INDEX

Explanations

references to advertisements and their characteristics

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

advertisement commercials

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

cleo

-0.44

 transférez

-0.41

 ISNI

-0.41

 sağ

-0.39

 AssemblyCulture

-0.39

rrggbb

-0.37

ilität

-0.36

 мәкал

-0.36

 referenties

-0.36

Down

-0.35

POSITIVE LOGITS

 advertisement

1.02

 advertisements

0.97

 Advertisement

0.92

 Advertisements

0.91

 adverts

0.88

ads

0.80

 iklan

0.80

 advertising

0.79

 advert

0.77

 commercials

0.75

Activations Density 0.010%

references to advertisements and their characteristics

advertisement commercials

No Comments

No Known Activations