INDEX

Explanations

HTML metadata related to character encoding

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

charset=

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

 InputDecoration

-0.74

mybatisplus

-0.66



-0.65

UserScript

-0.64

 BeautifulSoup

-0.64

 للمعارف

-0.62

 JpaRepository

-0.62

reactstrap

-0.59

GHIJKLM

-0.59

 utafitiHapana

-0.58

POSITIVE LOGITS

 ciudadana

0.31

 (...)

0.30

 साम

0.30

 citoy

0.29

 naciones

0.28

 Nationen

0.28

 mezclar

0.28

 chill

0.28

 légales

0.27

izarse

0.27

Activations Density 0.001%

HTML metadata related to character encoding

charset=

No Comments

No Known Activations