INDEX

Explanations

Clergy members

np_max-act · gemini-2.0-flash

The neuron detects terms referring to church offices or clergy roles (e.g. “minister,” “priest,” “vicar,” “chaplain,” etc.).

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

宏

-0.07

روس

-0.07

 Stanton

-0.07

->{_

-0.07

rea

-0.06

SequentialGroup

-0.06

 اد

-0.06

Day

-0.06

ilyn

-0.06

POSITIVE LOGITS

 каб

0.08

 Cave

0.07

 şans

0.07

 schematic

0.07

ocab

0.07

CAB

0.07

ництво

0.07

 SECOND

0.06

 mohla

0.06

erde

0.06

Activations Density 0.032%

Clergy members

The neuron detects terms referring to church offices or clergy roles (e.g. “minister,” “priest,” “vicar,” “chaplain,” etc.).

No Comments

No Known Activations

Clergy members

The neuron detects terms referring to church offices or clergy roles (e.g. “minister,” “priest,” “vicar,” “chaplain,” etc.).

No Comments

No Known Activations