INDEX

Explanations

references to specific domains or areas of focus in various aspects of life and their interconnections

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

domains, aspects, and areas

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

 React

-0.32

Ms

-0.32

 публика

-0.30

vlo

-0.30

 Withers

-0.30

 pretzels

-0.30

媒体

-0.29

Bes

-0.29

 Carl

-0.28

 izle

-0.28

POSITIVE LOGITS

setVerticalGroup

0.78

 disambiguazione

0.78

 Paglinawan

0.68

ValueStyle

0.67

 Roskov

0.64

sizeCache

0.62

Tembelea

0.62

MLLoader

0.62

 Bereiche

0.60

featureID

0.60

Activations Density 0.053%

references to specific domains or areas of focus in various aspects of life and their interconnections

domains, aspects, and areas

No Comments

No Known Activations