INDEX

Explanations

,

np_max-act · gemini-2.0-flash

This neuron strongly activates on imperative verbs and command‐style phrases (e.g. “Update:…,” “Go Beto!,” “Do not move or drive…,” “Vote like…,” “Nothing stopping…”), i.e. text that issues instructions or calls to action.

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

NewLabel

-0.07

 vuông

-0.07

という

-0.06

(Group

-0.06

 perfil

-0.06

vanished

-0.06

่าการ

-0.06

 peeled

-0.06

itur

-0.06

Chain

-0.06

POSITIVE LOGITS

арх

0.07

jh

0.07

 Comparison

0.07

depends

0.06

 psychological

0.06

字符

0.06

ιας

0.06

컴

0.06

ิทธ

0.06

طن

0.06

Activations Density 0.006%

,

This neuron strongly activates on imperative verbs and command‐style phrases (e.g. “Update:…,” “Go Beto!,” “Do not move or drive…,” “Vote like…,” “Nothing stopping…”), i.e. text that issues instructions or calls to action.

No Comments

No Known Activations

,

This neuron strongly activates on imperative verbs and command‐style phrases (e.g. “Update:…,” “Go Beto!,” “Do not move or drive…,” “Vote like…,” “Nothing stopping…”), i.e. text that issues instructions or calls to action.

No Comments

No Known Activations