INDEX

Explanations

code

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_19/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.19.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Jed

-0.07

-test

-0.06

 praising

-0.06

ecake

-0.06

 rectangles

-0.06

 astronauts

-0.06

_STOP

-0.06

 Tavern

-0.06

 tempList

-0.06

POSITIVE LOGITS

 comando

0.06

 خر

0.06

 сна

0.06

 disrupted

0.06

underscore

0.06

icrous

0.06

months

0.06

Wall

0.06

-establish

0.06

.drive

0.06

Activations Density 0.063%

code

No Comments

No Known Activations