INDEX

Explanations

structured data elements or formats

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

empêcher

-0.32

ByExample

-0.32

ząt

-0.29

 volta

-0.28

 razon

-0.28

 shifts

-0.28

 değiş

-0.28

 prestation

-0.27

ForResult

-0.27

eğ

-0.27

POSITIVE LOGITS

 autorytatywna

0.73

parsedMessage

0.66

featureID

0.64

Jeografia

0.63

 informée

0.63

FieldBuilder

0.63

Билгалдахарш

0.61

oredCriteria

0.60

SourceChecksum

0.60

AccessorTable

0.59

Activations Density 0.221%

structured data elements or formats

No Comments

No Known Activations