INDEX

Explanations

mathematical symbols and expressions

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

mathematics and code 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

存于互联网档案馆

-0.49

 FileWriter

-0.45

avigation

-0.44

 tanı

-0.43

 Aussagen

-0.43

 disambiguazione

-0.42

الإنجليزية

-0.41

phism

-0.41

cism

-0.40

ตัวเอง

-0.39

POSITIVE LOGITS

CppMethod

0.47

 ſch

0.47

 ſte

0.42

parsedMessage

0.41

 ſtre

0.40

 purpoſe

0.40

ſy

0.39

 ſta

0.39

Stu

0.39

Obrázky

0.38

Activations Density 0.028%

mathematical symbols and expressions

mathematics and code symbols

No Comments

No Known Activations