INDEX

Explanations

formatting elements commonly used in mathematical expressions or citations

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

of Theorem

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

LookAnd

-0.90

 betweenstory

-0.84

 queſta

-0.73

queryInterface

-0.72

 nonUne

-0.68

WriteTagHelper

-0.68

 gynhyrchwyd

-0.67

tangentMode

-0.67

oredCriteria

-0.66

NamedQueries

-0.65

POSITIVE LOGITS

 brancas

0.36

0.34

 <<<<<<<<<<<<<<

0.33

 this

0.31

 vermelhas

0.30

“

0.30

nameLabel

0.30

“

0.30

 recentes

0.30

《

0.28

Activations Density 0.000%

formatting elements commonly used in mathematical expressions or citations

of Theorem

No Comments

No Known Activations