INDEX

Explanations

terms and phrases related to security and insecurity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 security

-0.24

 Security

-0.23

_security

-0.23

security

-0.21

Security

-0.20

 securely

-0.19

 secure

-0.19

 SECURITY

-0.18

 sÃ©curitÃ©

-0.17

 sÃ©cur

-0.16

POSITIVE LOGITS

 measures

0.25

-conscious

0.25

 measure

0.23

 minded

0.23

 conscious

0.23

ContextHolder

0.22

-minded

0.22

 breach

0.22

 Measures

0.21

Bre

0.20

Activations Density 0.021%

terms and phrases related to security and insecurity

No Comments

No Known Activations