INDEX

Explanations

relationships and familial connections

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

repeated enumerations

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

 relâche

-0.41

 absolutely

-0.40

leby

-0.40

 absolu

-0.39

 absoluto

-0.38

 assolu

-0.36

 Mascot

-0.36

Parent

-0.35

ajur

-0.35

boy

-0.35

POSITIVE LOGITS

SharedCtor

0.60

PerformLayout

0.52

 Италијани

0.52

TemporalType

0.51

 CURIAM

0.50

 Administrativna

0.50

bcryptjs

0.49

EndInit

0.48

verwijspagina

0.48

OGND

0.47

Activations Density 0.000%

relationships and familial connections

repeated enumerations

No Comments

No Known Activations