INDEX

Explanations

references to media representations and portrayals, particularly related to LGBTQ+ themes

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

media influence and depictions

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

\{\\

-0.53

enterOuterAlt

-0.51

 propOrder

-0.50

 autorytatywna

-0.49

rrggbb

-0.49

AndEndTag

-0.49

الحياه

-0.47

 probation

-0.46

ษา

-0.45

terminated

-0.43

POSITIVE LOGITS

TV

0.47

 influencing

0.45

 television

0.44

 propaganda

0.44

 influence

0.42

tv

0.41

InjectMocks

0.41

 advertising

0.41

 images

0.40

 vettor

0.40

Activations Density 0.047%

references to media representations and portrayals, particularly related to LGBTQ+ themes

media influence and depictions

No Comments

No Known Activations