INDEX

Explanations

proper nouns, specifically names and titles related to characters and actors in films

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

actor names in cast lists

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

verwijspagina

-0.57

 transfieras

-0.54

 resourceCulture

-0.54

存于互联网档案馆

-0.51

 forbindelse

-0.50

makeConstraints

-0.50

ɵɵelementEnd

-0.49

 TRABAJO

-0.48

 plads

-0.48

AndEndTag

-0.48

POSITIVE LOGITS

 actor

0.42

 cast

0.41

ellido

0.38

 stars

0.38

 szín

0.36

 portraying

0.36

 actors

0.36

 plays

0.36

 aktor

0.35

 comedic

0.34

Activations Density 0.016%

proper nouns, specifically names and titles related to characters and actors in films

actor names in cast lists

No Comments

No Known Activations