INDEX

Explanations

numerical values and contact information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 ſein

-0.88

 ſei

-0.81

ſehen

-0.73

 increí

-0.72

acterísticas

-0.71

 plufieurs

-0.70

 ſeine

-0.70

mpagne

-0.70

 ſont

-0.70

 laſſen

-0.69

POSITIVE LOGITS

Autoritní

0.40

Xna

0.40

GEBURTSDATUM

0.39

IntoConstraints

0.37

VIAF

0.35

0.32

 nakalista

0.28

MessageTagHelper

0.27

Naissance

0.27

 LoggerFactory

0.26

Activations Density 0.059%

numerical values and contact information

No Comments

No Known Activations