INDEX

Explanations

various citations and references to scientific data or studies

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

foreign words and symbols

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

 Administrativna

-0.87

oredCriteria

-0.82

 kasarigan

-0.78

⤹

-0.74

UserScript

-0.70

 мәкал

-0.69

 gynhyrchwyd

-0.66

WebElementEntity

-0.66

 Италијани

-0.66

帖最后由

-0.65

POSITIVE LOGITS

 espalda

0.35

 wiederum

0.32

 also

0.32

 again

0.31

 Económica

0.29

MathML

0.28

<tbody>

0.28

 manzanas

0.28

 Ráp

0.28

jupiter

0.27

Activations Density 0.084%

various citations and references to scientific data or studies

foreign words and symbols

No Comments

No Known Activations