INDEX

Explanations

conjunctions, particularly the word "and."

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

Autoritní

-0.61

 autorytatywna

-0.52

 Mixed

-0.49

siąż

-0.49

FunctionFlags

-0.48

Mixed

-0.47

angor

-0.47

 mixed

-0.46

Havolalar

-0.46

 }}"></

-0.45

POSITIVE LOGITS

 StatefulWidget

0.86

 BoxFit

0.71

 OnInit

0.69

 StatelessWidget

0.69

StructEnd

0.68

endpush

0.67

íncia

0.65

 المعيارى

0.65

 jsPsych

0.63

-------------</

0.61

Activations Density 0.010%

conjunctions, particularly the word "and."

No Known Activations