INDEX

Explanations

mount/mounting followed by object

The neuron activates on occurrences of the word stem “mount,” including its variants like “mounting,” “mounted,” and “mount.”

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

-1.57

-1.38

-1.33

of

-1.32

that

-1.31

-1.27

-1.24

is

-1.23

POSITIVE LOGITS

 apret

1.25

 sabor

1.22

on

1.17

dáms

1.17

 zelen

1.16

 logr

1.15

 presion

1.13

esseurs

1.12

restaurante

1.12

 glorious

1.12

Activations Density 0.014%