INDEX

Explanations

significant occurrences and nuances in discussions about social structures and events

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

more importantly

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

XmlAccessType

-0.64

lalom

-0.52

 ligiloj

-0.51

typen

-0.50

hul

-0.49

ſammen

-0.49

 Turf

-0.48

㤵

-0.48

ulties

-0.48

urate

-0.48

POSITIVE LOGITS

Portail

0.39

VersionUID

0.38

featureID

0.38

何より

0.37

पया

0.37

 autorytatywna

0.34

Produzione

0.34

 poffible

0.34

autorest

0.33

artifactId

0.31

Activations Density 0.024%

significant occurrences and nuances in discussions about social structures and events

more importantly

No Comments

No Known Activations