INDEX

Explanations

descriptive states and attributes

This neuron detects commercial or promotional product-advertising language (e.g., purchase calls, donation appeals, and marketing descriptors).

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

ㅋㅋㅋㅋㅋㅋㅋㅋ

1.09

................

1.07

https

1.01

0.97

 https

0.96

0.95

================

0.93

----------------

0.86

****************

0.82

 gerçekleştir

0.81

POSITIVE LOGITS

 socalled

2.02

 wellknown

1.70

$%

1.55

 andRow

1.34

 longterm

1.30

 、,

1.23

 അദ്ദേ

1.14

 സെക്ര

1.08

 Tattha

1.04

Activations Density 0.164%