INDEX

Explanations

materials and products

This neuron specializes in spotting mentions of the raw components or substances used—words like “materials,” “ingredients,” 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

if

-1.86

now

-1.74

on

-1.56

 after

-1.55

 before

-1.52

 over

-1.51

an

-1.47

 about

-1.44

 just

-1.35

for

-1.34

POSITIVE LOGITS

to

1.66

Ostat

1.59

᛬

1.58

 THOSE

1.56

 hâte

1.53

ጧ

1.52

 asing

1.52

ああ

1.51

tą

1.51

のではないか

1.48

Activations Density 0.044%