INDEX

Explanations

instructions and methodologies related to processes or systems

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

 المعيارى

-0.42

 EconPapers

-0.41

:✨

-0.37

nezeu

-0.36

siębior

-0.34

SceneManagement

-0.34

Browsable

-0.34

걸

-0.34

Personensuche

-0.33

pinMode

-0.33

POSITIVE LOGITS

évaluateur

0.57

 Biôgrafia

0.56

TypedDataSet

0.56

Diweddarwch

0.52

 OMITBAD

0.52

parsedMessage

0.50

 useAppContext

0.50

>';

0.50

 &___

0.50

 snippetHide

0.49

Activations Density 0.013%

instructions and methodologies related to processes or systems

No Comments

No Known Activations