INDEX

Explanations

specific words followed by prepositions

The neuron activates on numeric tokens—i.e. digits or numbers (mileages, counts, scores, years, 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

 different

-1.30

 divent

-1.11

 diferentes

-1.08

 various

-1.06

↵

-1.05

</h2>

-1.04

 things

-1.03

 necessaria

-1.02

 posteriore

-1.02

További

-1.00

POSITIVE LOGITS

on

1.38

 with

1.27

 from

1.25

in

1.22

for

1.20

ïde

1.12

at

1.11

 الاجتماعية

1.11

panjang

1.10

럴

1.10

Activations Density 0.067%