INDEX

Explanations

elements related to image processing and manipulation

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

image related terms

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

 AssemblyCompany

-0.34

TextBoxColumn

-0.32

架

-0.28

때

-0.28

 please

-0.28

ValueGenerated

-0.28

 baş

-0.27

canza

-0.27

ibatis

-0.27

 faltar

-0.27

POSITIVE LOGITS

imread

0.72

 tartalomajánló

0.68

ImageContext

0.64

 eſſ

0.63

 ddelwed

0.61

 BoxFit

0.60

 BufferedImage

0.60

 imgs

0.60

BufferedImage

0.59

simage

0.59

Activations Density 0.064%

elements related to image processing and manipulation

image related terms

No Comments

No Known Activations