INDEX

Explanations

the

np_max-act · gemini-2.0-flash

This neuron detects mentions of editorial or publishing roles (e.g. “edited,” “publisher,” journal or newspaper titles).

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

 Contracts

-0.07

 договор

-0.06

ENARIO

-0.06

ク

-0.06

 Kramer

-0.06

 marathon

-0.06

ρέ

-0.06

 Illegal

-0.06

.fromJson

-0.06

 드라마

-0.05

POSITIVE LOGITS

~(

0.07

 Integr

0.07

 presenting

0.07

lng

0.06

aN

0.06

forth

0.06

ژ

0.06

bif

0.06

(bounds

0.06

-feature

0.06

Activations Density 0.019%

the

This neuron detects mentions of editorial or publishing roles (e.g. “edited,” “publisher,” journal or newspaper titles).

No Comments

No Known Activations