INDEX

Explanations

phrases that indicate size or limitations, particularly in relation to materials and physical properties

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

citations and technical terms

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

erunner

-0.50

PickerController

-0.49

celotti

-0.48

 gaji

-0.45

다는

-0.43

 Wappen

-0.42

 gebou

-0.41

 kosong

-0.41

 samman

-0.40

🏫

-0.39

POSITIVE LOGITS

 fragile

0.84

GEBURTSDATUM

0.84

QMetaType

0.81

 EconPapers

0.78

DockStyle

0.72

 kaarangay

0.71

ThroughAttribute

0.71

 heavy

0.68

 bulky

0.66

 delicate

0.66

Activations Density 0.169%

phrases that indicate size or limitations, particularly in relation to materials and physical properties

citations and technical terms

No Comments

No Known Activations