INDEX

Explanations

references to controlling or regulating various aspects of processes and energies

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

"Control" followed by certain words

np_acts-logits-general · gemini-2.0-flash

control

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

########.

-0.50

IContainer

-0.41

sca

-0.40

onActivityResult

-0.40

migrationBuilder

-0.40

umer

-0.39

 lenker

-0.39

spli

-0.39

Compat

-0.38

ur

-0.38

POSITIVE LOGITS

 control

0.84

 contrôler

0.79

 controlled

0.77

controlled

0.76

 CONTROL

0.76

 controlling

0.75

控制

0.73

Controlled

0.73

 controla

0.72

 controllable

0.71

Activations Density 0.523%

references to controlling or regulating various aspects of processes and energies

"Control" followed by certain words

control

No Comments

No Known Activations