INDEX

Explanations

or

np_max-act · gemini-2.0-flash

The neuron is specifically detecting the word “negative.”

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

 renting

-0.07

 succession

-0.07

 angles

-0.07

ز

-0.07

NavigationBar

-0.07

AllArgsConstructor

-0.06

-central

-0.06

폴

-0.06

_change

-0.06

 chocolates

-0.06

POSITIVE LOGITS

Iter

0.06

 arcane

0.06

oci

0.06

barcode

0.06

chied

0.06

hwnd

0.06

–

0.06

 Very

0.06

 tenía

0.06

�

0.06

Activations Density 0.005%

or

The neuron is specifically detecting the word “negative.”

No Comments

No Known Activations