INDEX

Explanations

elements related to addresses and locations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tagHelperRunner

-0.61

 kaarangay

-0.58

 Taktlose

-0.48

IsMutable

-0.47

OGND

-0.47

\{\\

-0.46

 iNdEx

-0.44

 gyhoeddwyd

-0.44

etheless

-0.43

 <=",

-0.43

POSITIVE LOGITS

 dwellings

0.52

 houses

0.51

 Address

0.50

 endereço

0.50

 properties

0.49

 endereco

0.49

 huset

0.48

 house

0.47

 address

0.47

 ADDRESS

0.47

Activations Density 0.338%

elements related to addresses and locations

No Comments

No Known Activations