INDEX

Explanations

date, Home, status, click, success

This neuron specializes in spotting human-readable string literals (display text) in code or docs.

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

 Drž

-1.38

 Livro

-1.37

 inginkan

-1.29

༷

-1.28

不知道

-1.27

 kompati

-1.24

 espet

-1.23

 Projektu

-1.21

にならない

-1.21

 silikon

-1.21

POSITIVE LOGITS

Hilsen

1.44

但却

1.35

 during

1.33

窺

1.28

紡

1.23

價值

1.19

معرفی

1.19

<bos>

1.19

などを

1.18

 after

1.17

Activations Density 0.005%