INDEX

Explanations

medical terms and acronyms related to diseases and healthcare

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

HostException

-0.78

awaiter

-0.74

ChildScrollView

-0.72

GEBURTSDATUM

-0.71

 ffilmiau

-0.71

RTLI

-0.70

DockStyle

-0.69

AnchorStyles

-0.68

 surla

-0.66

 CURIAM

-0.65

POSITIVE LOGITS

 femininas

0.34

 secco

0.33

 femininos

0.33

 scatola

0.31

 obdob

0.29

 verdade

0.29

wuchs

0.28

 typique

0.27

ForUser

0.27

폭

0.26

Activations Density 0.130%

medical terms and acronyms related to diseases and healthcare

No Comments

No Known Activations