INDEX

Explanations

the number of times an action or event is repeated in the text

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

copies and prior occurrences

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

 CreateTagHelper

-0.58

 مرئيه

-0.48

Filmographie

-0.48

isGrid

-0.46

DrawerToggle

-0.45

MethodManager

-0.45

richt

-0.44

rungsseite

-0.44

 ddelweddau

-0.44

libur

-0.44

POSITIVE LOGITS

ंदीखरीदारी

0.44

⋙

0.34

 rép

0.33

 копия

0.32

 نسخ

0.32

DECL

0.32

 بک

0.30

 copies

0.30

 esternos

0.30

 versions

0.30

Activations Density 0.114%

the number of times an action or event is repeated in the text

copies and prior occurrences

No Comments

No Known Activations