INDEX

Explanations

mathematical symbols and notation

This neuron flags tokens that come from LaTeX‐style math (i.e. math‐mode symbols, variables, superscripts/subscripts, and related notation).

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

 that

-1.18

ｿ

-1.01

Even

-1.01

ambut

-0.97

◒

-0.96

on

-0.94

 cápsulas

-0.94

 what

-0.94

şil

-0.93

 Verfassung

-0.91

POSITIVE LOGITS

^{

1.68

<sup>

1.43

^{\

1.41

^(

1.24

1.20

}^{

1.18

$^{

1.15

 toppen

1.04

 steen

1.01

了一会儿

0.97

Activations Density 0.053%