INDEX

Explanations

the repeated occurrence of the term "in."

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

NameInMap

-0.70

PerformLayout

-0.68

いな

-0.60

titleMargin

-0.56

In

-0.53

しゃ

-0.53

 Aidan

-0.52

MessageOf

-0.51

Diweddarwch

-0.50

 Desde

-0.50

POSITIVE LOGITS

原始内容存档于

0.67

 समीक्षक

0.65

^(@)

0.63

awsze

0.63

مراجع

0.61

=>"

0.60

"]="

0.60

 estekak

0.59

₁)

0.59

才知道

0.59

Activations Density 0.030%

the repeated occurrence of the term "in."

No Known Activations