INDEX

Explanations

place names and references to specific locations or jurisdictions

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

first patient

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

Hochspringen

-0.39

 capucha

-0.33

basicConfig

-0.32

 sostanze

-0.31

 huellas

-0.30

TypeInfo

-0.30

 OFFICE

-0.30

xtype

-0.29

hidrat

-0.29

 cotidiana

-0.28

POSITIVE LOGITS

verwijspagina

0.72

 BoxFit

0.68

 AssemblyVersion

0.64

StoryboardSegue

0.60

ChildScrollView

0.59

 فريبيس

0.56

RenderAtEndOf

0.54

setVerticalGroup

0.53

 nahilalakip

0.53

GHIJKLM

0.52

Activations Density 0.018%

place names and references to specific locations or jurisdictions

first patient

No Comments

No Known Activations