INDEX

Explanations

commands or actions used in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CloseOperation

-0.90

Obrázky

-0.82

 ddelweddau

-0.79

ngdoc

-0.78

 estekak

-0.75

XtraEditors

-0.71

LEncoder

-0.71

يميديا

-0.70

лтамалар

-0.70

styleType

-0.69

POSITIVE LOGITS

 your

0.80

ulate

0.78

tivate

0.66

 yourself

0.65

your

0.63

 Protect

0.63

 Save

0.62

 Promote

0.62

vate

0.60

 Spend

0.60

Activations Density 0.147%

commands or actions used in various contexts

No Comments

No Known Activations