INDEX

Explanations

phrases indicating changes or comparisons in health and scientific results

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

comparisons of good and bad

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

énario

-0.58

queryInterface

-0.57

 Autorisations

-0.56

تقاوى

-0.55

➯

-0.55

 gyhoeddwyd

-0.54

 referenties

-0.53

 препратки

-0.53

featureID

-0.50

 فريبيس

-0.50

POSITIVE LOGITS

反而

0.41

 justru

0.37

 tenang

0.35

 דו

0.34

RectangleBorder

0.34

 hujan

0.34

HttpDelete

0.32

 oddly

0.31

 उत्त

0.31

 paradox

0.30

Activations Density 0.242%

phrases indicating changes or comparisons in health and scientific results

comparisons of good and bad

No Comments

No Known Activations