INDEX

Explanations

Java annotations and repository-related concepts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

public class/interface 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

Negative Logits

OGND

-0.74

 betweenstory

-0.68

VersionUID

-0.68

aarrggbb

-0.65

ConstraintMaker

-0.64

GraphicsUnit

-0.63

IntoConstraints

-0.63

TypedDataSet

-0.61

majánló

-0.60

featureID

-0.60

POSITIVE LOGITS

 autorytatywna

0.42

 IBOutlet

0.33

ResumeLayout

0.26

fromUtf

0.25

Viitteet

0.25

abstractmethod

0.25

staticmethod

0.25

orm

0.24

󠁣

0.24

Attached

0.24

Activations Density 0.003%

Java annotations and repository-related concepts

public class/interface declarations

No Known Activations