INDEX

Explanations

author's transl

The neuron fires on tokens that label publication media or metadata (e.g. “electronic,” “ebk,” “transl”) in reference or bibliographic entries.

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

ichert

-1.48

%;

-1.38

 других

-1.37

 intrepid

-1.36

Mittwoch

-1.34

 той

-1.34

=\"

-1.34

い

-1.31

 از

-1.30

rawdę

-1.27

POSITIVE LOGITS

1.81

 allemaal

1.41

 hiciera

1.37

 should

1.34

monger

1.34

 kendisi

1.34

 観光

1.30

 вечера

1.29

اته

1.29

lishes

1.29

Activations Density 0.001%