INDEX

Explanations

references to movies and cinematic elements

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

"which" followed by a preposition or verb

np_acts-logits-general · gemini-2.0-flash

clauses describing relationships

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-16k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 مرئيه

-0.77

Personensuche

-0.70

󠁢

-0.62

+#+#

-0.61

 wireType

-0.61

 createState

-0.60

IGraphics

-0.58

存于互联网档案馆

-0.58

 beginnetje

-0.57

BagLayout

-0.57

POSITIVE LOGITS

 surla

0.67

culada

0.64

曖昧さ回避

0.55

EditorBrowsable

0.54

 affair

0.51

ési

0.51

UnusedPrivate

0.51

 thing

0.50

Жит

0.49

bria

0.49

Activations Density 1.588%

references to movies and cinematic elements

"which" followed by a preposition or verb

clauses describing relationships

No Comments

No Known Activations