INDEX

Explanations

phrases related to medical and scientific processes or conditions

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

 &___

-0.48

}{*}{}

-0.41

 dezelve

-0.40

 arbet

-0.36

 zelve

-0.34

 eenig

-0.33

 odborn

-0.32

 tiegħ

-0.32

 kepad

-0.32

 oglą

-0.32

POSITIVE LOGITS

 Administrativna

0.79

#+#

0.69

writeFieldEnd

0.68

GEBURTSDATUM

0.63

 kaarangay

0.59

Билгалдахарш

0.59

contentLoaded

0.59

 informée

0.57

 ivelany

0.56

LayoutStyle

0.56

Activations Density 1.908%

phrases related to medical and scientific processes or conditions

No Comments

No Known Activations