INDEX

Explanations

references to specific documents or online content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 jutaan

-0.53

.*")]

-0.48

GEBURTSDATUM

-0.46

 Infór

-0.45

RectangleBorder

-0.44

 słucha

-0.44

colgante

-0.43

 fakty

-0.43

 pengelola

-0.43

EndInit

-0.43

POSITIVE LOGITS

exitRule

0.49

Chham

0.45

expandindo

0.45

jsxFileName

0.43

VersionUID

0.41

ANTES

0.39

 PhpStorm

0.38

Jeografia

0.38

 ProtoMessage

0.38

 message

0.36

Activations Density 0.556%

references to specific documents or online content

No Comments

No Known Activations