INDEX

Explanations

titles and author names in scientific documents

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

lages

-0.49

alable

-0.48

 udaler

-0.47

ัล

-0.47

Uris

-0.47

tighet

-0.46

 sahip

-0.45

sią

-0.45

 fede

-0.44

descend

-0.44

POSITIVE LOGITS

出版年

0.94

تقاوى

0.82

 متعلقه

0.75

PreExecute

0.73

PhysRev

0.69

 snippetHide

0.66

 trajets

0.65

رشف

0.64

expandindo

0.63

Activations Density 0.004%

titles and author names in scientific documents

No Comments

No Known Activations