INDEX

Explanations

questions and answers

np_max-act · gemini-2.0-flash

This neuron selectively activates on the special header‐ID tokens that mark speaker or section boundaries in the text.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

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

No Comments

Negative Logits

frame

-0.07

 Modules

-0.06

 immediate

-0.06

-frame

-0.06

 condemning

-0.06

etus

-0.06

 bryster

-0.06

fe

-0.06

ahun

-0.06

ieder

-0.06

POSITIVE LOGITS

''.

0.06

 influencers

0.06

il

0.06

TexCoord

0.06

Canadian

0.06

 fullPath

0.06

Squared

0.06

_people

0.06

	elseif

0.06

,{"

0.06

Activations Density 0.060%

questions and answers

This neuron selectively activates on the special header‐ID tokens that mark speaker or section boundaries in the text.

No Comments

No Known Activations