INDEX

Explanations

references to dolls and action figures

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

dolls, figures, or puppets

np_acts-logits-general · gemini-2.0-flash

doll clothes or accessories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 Wynne

-0.40

 Ripple

-0.40

 Whitfield

-0.39

Bretagne

-0.37

 Galbraith

-0.35

岐

-0.35

 Rains

-0.35

Rant

-0.35

 Grenville

-0.35

 Bowen

-0.35

POSITIVE LOGITS

 doll

1.35

 dolls

1.27

doll

1.11

 Doll

1.09

 Dolls

1.05

Doll

1.02

 boneka

0.99

 DOLL

0.94

 boneca

0.93

 puppet

0.89

Activations Density 0.253%

references to dolls and action figures

dolls, figures, or puppets

doll clothes or accessories

No Comments

No Known Activations