INDEX

Explanations

elements indicating the publication and details of articles

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

Preceding words to articles or viewpoints

np_acts-logits-general · gemini-2.0-flash

article on following texts

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

queryInterface

-0.39

AccessorTable

-0.38

GOTREF

-0.36

toBeDefined

-0.35

 JpaRepository

-0.34

])).

-0.33

titleMargin

-0.32

ifikationer

-0.32

 echi

-0.32

hus

-0.31

POSITIVE LOGITS

oredCriteria

0.59

Населення

0.58

parsedMessage

0.57

HomeAsUpEnabled

0.56



0.55

 initComponents

0.54

ISODE

0.54

XmlType

0.52

matchCondition

0.52

 Article

0.51

Activations Density 0.449%

elements indicating the publication and details of articles

Preceding words to articles or viewpoints

article on following texts

No Comments

No Known Activations