INDEX

Explanations

references to political office and elections

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

INGER

-0.15

uling

-0.14

BTN

-0.14

chner

-0.13

ÑģÑĤÐ¸Ð½

-0.13

ertura

-0.13

interp

-0.13

.charCodeAt

-0.13

áº¡ng

-0.13

umping

-0.13

POSITIVE LOGITS

 public

0.35

public

0.30

 government

0.28

 pubb

0.24

 elected

0.24

 pÃºblico

0.23

 positions

0.23

 PUBLIC

0.23

-public

0.23

 elective

0.22

Activations Density 0.058%

references to political office and elections

No Comments

No Known Activations