INDEX

Explanations

proper nouns and geographical locations

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

Borneo geographies and Malaysian locations

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

 ſtand

-0.48

 gynhyrchwyd

-0.48

 Lovato

-0.47

ourge

-0.47

Datuak

-0.47

mercial

-0.47

 beſ

-0.46

힣

-0.46

-------

-0.46

🧀

-0.46

POSITIVE LOGITS

 Sarawak

0.72

 Borneo

0.67

imantan

0.64

 Kalimantan

0.57

wahati

0.42

 CreateTagHelper

0.40

 Brunei

0.36

 Kuch

0.36

bkz

0.35

RectangleBorder

0.34

Activations Density 0.012%

proper nouns and geographical locations

Borneo geographies and Malaysian locations

No Comments

No Known Activations