INDEX

Explanations

separate or independent entity

This neuron detects occurrences of words expressing separation or distinction (most strongly the word “separate” and its variants).

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

 також

-1.73

 where

-1.43

 around

-1.34

 zwłaszcza

-1.28

 дори

-1.27

 także

-1.27

 Menurut

-1.27

 Meski

-1.26

Despite

-1.25

瑧

-1.25

POSITIVE LOGITS

 than

2.02

tidig

1.40

thanx

1.34

{'

1.28

ఽ

1.27

‘

1.26

濤

1.21

ſſen

1.20

‣

1.19

Equ

1.16

Activations Density 0.116%