INDEX

Explanations

phrases that express methods or approaches

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

to wordsof words

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

tagHelperRunner

-0.52

sidemargin

-0.51

leyball

-0.50

 étoient

-0.49

ſelf

-0.47

LookAnd

-0.47

ómago

-0.47

fjspx

-0.46

 ainfi

-0.46

música

-0.46

POSITIVE LOGITS

UseVisualStyle

0.40

扛

0.34

.*")]

0.32

ÚBL

0.31

DeleteBehavior

0.30

internalType

0.30

лго

0.30

 Noor

0.30

ecore

0.30

WIB

0.29

Activations Density 0.045%

phrases that express methods or approaches

to wordsof words

No Comments

No Known Activations