INDEX

Explanations

references to emptiness or voids in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 AssemblyCompany

-0.63

 RELATIONSHIP

-0.54

relationship

-0.54

XmlAccessorType

-0.52

 Inscrivez

-0.50

chufe

-0.50

+#+

-0.50

 GenerationType

-0.49

 yür

-0.49

 Relationship

-0.49

POSITIVE LOGITS

empty

0.82

isEmpty

0.82

 Empty

0.80

Empty

0.79

IsNullOrEmpty

0.75

 empty

0.73

EMPTY

0.70

 EMPTY

0.68

OrWhiteSpace

0.68

 isEmpty

0.68

Activations Density 0.469%

references to emptiness or voids in various contexts

No Comments

No Known Activations