INDEX

Explanations

starting new phases or topics

The neuron fires on long runs of leading spaces (indentation) at the starts of lines.

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

 with

-1.78

 nestled

-1.75

 Providing

-1.70

if

-1.63

in

-1.62

之类的

-1.56

at

-1.54

臧

-1.52

か月

-1.47

-1.46

POSITIVE LOGITS

咴

2.22

Ꭳ

2.05

 isolamento

1.97

 alcune

1.90

 semelh

1.79

蚡

1.78

菒

1.78

硖

1.77

uſe

1.75

巂

1.73

Activations Density 0.003%