INDEX

Explanations

references to killing and murder within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 Farah

-0.76

Applicant

-0.75

 vache

-0.72

 متحده

-0.71

 Manly

-0.70

 Weeks

-0.69

 grotte

-0.69

cerne

-0.68

ngl

-0.67

cheme

-0.66

POSITIVE LOGITS

 kill

1.45

 kills

1.41

 KILL

1.38

 Kill

1.36

kill

1.29

 killing

1.28

 killed

1.26

 Kills

1.21

kills

1.20

 killings

1.20

Activations Density 0.065%

references to killing and murder within the text

No Comments

No Known Activations