INDEX

Explanations

references to geographical locations and associated entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

WebElementEntity

-0.46

canestro

-0.45

EndContext

-0.43

Toponymie

-0.40

uesia

-0.38

 DbType

-0.34

 العاده

-0.32

 الرخصة

-0.31

 Dostupné

-0.31

kuji

-0.30

POSITIVE LOGITS

 Maldives

1.02

Mal

0.71

 Male

0.68

 MALE

0.68

Ato

0.67

Male

0.65

Mal

0.65

ato

0.65

 Abdulla

0.63

mal

0.60

Activations Density 0.107%

references to geographical locations and associated entities

No Comments

No Known Activations