INDEX

Explanations

references to censorship and banned works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

-0.15

loff

-0.15

abyrin

-0.15

pev

-0.15

íħ

-0.15

@student

-0.14

igne

-0.14

 Baghd

-0.14

uvw

-0.14

 abandonment

-0.14

POSITIVE LOGITS

 censorship

0.44

 censor

0.43

0.34

ensor

0.33

ensored

0.31

cen

0.26

ensors

0.25

 banning

0.23

 bans

0.23

ban

0.23

Activations Density 0.075%

references to censorship and banned works

No Comments

No Known Activations