INDEX

Explanations

variations of the word "profanity"

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

edes

-0.17

arch

-0.16

ENS

-0.16

dbl

-0.16

 cust

-0.15

ì£¼ëĬĶ

-0.15

ero

-0.15

oggles

-0.15

unched

-0.14

 Dynamo

-0.14

POSITIVE LOGITS

anity

0.28

 prof

0.27

ane

0.20

essed

0.19

essional

0.19

ANE

0.18

essions

0.18

aned

0.18

 PROF

0.17

umo

0.17

Activations Density 0.007%

variations of the word "profanity"

No Comments

No Known Activations