INDEX

Explanations

terms related to security and risk management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

usercontent

-0.18

kte

-0.17

nez

-0.16

_SRV

-0.16

á»ģn

-0.15

zyst

-0.15

adan

-0.15

DebugEnabled

-0.15

íĸī

-0.14

ossip

-0.14

POSITIVE LOGITS

 security

0.22

-security

0.17

Security

0.17

 SECURITY

0.17

security

0.17

 safety

0.16

 guard

0.15

 guards

0.15

 sÃ©curitÃ©

0.15

'en

0.15

Activations Density 0.008%

terms related to security and risk management

No Comments

No Known Activations