INDEX

Explanations

programming terms related to creating and manipulating objects or classes

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

Constructor or method calls with 'from'

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

 betweenstory

-0.62

 AssemblyProduct

-0.60

出版年

-0.54

WaitGroup

-0.47

tagHelper

-0.44

awtextra

-0.43

Personendaten

-0.42

FXML

-0.41

getItemId

-0.41

ウンサー

-0.41

POSITIVE LOGITS

 colorWith

0.72

Portale

0.60

 numberWith

0.56

 findBy

0.55

findBy

0.52

createFrom

0.50

 initWith

0.50

yntaxException

0.49

 GoogleFonts

0.49

getBy

0.48

Activations Density 0.013%

programming terms related to creating and manipulating objects or classes

Constructor or method calls with 'from'

No Comments

No Known Activations