INDEX

Explanations

space

np_max-act · gemini-2.0-flash

This neuron responds to whitespace‐related constructs, such as space/newline characters and character‐class patterns matching whitespace.

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

	UPROPERTY

-0.06

_REFERER

-0.06

Sou

-0.06

Isl

-0.06

.batch

-0.06

 Sevent

-0.06

_argument

-0.06

>{@

-0.06

_should

-0.06

Specifies

-0.06

POSITIVE LOGITS

.xhtml

0.06

 століття

0.06

DEFINE

0.06

 Holder

0.06

 швидко

0.06

ometer

0.06

ivityManager

0.06

SSR

0.06

Enemies

0.06

 καθ

0.06

Activations Density 0.006%

space

This neuron responds to whitespace‐related constructs, such as space/newline characters and character‐class patterns matching whitespace.

No Comments

No Known Activations