INDEX

Explanations

references to "voices" and their relationships to authority or groups

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

"voice" and related contexts

np_acts-logits-general · gemini-2.0-flash

voice communication and mail

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 oprot

-0.35

CloseOperation

-0.33

PickerController

-0.33

 rootReducer

-0.32

érables

-0.30

 MenuView

-0.30

 gắng

-0.29

 دا

-0.29

setVerticalGroup

-0.29

 stationnement

-0.29

POSITIVE LOGITS

قایناق‌لار

0.67

delwed

0.65

\{\\

0.63

ValueStyle

0.63

Voice

0.59

="@+

0.57

 Voice

0.56

mails

0.56

 hoarse

0.55

 nahilalakip

0.54

Activations Density 0.064%

references to "voices" and their relationships to authority or groups

"voice" and related contexts

voice communication and mail

No Comments

No Known Activations