INDEX

Explanations

numerical identifiers or references in patent applications

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

patent numbers

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

GEBURTSDATUM

-0.68

]")]

-0.62

 désolés

-0.61

 تضيفلها

-0.60

تقاوى

-0.59

帖最后由

-0.57

 дописавши

-0.56

ParallelGroup

-0.56

цездатний

-0.56

LayoutStyle

-0.54

POSITIVE LOGITS

 enfans

0.45

 bénéfice

0.41

 aperçu

0.40

 argint

0.38

 bénéfices

0.38

jalá

0.38

tamment

0.38

 Berührung

0.37

paravant

0.37

 discorso

0.37

Activations Density 0.111%

numerical identifiers or references in patent applications

patent numbers

No Comments

No Known Activations