INDEX

Explanations

references to figures and tables in a document

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

reference markers

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

Personensuche

-0.74

KURZBESCHREIBUNG

-0.73

 queſta

-0.71

 pinulongan

-0.67

SourceChecksum

-0.63

piram

-0.60

Географиясе

-0.60

 autorytatywna

-0.60

 Autorizaciones

-0.60

 AssemblyCulture

-0.59

POSITIVE LOGITS

 likewise

0.54

 ebenfalls

0.53

 similarly

0.47

 equally

0.47

 szint

0.47

 altrett

0.43

Similarly

0.43

 selben

0.43

Likewise

0.43

同样

0.42

Activations Density 0.204%

references to figures and tables in a document

reference markers

No Comments

No Known Activations