INDEX

Explanations

specific terms and phrases related to labeling and denoting information

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

\{\\

-0.56

 impact

-0.48

Go

-0.47

May

-0.47

جوايز

-0.47

 bill

-0.46

imp

-0.45

bill

-0.44

ழி

-0.44

Be

-0.44

POSITIVE LOGITS

FunctionFlags

0.70

Personensuche

0.68

principalColumn

0.68

latego

0.65

 pleaſure

0.64

 spéciaux

0.63

Efq

0.62

 defStyleAttr

0.61

featureID

0.60

 Meksi

0.59

Activations Density 0.273%

specific terms and phrases related to labeling and denoting information

No Comments

No Known Activations