INDEX

Explanations

references to LGBTQ+ topics, particularly regarding media representation and gender-neutral facilities

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

gender identity and sexuality

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

awaiter

-0.43

TargetException

-0.39

expandindo

-0.38

Numerology

-0.37

InstanceState

-0.36

polazione

-0.36

 nouve

-0.36

 ayer

-0.35

 Nauk

-0.35

neck

-0.34

POSITIVE LOGITS

 autorytatywna

0.54

 EconPapers

0.52

:✨

0.52

tonsoft

0.49

 Wicidata

0.48

ⓧ

0.44

 transgender

0.43

ulongan

0.43

 transférez

0.42

 disambiguazione

0.42

Activations Density 0.014%

references to LGBTQ+ topics, particularly regarding media representation and gender-neutral facilities

gender identity and sexuality

No Comments

No Known Activations