INDEX

Explanations

This neuron detects comparative-analogy cue words (e.g. “similar,” “like,” “same”) that introduce analogies or similes.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 that

-1.55

闈

-1.48

軀

-1.41

↵↵

-1.39

 able

-1.38

 pantal

-1.34

đầm

-1.33

雠

-1.30

 femei

-1.29

 île

-1.27

POSITIVE LOGITS

{~

1.48

tisgarh

1.33

我们在

1.24

 июля

1.21

 навіть

1.20

 bezeich

1.20

fragt

1.19

釓

1.19

 descrit

1.18

Namara

1.17

Activations Density 0.033%