INDEX

Explanations

unpacking gender, romance, diverse

The neuron detects high-level business‐and‐tech jargon and action terms commonly used in product, strategy, and growth pitches.

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

ってない

-1.34

 дорогой

-1.30



-1.16

⤒

-1.16

 béisbol

-1.16

ったのは

-1.16

※

-1.14

 indios

-1.13

其他

-1.13

 earlier

-1.12

POSITIVE LOGITS

my

2.02

1.83

1.16

 также

1.14

 również

1.13

 всегда

1.09

ціє

1.07

 également

1.06

 también

1.05

 моя

1.04

Activations Density 0.067%