INDEX

Explanations

sections related to authorship, contributions, and conflict of interest disclosures

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

disclosure of conflicts of interest

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

Personendaten

-0.57

AndEndTag

-0.56

商品説明

-0.48

kháu

-0.48

شهاد

-0.42

 Вікіпе

-0.41

SPJ

-0.41

lait

-0.41

الحياه

-0.41

Vra

-0.40

POSITIVE LOGITS

 vectorielle

0.52

 gynhyrchwyd

0.45

 huelga

0.45

osoba

0.44

kwds

0.44

InputBorder

0.43

 humanidad

0.42

 humedad

0.42

 materiál

0.41

 prisonniers

0.41

Activations Density 0.466%

sections related to authorship, contributions, and conflict of interest disclosures

disclosure of conflicts of interest

No Comments

No Known Activations