INDEX

Explanations

instances of moral judgment or wrongdoing

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

Not in Any Lists

No Comments

Negative Logits

apa

-0.16

.qual

-0.16

Ø§ÙĨÙĪ

-0.16

antar

-0.16

uke

-0.16

lsa

-0.15

ipy

-0.15

anki

-0.15

rise

-0.15

arters

-0.14

POSITIVE LOGITS

fully

0.34

headed

0.34

/right

0.29

-headed

0.29

er

0.26

ed

0.26

wrong

0.25

 wrong

0.25

do

0.24

Activations Density 0.055%

instances of moral judgment or wrongdoing

No Comments

No Known Activations