INDEX

Explanations

introduces, notified, able, responsible, annoying, listed, brings

The neuron strongly responds to second-person pronouns—especially “you”—and other common function words used when directly addressing or instructing the reader.

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

-0.90

by

-0.88

eaked

-0.88

be

-0.87

 그의

-0.86

issaient

-0.86

냇

-0.83

 Handwerk

-0.81

zyst

-0.81

 ACTUAL

-0.80

POSITIVE LOGITS

 will

2.75

都会

1.36

 будет

1.34

 would

1.31

会

1.26

會

1.25

 sẽ

1.21

 θα

1.20

🥇

1.12

จะ

1.11

Activations Density 0.087%