INDEX

Explanations

references to scientific or mathematical formats and symbols

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

mathematical expressions

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

 Erde

-0.33

 Herzen

-0.32

um

-0.31

 Heiligen

-0.30

 cuerda

-0.30

movieId

-0.30

 Spitzen

-0.29

pas

-0.28

 chickens

-0.28

 grand

-0.28

POSITIVE LOGITS

GEBURTSDATUM

0.79

 surla

0.66

RegistryLite

0.62

Hauptartikel

0.61

DockStyle

0.60

Diwedd

0.60

ंदीखरीदारी

0.59

transQ

0.59

 kasarigan

0.59

 Infórmanos

0.58

Activations Density 0.014%

references to scientific or mathematical formats and symbols

mathematical expressions

No Comments

No Known Activations