INDEX

Explanations

security-related terms and phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ju

-0.34

 createSprite

-0.34

 uLocal

-0.34

hit

-0.33

Ju

-0.31

 revived

-0.31

 Bons

-0.30

hit

-0.30

he

-0.30

Sub

-0.30

POSITIVE LOGITS

 security

0.97

security

0.82

Security

0.79

 SECURITY

0.77

 privacy

0.76

 Security

0.75

セキュリティ

0.71

 sécurité

0.70

 seguridad

0.69

privacy

0.69

Activations Density 0.507%

security-related terms and phrases

No Comments

No Known Activations