INDEX

Explanations

references to cleanliness and maintenance in various contexts

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

actions associated with maintenance and cleaning

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

cleaning and maintenance

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 cleaning

-1.36

clean

-1.34

 clean

-1.31

 Cleaning

-1.24

 Clean

-1.23

Clean

-1.23

cleaning

-1.20

Cleaning

-1.20

 CLEAN

-1.20

 cleaned

-1.16

POSITIVE LOGITS

sphase

0.58

 McCartney

0.57

 kasarigan

0.56

omie

0.53

cplusplus

0.53

IsMutable

0.52

 nori

0.51

OrderID

0.51

 thơm

0.50

osoba

0.50

Activations Density 0.017%

references to cleanliness and maintenance in various contexts

actions associated with maintenance and cleaning

cleaning and maintenance

No Comments

No Known Activations