INDEX

Explanations

phrases related to law enforcement or criminal activity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.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



-0.64

Personendaten

-0.63

 poveznice

-0.63

 Италијани

-0.58

ništ

-0.55

+#+#

-0.54

MockBean

-0.53

 дописавши

-0.52

tershire

-0.51

RegistryLite

-0.51

POSITIVE LOGITS

 mask

2.86

 masks

2.64

 Mask

2.44

mask

2.33

Mask

2.30

 Masks

2.20

 MASK

2.14

Masks

1.98

masks

1.98

 masked

1.98

Activations Density 0.382%

phrases related to law enforcement or criminal activity

No Comments

No Known Activations