INDEX

Explanations

answer colon

This neuron detects numeric tokens (especially multi-digit and decimal numbers).

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

 even

-2.05

of

-2.05

-1.83

but

-1.77

-1.51

 nawet

-1.48

for

-1.47

 with

-1.45

-1.44

-1.43

POSITIVE LOGITS

考え方

1.55

ización

1.53

 differing

1.50

褙

1.50

熥

1.45

 fêtes

1.45

ありますか

1.43

ᵴ

1.42

 Какие

1.41

說道

1.41

Activations Density 0.071%