INDEX

Explanations

UNITED

np_max-act · gemini-2.0-flash

This neuron detects the uppercase word “UNITED” (as in “UNITED STATES”) in legal document headers.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

(row

-0.08

aci

-0.06

#####↵

-0.06

 сим

-0.06

 quirky

-0.06

bit

-0.06

 کردند

-0.06

gim

-0.06

 broadcasts

-0.06

 Mick

-0.06

POSITIVE LOGITS

 Raise

0.07

 Create

0.07

 Wishlist

0.07

 automobile

0.06

OURSE

0.06

={"/

0.06

 requirements

0.06

\Entity

0.06

्ग

0.06

يران

0.06

Activations Density 0.001%

UNITED

This neuron detects the uppercase word “UNITED” (as in “UNITED STATES”) in legal document headers.

No Comments

No Known Activations