INDEX

Explanations

terms that indicate naming, labeling, or describing something

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

örté

-0.53

inerja

-0.48

AntiForgeryToken

-0.47

addGap

-0.45

{}'.

-0.45

getParameters

-0.44

 сал

-0.42

avimas

-0.41

 gắng

-0.41

bių

-0.41

POSITIVE LOGITS

 called

1.34

called

1.24

 CALLED

1.16

 Called

1.10

 termed

1.09

Called

1.09

 llamado

1.09

叫做

1.05

 sogenannten

0.97

 llamados

0.97

Activations Density 1.592%

terms that indicate naming, labeling, or describing something

No Comments

No Known Activations