INDEX

Explanations

structured data format indicators, likely related to markup or programming elements

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

addPreferredGap

-0.58

CardBody

-0.53

endphp

-0.49

 JpaRepository

-0.49

 Económica

-0.48

guarded

-0.47

HandlerContext

-0.46

 createState

-0.46

呢

-0.45

жи

-0.45

POSITIVE LOGITS

 pinulongan

0.87

Personendaten

0.77



0.71

adpleegd

0.66

ArrowToggle

0.65

parsedMessage

0.64

 SwitchCompat

0.63

 estimés

0.62

ieteur

0.62

 newOwner

0.62

Activations Density 0.249%

structured data format indicators, likely related to markup or programming elements

No Comments

No Known Activations