INDEX

Explanations

mentions of specific individuals, particularly associated with controversial figures or topics

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

Osama Laden, Cambridge University Press

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

awtextra

-0.55

новништво

-0.51

calaure

-0.50

 tramonto

-0.48

 miniaturka

-0.47

lück

-0.47

approche

-0.47

 laajenta

-0.46

 ilustracji

-0.46

carteles

-0.45

POSITIVE LOGITS

 Laden

1.13

eryl

0.90

 onlyOwner

0.59

hradun

0.55

quiao

0.52

zarchiwizowane

0.51

xnn

0.50

ilene

0.50

})->

0.48

 Gleaner

0.48

Activations Density 0.003%

mentions of specific individuals, particularly associated with controversial figures or topics

Osama Laden, Cambridge University Press

No Comments

No Known Activations