INDEX

Explanations

computers

np_max-act · gemini-2.0-flash

This neuron fires on hardware‐specification tokens—particularly GPU/CPU model names and numeric values (e.g. “RTX,” “2060,” decimal numbers) in the text.

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

-0.06

-hide

-0.06

mul

-0.06

egative

-0.06

ыш

-0.06

 discomfort

-0.06

occer

-0.06

 jelly

-0.06

ZX

-0.06

	conf

-0.06

POSITIVE LOGITS

 renov

0.07

 blessings

0.07

![

0.06

 dateTime

0.06

aser

0.06

.openConnection

0.06

/images

0.06

=h

0.06

++){↵↵

0.06

清

0.06

Activations Density 0.022%

computers

This neuron fires on hardware‐specification tokens—particularly GPU/CPU model names and numeric values (e.g. “RTX,” “2060,” decimal numbers) in the text.

No Comments

No Known Activations