INDEX

Explanations

expressions of fear and risk related to personal safety

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

ehler

-0.16

ĮĴ

-0.15

 culprit

-0.15

Ø®Ø´

-0.14

Įĵ

-0.14

áº¬

-0.14

 guilty

-0.14

insi

-0.13

affen

-0.13

 Regards

-0.13

POSITIVE LOGITS

 being

0.40

being

0.34

 Being

0.29

Being

0.29

 becoming

0.29

 scrutiny

0.28

è¢«

0.27

 detection

0.25

 losing

0.24

 attack

0.23

Activations Density 0.227%

expressions of fear and risk related to personal safety

No Comments

No Known Activations