INDEX

Explanations

programming-related keywords and declarations

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

return statements

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

withOpacity

-0.64

 Meksiku

-0.48

发表于

-0.47

دانشنامهٔ

-0.46

PreferredItem

-0.45

awtextra

-0.44

UIControlState

-0.43

IonicModule

-0.43

 MainAxisSize

-0.43

 SwiftUI

-0.42

POSITIVE LOGITS

httphttps

0.47

RenderAtEndOf

0.44

DriverManager

0.38

QMetaType

0.38

 useContext

0.36

omock

0.36

SequentialGroup

0.34

ViewFeatures

0.34

contentLoaded

0.33

 تضيفلها

0.33

Activations Density 0.002%

programming-related keywords and declarations

return statements

No Comments

No Known Activations