INDEX

Explanations

novel followed by a descriptor

The neuron is tuned to detect mentions of the word “novel” (and its close variants like “novels”) in the text.

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

 houſe

-1.13

考虑到

-0.94

ियों

-0.91

 Minangkabau

-0.90

楼

-0.89

 bastion

-0.86

 Inſ

-0.85

은

-0.84

dann

-0.84

Zdrav

-0.82

POSITIVE LOGITS

 coronavirus

1.59

ties

1.59

istic

1.34

 approach

1.28

las

1.17

Coronavirus

1.09

 Coronavirus

1.06

ty

1.02

tys

1.02

Approach

1.00

Activations Density 0.021%