INDEX

Explanations

fish and chips

The neuron fires on occurrences of the words “fish” or “chip(s),” especially in the context of “fish and chips.”

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

 that

-1.75

 definido

-1.66

瓩

-1.55

 їх

-1.51

 และ

-1.49

 princípios

-1.48

 españa

-1.45

妤

-1.44

fieras

-1.39

ジュニア

-1.39

POSITIVE LOGITS

мам

1.52

有很多

1.39

Опис

1.38

ный

1.38

нето

1.33

кто

1.30

 Daarna

1.30

리는

1.29

 manchen

1.29

有不少

1.28

Activations Density 0.002%