two

1.70

 three

1.43

the

1.36

for

1.35

 four

1.34

any

1.29

on

1.26

 went

1.19

("-",

1.19

Activations Density 0.098%