INDEX

Explanations

mathematical and programming symbols or notations

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

special characters and math symbols

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

Personendaten

-0.78

parsedMessage

-0.66

ftagPool

-0.63

DockStyle

-0.60

GOTREF

-0.59

WriteTagHelper

-0.58

UrlResolution

-0.58

GEBURTS

-0.56

 فريبيس

-0.50

 constitucional

-0.50

POSITIVE LOGITS

esModule

0.36

 shoes

0.36

TagMode

0.36

DAR

0.35

hí

0.34

 Giles

0.34

Zaj

0.34

 suitcases

0.34

SessionId

0.34

 Shoes

0.33

Activations Density 0.007%

mathematical and programming symbols or notations

special characters and math symbols

No Comments

No Known Activations