INDEX

Explanations

specific initial tokens followed by specific punctuation or words

The neuron detects numeric literals (both integers and floating‐point numbers) in the text.

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

‘

-1.41

ـ

-1.39

-1.37

£

-1.37

gestone

-1.35

at

-1.34

}}$

-1.34

-1.32

POSITIVE LOGITS

for

2.34

 where

1.54

 divertido

1.47

 encantador

1.44

 through

1.42

 that

1.42

ilustra

1.42

 hábitat

1.42

ganggu

1.42

 destacado

1.41

Activations Density 0.001%