INDEX

Explanations

the word "floor" along with associated words

oai_token-act-pair · gemini-2.0-flash

floor

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

homepage

-0.56

 gyhoeddwyd

-0.52

arqu

-0.50

krivelse

-0.49

igshid

-0.49

abestanden

-0.48

 Toch

-0.47

 Acceptable

-0.47

ยว

-0.47

òria

-0.47

POSITIVE LOGITS

 BoxDecoration

0.73

EndProject

0.65

DoubleQuotes

0.62

XmlAccessorType

0.56

WebServlet

0.52

WithMany

0.51

 inoxydable

0.51

mergeFrom

0.49

če

0.47

 RIPRODUZIONE

0.47

Activations Density 1.373%

the word "floor" along with associated words

floor

No Comments

No Known Activations