INDEX

Explanations

quantitative expressions and calculations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

 tumba

-0.50

 Jerusalén

-0.47

 Jérusalem

-0.46

 Италијани

-0.45

 dioses

-0.45

 desay

-0.44

 Metzger

-0.43

 Gerechtigkeit

-0.43

 AssemblyCulture

-0.43

 remotely

-0.43

POSITIVE LOGITS

 utafitiHapana

0.60

 jspb

0.48

PerformLayout

0.44

 للمعارف

0.43

 EconPapers

0.42

 autorytatywna

0.42

 Amigo

0.42

EndContext

0.41

"}")

0.40

POI

0.40

Activations Density 0.200%

quantitative expressions and calculations

No Comments

No Known Activations

quantitative expressions and calculations

No Comments

No Known Activations