INDEX

Explanations

named entities related to places, specifically theatres, languages, and indigenous groups

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

proper nouns or specific entities

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

ImageContext

-0.63

Diwedd

-0.60

ंदीखरीदारी

-0.60

IUrlHelper

-0.57

 الدولى

-0.56

principalColumn

-0.56

Brainz

-0.56

RefNanny

-0.54

hassee

-0.54

tich

-0.54

POSITIVE LOGITS

Demografie

0.36

Collegamenti

0.31

 bientôt

0.31

 nôtre

0.30

 StatefulWidget

0.30

 Höhe

0.30

 calme

0.29

страницу

0.29

updateOne

0.29

 franca

0.29

Activations Density 0.105%

named entities related to places, specifically theatres, languages, and indigenous groups

proper nouns or specific entities

No Comments

No Known Activations