INDEX

Explanations

investigative

np_max-act · gemini-2.0-flash

This neuron detects mentions of investigative journalism roles and awards (e.g., “investigative,” “reporter,” “Pulitzer”).

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

ASM

-0.07

 disorder

-0.07

quina

-0.06

stype

-0.06

.RE

-0.06

ZD

-0.06

coop

-0.06

earch

-0.06

 THEN

-0.06

 rounding

-0.06

POSITIVE LOGITS

ّة

0.07

 Position

0.07

 masterpiece

0.06

Recently

0.06

 důsled

0.06

').↵

0.06

 United

0.06

yz

0.06

Activations Density 0.023%

investigative

This neuron detects mentions of investigative journalism roles and awards (e.g., “investigative,” “reporter,” “Pulitzer”).

No Comments

No Known Activations