INDEX

Explanations

references to Sweden and related terms

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

Sweden or Swedish

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

SourceChecksum

-0.49

IntoConstraints

-0.46

AddHtmlAttribute

-0.45

ьаж

-0.45

 متعلقه

-0.45

beutel

-0.44

 okru

-0.44

itizer

-0.44

itza

-0.44

 dima

-0.44

POSITIVE LOGITS

 Sweden

1.98

Sweden

1.83

 Swedish

1.79

Swedish

1.65

 Swedes

1.58

 sweden

1.45

 Suède

1.45

 Suecia

1.44

 schwed

1.37

 Schweden

1.31

Activations Density 0.005%

references to Sweden and related terms

Sweden or Swedish

No Comments

No Known Activations