INDEX

Explanations

curse

np_max-act · gemini-2.0-flash

The neuron primarily detects the word “curse,” especially when it appears in the phrase “curse of.”

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

坦

-0.07

ibBundleOrNil

-0.07

学会

-0.06

aporation

-0.06

 Eaton

-0.06

Gay

-0.06

 brighter

-0.06

 Backend

-0.06

()})↵

-0.06

abstractmethod

-0.06

POSITIVE LOGITS

 curse

0.13

 cursed

0.10

 Curse

0.09

 curses

0.09

ursed

0.07

RegularExpression

0.07

чки

0.07

 sacred

0.07

ку

0.07

 psychic

0.07

Activations Density 0.003%

curse

The neuron primarily detects the word “curse,” especially when it appears in the phrase “curse of.”

No Comments

No Known Activations