INDEX

Explanations

specific products and their features or specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 40-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_40/width_16k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.40.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

 Nachfrage

-0.46

 Sünde

-0.46

 Aufwand

-0.43

 peores

-0.42

 Tatsache

-0.42

 Zustimmung

-0.41

 Anordnung

-0.40

 Umständen

-0.40

 Voraussetzungen

-0.40

 Herkunft

-0.40

POSITIVE LOGITS

 kayaks

1.03

 subwoofer

1.02

 bikes

1.00

 kayak

0.98

 bike

0.95

 headlights

0.94

 binoculars

0.94

 skis

0.94

 headphones

0.93

 headlight

0.93

Activations Density 1.732%

specific products and their features or specifications

No Comments

No Known Activations