INDEX

Explanations

references to law enforcement and police-related actions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

by police

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ждую

-0.53

 cív

-0.53

 ciudadana

-0.47

 gouvernements

-0.45

'{@

-0.44

 cerâmica

-0.44

harapkan

-0.43

 kesan

-0.43

 Verhältnis

-0.43

 céramique

-0.43

POSITIVE LOGITS

 police

0.56

 Police

0.52

 detectives

0.50

 NYPD

0.47

FBI

0.47

Police

0.46

 RCMP

0.46

PropertyChanging

0.46

 suspect

0.45

police

0.43

Activations Density 0.015%

references to law enforcement and police-related actions

by police

No Known Activations