INDEX

Explanations

references to Australia

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 Spill

-0.71

</b>

-0.68

="#

-0.67

 Erie

-0.66

 Knicks

-0.66

 Berks

-0.65

Spill

-0.64

</i>

-0.64

endregion

-0.63

accio

-0.61

POSITIVE LOGITS

 Australia

1.52

 Australian

1.40

Australia

1.40

 Australians

1.34

 australian

1.30

ALIAN

1.27

 AUSTRALIA

1.25

Australie

1.24

 australia

1.23

 Aussie

1.22

Activations Density 0.095%

references to Australia

No Comments

No Known Activations