INDEX

Explanations

programming keywords and modifiers related to access controls, visibility, and annotations

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

public static final declarations

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

 préfé

-0.50

 asesino

-0.49

ंदीखरीदारी

-0.49

atguigu

-0.48

 StringTokenizer

-0.44

 محفوظة

-0.44

onOptions

-0.43

createStatement

-0.43

enumi

-0.43

BeginContext

-0.43

POSITIVE LOGITS

 butterknife

0.58

 IBOutlet

0.46

adaptiveStyles

0.45

 SafeMath

0.44

 ombre

0.42

ніципалі

0.42

 silks

0.41

WebElementEntity

0.41

kmale

0.41

 interfaces

0.40

Activations Density 0.001%

programming keywords and modifiers related to access controls, visibility, and annotations

public static final declarations

No Comments

No Known Activations