INDEX

Explanations

patterns related to image naming and Android resource files

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

image and string resources

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

 CreateTagHelper

-0.48

RectangleBorder

-0.45

HtmlAttribute

-0.43

########.

-0.42

expandindo

-0.37

 समीक्षाओं

-0.35

providedIn

-0.35

webElementXpaths

-0.34

 independ

-0.34

 administrativos

-0.33

POSITIVE LOGITS

 imageNamed

0.64

 astore

0.54

 Majefty

0.50

adaptiveStyles

0.48

 IBRARY

0.48

 actionMode

0.48

 للاسماء

0.48

 cartoons

0.47

 resourceCulture

0.47

 Numerade

0.47

Activations Density 0.005%

patterns related to image naming and Android resource files

image and string resources

No Comments

No Known Activations