INDEX

Explanations

phrases indicating the start of a new section or topic

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

出版年

-0.48

lapsingToolbar

-0.40

 defaultstate

-0.40

homonymie

-0.38

 समीक्षक

-0.37

 nahilalakip

-0.35

IntoConstraints

-0.35

 linkovi

-0.35

TypedDataSet

-0.34

Externé

-0.34

POSITIVE LOGITS

 дописавши

0.50

 gesetzlichen

0.49

CPtr

0.49

tanleria

0.48

 Савезне

0.48

beri

0.48

multer

0.47

kit

0.47

 mocht

0.47

 Risiken

0.46

Activations Density 0.108%

phrases indicating the start of a new section or topic

No Comments

No Known Activations