INDEX

Explanations

quotation marks and thanks

The neuron strongly activates on the opening and closing quotation‐mark punctuation (and adjacent sentence boundaries) that frame user testimonial quotes.

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

 BSEB

-1.13

 guardare

-1.12

https

-1.09

 domani

-1.09

 beliebt

-1.07

为了

-1.05

:—

-1.04

在这个

-1.03

ditionally

-1.03

如果

-1.03

POSITIVE LOGITS

 your

2.67

 thank

1.56

1.55

 Your

1.48

you

1.44

你们的

1.40

 Thank

1.27

 votre

1.23

 thanks

1.21

 vaše

1.20

Activations Density 0.031%