INDEX

Explanations

references to specific brands and manufacturers

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

organizations and legal citations

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



-0.61

vician

-0.60

AISSEE

-0.54

argout

-0.50

smithy

-0.50

featureID

-0.49

?></

-0.47

dAtA

-0.45

NameInMap

-0.45

InstrumentedTest

-0.45

POSITIVE LOGITS

 organizations

0.40

Emerson

0.37

 Organisationen

0.35

 Organizations

0.35

 organisations

0.34

 Zusammenarbeit

0.33

 Том

0.33

recognized

0.32

 Kooperation

0.32

 velkommen

0.32

Activations Density 0.318%

references to specific brands and manufacturers

organizations and legal citations

No Comments

No Known Activations