INDEX

Explanations

concepts and terms related to scientific methods and modeling

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

news, setup, relevant, coli

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

__":

-0.35

Discussion

-0.35

verständlich

-0.34

NVIC

-0.34

 prega

-0.34

anschluss

-0.33

ridge

-0.32

LayoutInflater

-0.32

__":

-0.32

ต่อไป

-0.31

POSITIVE LOGITS

 Италијани

0.81

 Савезне

0.79

 disambiguazione

0.74

 AssemblyTitle

0.66

 Italijanski

0.65

 Administrativna

0.63

ItemBackground

0.62

 المعيارى

0.60

 otomatig

0.59

 AssemblyCulture

0.59

Activations Density 0.025%

concepts and terms related to scientific methods and modeling

news, setup, relevant, coli

No Comments

No Known Activations