INDEX

Explanations

references to prior experience or actions taken by the user

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

have followed by a verb

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

It detects first-person self-references or statements about the speaker (tokens like "I", "have", "I'm", "don't").

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

:✨

-0.48

KURZBESCHREIBUNG

-0.48

 defaultstate

-0.47

Personensuche

-0.47

 Superan

-0.45

Autoritní

-0.44

 Administrativna

-0.41

utilisons

-0.40

 Italijanski

-0.40

 bancaire

-0.40

POSITIVE LOGITS

urgis

0.45

rosin

0.41

ACTED

0.41

MemoryWarning

0.40

Legenda

0.40

vegas

0.39

urgeon

0.37

engeti

0.37

istocene

0.37

 समीक्षक

0.36

Activations Density 0.133%

references to prior experience or actions taken by the user

have followed by a verb

It detects first-person self-references or statements about the speaker (tokens like "I", "have", "I'm", "don't").

No Comments

No Known Activations

references to prior experience or actions taken by the user

have followed by a verb

It detects first-person self-references or statements about the speaker (tokens like "I", "have", "I'm", "don't").

No Comments

No Known Activations