INDEX

Explanations

terms related to LGBTQ+ identities and issues

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

sexual orientation and identity

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

 WaitForSeconds

-0.41

 Grain

-0.41

 recalls

-0.40

efforts

-0.39

Grain

-0.39

 Coppola

-0.38

 deliberation

-0.37

 recalled

-0.37

 granular

-0.36

颦

-0.35

POSITIVE LOGITS

gay

0.93

 homosexual

0.92

 homosexuality

0.90

 lesbian

0.88

 homophobic

0.86

gay

0.83

 gays

0.81

 lesbians

0.80

 sexuality

0.80

 LGBTQ

0.80

Activations Density 0.031%

terms related to LGBTQ+ identities and issues

sexual orientation and identity

No Comments

No Known Activations