INDEX

Explanations

references to royal family events and individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

allel

-0.16

ampaign

-0.15

etroit

-0.15

íĺ¸

-0.14

aub

-0.14

iaux

-0.14

ohana

-0.14

mrb

-0.13

 NSMutable

-0.13

esome

-0.13

POSITIVE LOGITS

 royal

0.36

 Royal

0.32

 Prince

0.32

roy

0.32

roy

0.31

 Royals

0.30

Royal

0.29

 Princess

0.29

 prince

0.28

Prince

0.28

Activations Density 0.141%

references to royal family events and individuals

No Comments

No Known Activations