INDEX

Explanations

deterrence

np_max-act · gemini-2.0-flash

The neuron responds specifically to the word “deterrence” (and its variants like “deterrent”), activating on the morphemes “deter-” and “-rence.”

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

ern

-0.08

-Mar

-0.07

/people

-0.06

 Keller

-0.06

-default

-0.06

imiento

-0.06

(sem

-0.06

pre

-0.06

SW

-0.06

POSITIVE LOGITS

HOR

0.07

 glColor

0.07

 căn

0.06

pname

0.06

.getUsername

0.06

 splitter

0.06

 carn

0.06

 Challenger

0.06

 phil

0.06

/connection

0.06

Activations Density 0.028%

deterrence

The neuron responds specifically to the word “deterrence” (and its variants like “deterrent”), activating on the morphemes “deter-” and “-rence.”

No Comments

No Known Activations