INDEX

Explanations

references to cyber threats and attacks, particularly involving hackers and malicious organizations

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

Threats, attacks, or invasions

np_acts-logits-general · gemini-2.0-flash

Threatening invaders

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Sprintf

-0.42

 moest

-0.40

 mengatasi

-0.39

 így

-0.39

 blushed

-0.39

 queda

-0.39

mace

-0.39

çois

-0.38

 contribu

-0.38

 helps

-0.38

POSITIVE LOGITS

 targeting

0.99

StructEnd

0.91

 threatening

0.90

 attacking

0.89

 attacks

0.87

 threaten

0.86

 Attacks

0.86

EDEFAULT

0.84

targeting

0.83

 threatens

0.83

Activations Density 0.609%

references to cyber threats and attacks, particularly involving hackers and malicious organizations

Threats, attacks, or invasions

Threatening invaders

No Comments

No Known Activations