INDEX

Explanations

programming-related keywords and syntax components, particularly those associated with object-oriented programming and type checking

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

object type checking

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

 nonUne

-0.51

WillAppear

-0.47

/**

-0.46

blico

-0.46

SBATCH

-0.46

RenderAtEndOf

-0.45

idän

-0.45

prepareStatement

-0.45

 colectiva

-0.45

 BoxFit

-0.44

POSITIVE LOGITS

Personendaten

0.56

Хро

0.51

AnimationsModule

0.49

ثيق

0.46

 <>",

0.44

 tartalomajánló

0.41

GTCX

0.41

oa̍t

0.40

hoedd

0.40

 censiti

0.40

Activations Density 0.010%

programming-related keywords and syntax components, particularly those associated with object-oriented programming and type checking

object type checking

No Comments

No Known Activations