INDEX

Explanations

brand names and retail locations

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

store brands

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

htra

-0.34

見ても

-0.33

utul

-0.32

หน้าน

-0.32

FunctionFlags

-0.32

 traits

-0.31

تحميل

-0.31

митрий

-0.31

 realities

-0.31

 subtle

-0.31

POSITIVE LOGITS

 BoxFit

0.57

ंदीखरीदारी

0.56

تقاوى

0.54

bootstrapcdn

0.53

initializeApp

0.50

 WebDriverWait

0.50

 utafitiHapana

0.49

 itſelf

0.48

 Houſe

0.48

 оригіналу

0.48

Activations Density 0.008%

brand names and retail locations

store brands

No Comments

No Known Activations