INDEX

Explanations

quantifiers and references to groups or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

GEBURTSDATUM

-0.92

setVerticalGroup

-0.81

 kasarigan

-0.74

клопе

-0.73

ChildScrollView

-0.72

BeginContext

-0.68

 мәкал

-0.66

StructEnd

-0.64

 AssemblyVersion

-0.63

WebElementEntity

-0.62

POSITIVE LOGITS

Others

0.49

 emperador

0.48

 Others

0.47

 antaranya

0.47

 extranjero

0.46

 Económica

0.44

 Kerja

0.44

 Mereka

0.44

 ninguno

0.44

 diantaranya

0.42

Activations Density 0.399%

quantifiers and references to groups or classifications

No Comments

No Known Activations