INDEX

Explanations

potential threats

np_max-act · gemini-2.0-flash

The neuron is triggered by words signaling security or safety alerts—especially “threat,” “bomb,” and other suspicious or evacuation-related terms.

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

.Play

-0.06

_activ

-0.06

什

-0.06

 Percy

-0.06

oters

-0.06

.words

-0.06

權

-0.06

_agg

-0.06

po

-0.06

-0.05

POSITIVE LOGITS

 adı

0.07

 mám

0.07

ฉ

0.07

 анти

0.07

Appearance

0.07

 уг

0.06

_srv

0.06

 membr

0.06

%'↵

0.06

.nr

0.06

Activations Density 0.025%

potential threats

The neuron is triggered by words signaling security or safety alerts—especially “threat,” “bomb,” and other suspicious or evacuation-related terms.

No Comments

No Known Activations