INDEX

Explanations

specific domain terms

The neuron is detecting numeric tokens, especially decimal 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

 його

-1.55

it

-1.43

 его

-1.38

 where

-1.33

its

-1.29

してました

-1.23

 التي

-1.21

 armados

-1.18

 排行

-1.17

 their

-1.16

POSITIVE LOGITS

更为

1.43

眼前的

1.43

delse

1.42

vartis

1.37

onomies

1.33

ghalaya

1.30

greSQL

1.30

ttps

1.28

ască

1.27

 саны

1.27

Activations Density 0.096%