INDEX

Explanations

references to vampires in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post

Features

65,536

Data Type

float32

Hook Name

blocks.25.hook_resid_post

Hook Layer

Architecture

gated

Context Size

1,024

Dataset

Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Dwarf

-0.15

 Corpor

-0.15

ventas

-0.15

_px

-0.15

.px

-0.14

celand

-0.14

erken

-0.13

 autistic

-0.13

ruk

-0.13

áº·t

-0.13

POSITIVE LOGITS

 vampire

0.50

 vamp

0.46

 Vampire

0.46

 vampires

0.45

acula

0.34

ampire

0.33

wer

0.33

Wer

0.28

 blood

0.28

 Blood

0.27

Activations Density 0.020%

references to vampires in various contexts

No Comments

No Known Activations