INDEX

Explanations

email addresses or links associated with logins

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

alach

-0.18

ange

-0.18

.fix

-0.15

haled

-0.15

eday

-0.14

 Mahmoud

-0.14

aval

-0.14

irim

-0.14

zsche

-0.14

_salt

-0.14

POSITIVE LOGITS

 protected

0.39

protected

0.31

 Protected

0.26

Protected

0.24

 protection

0.23

 protect

0.23

 protecting

0.22

 protective

0.21

ä¿ĿæĬ¤

0.21

	protected

0.19

Activations Density 0.004%

email addresses or links associated with logins

No Comments

No Known Activations