INDEX

Explanations

personal pronouns and sentence-ending Japanese grammatical particles, as well as Java-related terms

oai_token-act-pair · gemini-2.0-flash

させ

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

principalTable

-0.49

inf

-0.46

fromCharCode

-0.43

されていた

-0.40

 Roskov

-0.39

丁目

-0.39

 crew

-0.38

ustris

-0.38

 Order

-0.38

Was

-0.38

POSITIVE LOGITS

SequentialGroup

0.91

ChildScrollView

0.75

 متعلقه

0.72

 bezeichneter

0.65

 שוליים

0.64

 OnInit

0.62

 تضيفلها

0.61

providedIn

0.60

DockStyle

0.60

Skocz

0.59

Activations Density 0.235%

personal pronouns and sentence-ending Japanese grammatical particles, as well as Java-related terms

させ

No Comments

No Known Activations