INDEX

Explanations

instances of the word "do" and variations thereof in various contexts

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ChildScrollView

-0.87

 Jann

-0.81

Ise

-0.79

stateProvider

-0.76

 istanbul

-0.75

Portail

-0.74

ttino

-0.74

</tfoot>

-0.73

 <<<<<<<<<<<<<<

-0.72

 Nere

-0.72

POSITIVE LOGITS

 done

1.49

do

1.38

Do

1.27

 does

1.24

 DONE

1.23

Do

1.18

done

1.14

 Does

1.09

DO

1.07

did

1.06

Activations Density 1.384%

instances of the word "do" and variations thereof in various contexts

No Comments

No Known Activations