INDEX

Explanations

references to infamous criminal incidents and locations

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

visor

-0.15

anela

-0.15

cbc

-0.14

å¨ģ

-0.14

 Annunci

-0.14

ivid

-0.14

duk

-0.13

.animation

-0.13

indre

-0.13

azeera

-0.13

POSITIVE LOGITS

 Manson

0.40

 Tate

0.32

Hel

0.30

 Charles

0.27

 cult

0.27

 Sharon

0.26

hel

0.26

Hel

0.25

 Cult

0.24

Charles

0.23

Activations Density 0.002%

references to infamous criminal incidents and locations

No Known Activations