INDEX

Explanations

code contribution and plugins

The neuron is triggered by tokens forming the substring “contrib,” i.e. words related to “contribute,” “contribution,” or “contributor.”

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

 other

-2.42

 time

-2.06

⸰

-2.02

chemicals

-1.91

 types

-1.91

стреча

-1.87

苹

-1.84

всегда

-1.83

 inaugural

-1.81

 checkout

-1.79

POSITIVE LOGITS

you

2.38

 こんな

2.25

 innova

2.06

焿

1.97

ции

1.96

絰

1.90

様な

1.89

ations

1.88

ной

1.87

Activations Density 0.002%