INDEX

Explanations

events related to murder and violent acts

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

joba

-0.45

nextProps

-0.43

 quienes

-0.42

 berücksichtigen

-0.41

 Ritz

-0.41

Butt

-0.41

AppContext

-0.41

galus

-0.41

XCTAssert

-0.40

resizingMask

-0.40

POSITIVE LOGITS

 attack

0.76

 violence

0.71

 crime

0.66

 CRIME

0.63

 violen

0.63

attack

0.63

 Violence

0.62

 murder

0.62

 attaque

0.61

 Attack

0.61

Activations Density 0.754%

events related to murder and violent acts

No Comments

No Known Activations