INDEX

Explanations

structured data and format representations commonly used in programming or API documentation

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

numbers and symbols

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

AntiForgeryToken

-0.65

GEBURTS

-0.63

RefNanny

-0.53

 صوتيه

-0.52

OGND

-0.52

 AssemblyCulture

-0.51

 disambiguazione

-0.50

脚注の使い方

-0.50

makeConstraints

-0.49

 ProtoMessage

-0.49

POSITIVE LOGITS

 pinulongan

0.47

Бахар

0.42

Bibliograf

0.36

səhifə

0.35

SCHEMA

0.35

ambar

0.35

wingen

0.34

 sorpresas

0.34

 Grüsse

0.34

zkiem

0.34

Activations Density 0.129%

structured data and format representations commonly used in programming or API documentation

numbers and symbols

No Comments

No Known Activations