INDEX

Explanations

possessive pronouns and related terms indicating ownership or connection

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

possessive pronouns in multiple languages

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

 honneur

-0.44

jalá

-0.42

 pieniądze

-0.42

 tillegg

-0.40

 varandra

-0.39

 Städte

-0.39

 Dingen

-0.39

 kvinnor

-0.38

 kohta

-0.37

 näm

-0.37

POSITIVE LOGITS

own

1.11

niſſe

0.78

⤹

0.77

自己的

0.76

Your

0.76

my

0.75

 अपना

0.75

Own

0.74

 deſſen

0.73

 your

0.73

Activations Density 0.004%

possessive pronouns and related terms indicating ownership or connection

possessive pronouns in multiple languages

No Comments

No Known Activations