INDEX

Explanations

references to residential living spaces, specifically apartments and flats

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 propOrder

-0.48

 Winf

-0.46

LoggerFactory

-0.42

Czas

-0.42

 Micho

-0.42

desta

-0.42

 Besch

-0.40

 menger

-0.40

 trasc

-0.40

QRST

-0.39

POSITIVE LOGITS

 apartment

0.82

 apartments

0.77

apartment

0.74

 appartement

0.73

Apartment

0.72

 Apartment

0.71

Apartments

0.70

 appartements

0.69

wohnung

0.68

 Apartments

0.66

Activations Density 0.009%

references to residential living spaces, specifically apartments and flats

No Comments

No Known Activations