INDEX

Explanations

terms related to violence and criminal activity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 actually

-0.51

 nämlich

-0.48

 bandoulière

-0.47

 mijne

-0.47

unschweig

-0.47

 dingen

-0.47

ملكة

-0.46

 Vorteil

-0.46

 Española

-0.46

 rangka

-0.46

POSITIVE LOGITS

########.

0.76

EndContext

0.72

tonode

0.69

withIdentifier

0.67

awtextra

0.64

__':

0.62

dicionado

0.59

MessageTagHelper

0.58

 snippetHide

0.57

 Numerade

0.57

Activations Density 0.900%

terms related to violence and criminal activity

No Comments

No Known Activations