INDEX

Explanations

book titles/sections

The neuron activates on numeric tokens used as page or chapter numbers in tables of contents.

New Auto-Interp

Configuration

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 denně

-0.07

 дії

-0.07

ice

-0.07

 kvinne

-0.07

 рам

-0.06

ním

-0.06

 нав

-0.06

gay

-0.06

(bin

-0.06

ैय

-0.06

POSITIVE LOGITS

0.09

0.08

↵

0.08

0.07

Activations Density 0.007%