INDEX

Explanations

control

np_max-act · gemini-2.0-flash

This neuron detects mentions of having “control over” aspects like security, privacy, or configuration (i.e. advantage statements emphasizing control).

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

 motel

-0.07

 FUNCT

-0.06

bitset

-0.06

สล

-0.06

 новых

-0.06

AZ

-0.06

Longitude

-0.06

 раздел

-0.06

Sv

-0.06

.loadData

-0.06

POSITIVE LOGITS

mort

0.07

 Studio

0.07

全面

0.07

 Adding

0.07

 screw

0.07

 TASK

0.07

(INT

0.06

/control

0.06

 Control

0.06

 "")↵

0.06

Activations Density 0.015%

control

This neuron detects mentions of having “control over” aspects like security, privacy, or configuration (i.e. advantage statements emphasizing control).

No Comments

No Known Activations