INDEX

Explanations

phrases indicating warnings or threats

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

followed by prepositions

np_acts-logits-general · gemini-2.0-flash

formal statements and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

DockStyle

-0.49

 esternos

-0.46

 yapılan

-0.43

 scattata

-0.41

ISupport

-0.41

 efectu

-0.40

IsContent

-0.39

 gemaakt

-0.39

Effec

-0.38

 isInitialized

-0.38

POSITIVE LOGITS

kepada

0.49

 vnto

0.49

addCriterion

0.48

樣的

0.48

 throughout

0.48

 indisponible

0.47

 through

0.46

principalColumn

0.45

 autorytatywna

0.45

 AssemblyCompany

0.44

Activations Density 0.811%

phrases indicating warnings or threats

followed by prepositions

formal statements and outcomes

No Comments

No Known Activations