INDEX

Explanations

div `dx` `dy`

The neuron fires on closing delimiters that mark the end of code or markup blocks—e.g. closing braces, endfor/end tags, and closing 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

ḱ

-1.02

íqu

-0.98

 katk

-0.96

Santé

-0.95

 kişilik

-0.94

اعي

-0.94

biete

-0.91

энер

-0.91

 classifica

-0.90

RemindMe

-0.89

POSITIVE LOGITS

乐观

0.90

0.85

 się

0.79

bih

0.79

 второй

0.78

 поддер

0.78

Then

0.77

itored

0.77

Activations Density 0.001%