INDEX

Explanations

complex numerical data or specific configurations related to technology or systems

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

non-english words

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

RenderAtEndOf

-1.04

elemField

-0.77

+#+#

-0.77

featureID

-0.73

 Administrativna

-0.73

 betweenstory

-0.71

pecabe

-0.70

twimg

-0.70

parsedMessage

-0.69

writeFieldEnd

-0.68

POSITIVE LOGITS

 могу

0.31

 Jego

0.30

ább

0.29

OrNull

0.28

WEB

0.28

 Incluso

0.26

 Stille

0.26

モール

0.25

erdings

0.24

 можна

0.24

Activations Density 0.081%

complex numerical data or specific configurations related to technology or systems

non-english words

No Comments

No Known Activations