INDEX

Explanations

starts of phrases

The neuron consistently lights up on very common short function words (stopwords) such as “is,” “in,” “of,” “it,” contractions like “‘s” or “’t,” “can,” “out,” etc.

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

-1.29

 другую

-1.26

 llena

-1.26

 опять

-1.22

чем

-1.20

 ежедневно

-1.20

 irgend

-1.18

--

-1.18

𝑂

-1.17

).

-1.16

POSITIVE LOGITS

 they

1.44

by

1.41

you

1.40

it

1.34

 have

1.30

鎔

1.27

נם

1.27

so

1.27

velli

1.27

⊺

1.26

Activations Density 0.116%