INDEX

Explanations

references to passports

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

passport application

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 elétrica

-0.48

 negócio

-0.46

effects

-0.45

usual

-0.44

 mesmas

-0.44

wę

-0.42

evil

-0.41

 inimigo

-0.40

 Howe

-0.39

Big

-0.39

POSITIVE LOGITS

 passport

1.24

 Passport

1.22

Passport

1.13

 passports

1.05

passport

1.00

 паспорт

0.67

 faſt

0.62

 peregr

0.61

 nahilalakip

0.61

StoryboardSegue

0.61

Activations Density 0.005%

references to passports

passport application

No Comments

No Known Activations