INDEX

Explanations

more than

This neuron detects comparative quantity phrases, especially “more than …,” that indicate amounts exceeding a specified number.

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

and

-1.70

new

-1.45

 hábiles

-1.40

-1.39

 from

-1.36

følge

-1.30

 with

-1.30

-1.26

)+"

-1.23

嗥

-1.20

POSITIVE LOGITS

vê

1.41

 случае

1.41

 preponder

1.39

 importantly

1.39

nthetic

1.38

 době

1.36

تیجه

1.35

컹

1.35

initas

1.35

ně

1.34

Activations Density 0.017%