INDEX

Explanations

assessing worthiness

The neuron detects references to monetary value or cost–value judgments (prices, spending, worth, money amounts).

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

டுக

-1.23

中心的

-1.20

zzet

-1.20

 beberapa

-1.16

scape

-1.15

 تغي

-1.13

 bridge

-1.12

媗

-1.10

펜

-1.09

POSITIVE LOGITS

1.45

1.31

 próprias

1.16

1.14

1.12

 можна

1.07

<blockquote>

1.07

讐

1.06

 regardless

1.05

Activations Density 0.029%