INDEX

Explanations

young [group/entity]

The neuron detects nouns that immediately follow the adjective “young” (i.e. the head words of “young X” phrases).

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

-1.55

 another

-1.34

another

-1.29

 particular

-1.27

 normal

-1.24

 usual

-1.23

 aguj

-1.21

 human

-1.20

the

-1.20

 food

-1.19

POSITIVE LOGITS

of

1.85

(<

1.52

 will

1.48

 altercation

1.32

outchou

1.21

писы

1.20

chés

1.18

 новый

1.18

uffff

1.18

ESTADO

1.16

Activations Density 0.043%