INDEX

Explanations

references to positions, roles, or work settings

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

job position

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

icoot

-0.40

orden

-0.39

ionario

-0.39

 gyhoeddwyd

-0.36

xsi

-0.36

 certain

-0.36

SequentialGroup

-0.35

 vêtement

-0.35

 Footprint

-0.35

 intuiti

-0.35

POSITIVE LOGITS

nextProps

0.51

ⓧ

0.50

 wireType

0.48

MeasureSpec

0.46

CloseOperation

0.46

SourceChecksum

0.46

resave

0.45

featureID

0.45

styleable

0.44

 nonUne

0.43

Activations Density 0.187%

references to positions, roles, or work settings

job position

No Comments

No Known Activations