INDEX

Explanations

commentary/advice

np_max-act · gemini-2.0-flash

The neuron responds to normative or advisory language—tokens in the assistant’s moral-guidance or encouragement statements.

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

 hide

-0.08

 Popup

-0.06

year

-0.06

mee

-0.06

,T

-0.06

ุ์

-0.06

hen

-0.06

clud

-0.06

 Hide

-0.06

 Crew

-0.06

POSITIVE LOGITS

.Clamp

0.07

lya

0.06

imet

0.06

.atan

0.06

 tess

0.06

_INT

0.06

Esp

0.06

ском

0.06

 Plaint

0.06

longleftrightarrow

0.06

Activations Density 0.090%

commentary/advice

The neuron responds to normative or advisory language—tokens in the assistant’s moral-guidance or encouragement statements.

No Comments

No Known Activations