INDEX

Explanations

questions and expressions of inquiry

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

Following certain punctuation or sentence starters

np_acts-logits-general · gemini-2.0-flash

technical terms and phrases

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 estancias

-0.35

 surface

-0.32

 capacities

-0.32

 phase

-0.32

Özel

-0.32

 superfície

-0.31

 aprendiz

-0.31

 canale

-0.31

 sobą

-0.31

 protectora

-0.30

POSITIVE LOGITS

 okuyayım

0.68

migrationBuilder

0.57

Données

0.56

 <>",

0.55

 Савезне

0.54

 autorytatywna

0.52

 Roskov

0.51

Diwedd

0.51

ollectionView

0.50

SerializedName

0.50

Activations Density 0.125%

questions and expressions of inquiry

Following certain punctuation or sentence starters

technical terms and phrases

No Comments

No Known Activations