INDEX

Explanations

prepositional phrases

The neuron primarily detects numerical values written with decimal points (i.e., floating-point numbers).

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

can

-1.49

 would

-1.43

 allows

-1.34

 from

-1.30

 what

-1.29

 americana

-1.23

by

-1.21

 allow

-1.20

of

-1.19

 like

-1.19

POSITIVE LOGITS

℣

1.82

the

1.64

They

1.59

Their

1.48

嵓

1.41

不管是

1.38

 écoles

1.34

 toujours

1.32

nto

1.31

(<

1.30

Activations Density 0.190%