INDEX

Explanations

numerical data and specific counts or frequencies within texts

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

parsedMessage

-0.46

berdayakan

-0.45

featureID

-0.43

 كومونز

-0.43

Ỳ

-0.41

Diweddarwch

-0.40

Indented

-0.39

 zapato

-0.38



-0.38

printStackTrace

-0.38

POSITIVE LOGITS

 democracies

0.40

енча

0.40

velocityX

0.39

 mouseClicked

0.39

hug

0.39

charCodeAt

0.39

Portale

0.37

EDEFAULT

0.37

ègre

0.37

 <<<<<<<<<<<<<<

0.36

Activations Density 0.491%

numerical data and specific counts or frequencies within texts

No Comments

No Known Activations

numerical data and specific counts or frequencies within texts

No Comments

No Known Activations