INDEX

Explanations

style="color:blue"href="

The neuron detects references to CSS stylesheets (e.g. the “stylesheet” attribute or “.css” in HTML tags).

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

 rafraî

-2.44

 rafra

-2.23

اقتصاد

-2.20

مواد

-2.19

 régal

-2.16

 間取り

-2.16

鵟

-2.16

Или

-2.14

疟

-2.14

Krank

-2.11

POSITIVE LOGITS

we

2.33

was

2.31

in

2.30

 along

2.19

is

2.09

 farben

2.08

 same

2.05

↵↵

2.03

1.99

 with

1.97

Activations Density 0.051%