INDEX

Explanations

references to auxiliary components or structures in computational 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

Negative Logits

 mourut

-0.77

 déclar

-0.74

aé

-0.72

 leaſt

-0.72

Eſ

-0.72

 greateſt

-0.72

 Monfieur

-0.71

 vérit

-0.71

ratislava

-0.71

 arrêté

-0.70

POSITIVE LOGITS

Aux

1.38

AUX

1.32

aux

1.29

aux

1.26

AUX

1.23

Aux

1.22

Auxiliary

1.18

 auxiliary

1.14

 Auxiliary

1.13

 auxili

1.07

Activations Density 0.010%

references to auxiliary components or structures in computational contexts

No Known Activations