INDEX

Explanations

places of refuge and escape

The neuron is triggered by nouns denoting idyllic or peaceful places/spaces (e.g., haven, paradise, oasis, retreat, solace).

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

 tinge

-1.63

檮

-1.58

 kwietnia

-1.56

 czerwca

-1.55

觧

-1.54

dịch

-1.49

 dieną

-1.46

 poži

-1.41

很赞

-1.41

 stycznia

-1.41

POSITIVE LOGITS

 haven

1.82

他の

1.73

↵↵

1.64

of

1.56

 extranjera

1.55

1.45

by

1.45

 extranjeros

1.41

(?)

1.40

にして

1.36

Activations Density 0.026%