INDEX

Explanations

formal

np_max-act · gemini-2.0-flash

This neuron detects instructions or prompts requesting a formal writing style (e.g. the word “formal” or related style‐specification tokens).

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

 Louisiana

-0.07

 uploading

-0.07

ramento

-0.07

Skills

-0.06

互

-0.06

zelf

-0.06

 fertile

-0.06

 Indonesia

-0.06

oses

-0.06

 nostra

-0.06

POSITIVE LOGITS

Dal

0.07

.Raise

0.06

!“

0.06

 DataContext

0.06

 नक

0.06

 celebr

0.06

NAMESPACE

0.06

ฐาน

0.06

 dành

0.06

Activations Density 0.029%

formal

This neuron detects instructions or prompts requesting a formal writing style (e.g. the word “formal” or related style‐specification tokens).

No Comments

No Known Activations

formal

This neuron detects instructions or prompts requesting a formal writing style (e.g. the word “formal” or related style‐specification tokens).

No Comments

No Known Activations