INDEX

Explanations

sentimentalism and sentimentality

The neuron detects occurrences of words built from the “sentiment / senti-” root (e.g. “sentimental,” “sentimentality,” 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

ędzynarod

-2.41

merkmale

-2.41

⡦

-2.31

ritsar

-2.23

 mangled

-2.20

adillas

-2.08

腕

-2.06

잰

-2.05

emptive

-2.05

 postpon

-2.05

POSITIVE LOGITS

ization

3.25

ata

2.97

you

2.88

ский

2.30

ção

2.28

”,

2.23

</h3>

2.22

</strong>

2.16

un

2.14

on

2.11

Activations Density 0.003%