INDEX

Explanations

structured mathematical terminology and relationships

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

terms and contributions

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

 سكانية

-0.41

 Administrativna

-0.40

лтемелер

-0.38

 الرخصة

-0.35

 विश्वसनीयता

-0.33

 sceptre

-0.33

exitRule

-0.33

 perfons

-0.32

 studio

-0.31

POSITIVE LOGITS

 contribution

0.56

CPtr

0.56

Portion

0.51

 contributo

0.51

RTDA

0.49

 aanv

0.49

 CONTRIBUTION

0.48

 verantwoorde

0.46

orithmic

0.46

 contributory

0.46

Activations Density 0.345%

structured mathematical terminology and relationships

terms and contributions

No Comments

No Known Activations