INDEX

Explanations

terms and elements related to data, specifically in statistical, scientific, or technical contexts

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

 McIntyre

-0.34

Referencie

-0.34

 inter

-0.32

Reparto

-0.32

Officer

-0.32

enterOuterAlt

-0.32

vician

-0.32

а

-0.31

Sub

-0.31

craina

-0.31

POSITIVE LOGITS

GEBURTSDATUM

0.59

principalColumn

0.50

 للاسماء

0.49

InjectAttribute

0.47

ⓧ

0.46

 canst

0.45

OGND

0.44

Хьажоргаш

0.44

 linkovi

0.43

saraba

0.43

Activations Density 1.192%

terms and elements related to data, specifically in statistical, scientific, or technical contexts

No Comments

No Known Activations