INDEX

Explanations

usernames or email addresses

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

@ domain names

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

setVerticalGroup

-0.38

tonsoft

-0.36

 TestBed

-0.36

يميديا

-0.35

saraba

-0.34

bpy

-0.34

="@+

-0.33

ⓧ

-0.32

+:+

-0.31

 Familienname

-0.31

POSITIVE LOGITS

OGND

0.69

 miniaturka

0.58

المناصب

0.57

ValueStyle

0.57

 camiset

0.53

 мәкал

0.52

хьтан

0.52

Bibliograf

0.52

 собі

0.50

 spesa

0.49

Activations Density 0.000%

usernames or email addresses

@ domain names

No Comments

No Known Activations