INDEX

Explanations

punctuation marks, specifically commas

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

 financières

-0.53

 razón

-0.48

 forno

-0.45

baiki

-0.45

_,

-0.45

 pouvaient

-0.44

 énergé

-0.44

</u>

-0.44

siapkan

-0.44

 juguetes

-0.44

POSITIVE LOGITS

Filmografie

1.16

 تضيفلها

0.90

+#+

0.82

 ویکی‌پدیا

0.82

DockStyle

0.81



0.80

ГЛА

0.79

Personensuche

0.76

참고

0.75

 الحره

0.73

Activations Density 0.001%

punctuation marks, specifically commas

No Comments

No Known Activations