INDEX

Explanations

evokes comparisons and memories

The neuron detects the common comparative phrase “reminds (or reminded) me of,” i.e. when the text uses “reminds me of” to draw a similarity.

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

-1.89

 lakini

-1.73

-1.70

 щоб

-1.66

蹿

-1.65

-1.64

箆

-1.59

჻

-1.59

鯔

-1.56

してくれた

-1.55

POSITIVE LOGITS

ጵ

1.58

？」

1.45

guerite

1.41

庥

1.41

 pantai

1.38

 sözler

1.38

 ſame

1.30

 şöyle

1.30

fancybox

1.29

«

1.28

Activations Density 0.005%