INDEX

Explanations

words and phrases related to marking or categorizing items

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

Czytaj

-0.79

LEGGI

-0.76

 Wallis

-0.74

źródło

-0.74

 MacGregor

-0.73

 BoxFit

-0.72

ぎり

-0.71

bewah

-0.71

 ApiService

-0.71

</tfoot>

-0.71

POSITIVE LOGITS

Mark

0.88

 Marks

0.88

Marks

0.85

 MARKS

0.81

 Mark

0.80

 Markle

0.75

 marks

0.74

manship

0.74

 resultCode

0.74

mark

0.74

Activations Density 0.019%

words and phrases related to marking or categorizing items

No Comments

No Known Activations