INDEX

Explanations

.

np_max-act · gemini-2.0-flash

The neuron responds most strongly to words describing violent or threatening actions (e.g. “execution,” “set afire,” “firing squad,” “match”). It flags terms that evoke physical harm or acts of aggression.

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

URED

-0.07

evt

-0.07

ATURE

-0.06

 baseline

-0.06

 fragile

-0.06

 Rita

-0.06

�

-0.06

SL

-0.06

方向

-0.06

 twice

-0.06

POSITIVE LOGITS

={(

0.07

_tcp

0.06

atables

0.06

',{'

0.06

-scenes

0.06

 jsonObj

0.06

 수강

0.06

欲

0.06

());
↵

0.06

.training

0.06

Activations Density 0.105%

.

The neuron responds most strongly to words describing violent or threatening actions (e.g. “execution,” “set afire,” “firing squad,” “match”). It flags terms that evoke physical harm or acts of aggression.

No Comments

No Known Activations