INDEX

Explanations

references to versioning and licensing terms

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

file headers and code 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

 surla

-0.56

@[+][

-0.54

ɵɵelementStart

-0.50

 بيها

-0.48

ɵɵ

-0.48

typeorm

-0.48

medriver

-0.48

$_(

-0.48

səhifə

-0.47

 PeEnEo

-0.47

POSITIVE LOGITS

 Familienname

0.46

paravant

0.39

EndTag

0.39

 esprits

0.38

eseorang

0.38

 avoient

0.38

 arşivlendi

0.38

 AssemblyCompany

0.37

исленность

0.37

Authority

0.37

Activations Density 0.001%

references to versioning and licensing terms

file headers and code markers

No Comments

No Known Activations