INDEX

Explanations

references to automatic processes or features in software settings

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

Tokens starting with "auto"

np_acts-logits-general · gemini-2.0-flash

auto with modifiers

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

addMessage

-0.48

 Geden

-0.46

 promocional

-0.39

 Berlín

-0.38

 Regionen

-0.37

 menuItem

-0.36

 colgante

-0.36

 Económica

-0.36

 Bidang

-0.36

 ideia

-0.36

POSITIVE LOGITS

 AUTO

1.10

 Auto

1.08

Auto

1.05

 auto

1.02

Autowired

1.00

setAuto

0.99

 Autom

0.96

Automatic

0.94

Autos

0.93

autom

0.92

Activations Density 0.097%

references to automatic processes or features in software settings

Tokens starting with "auto"

auto with modifiers

No Comments

No Known Activations