INDEX

Explanations

symbols or indicators of formatting or special emphasis in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 useDispatch

-0.68

 Majefty

-0.65

Hochspringen

-0.64

TextWatcher

-0.62

didSet

-0.60

 tslint

-0.58

 فريبيس

-0.58

 Houſe

-0.57

 textStatus

-0.57

 houſe

-0.57

POSITIVE LOGITS

 للاسماء

0.86

 تكبرها

0.83

 tartalomajánló

0.66

aarrggbb

0.62

 дописавши

0.62

ThemeOverlay

0.60

ंदीखरीदारी

0.58

 kasarigan

0.55

ViewFeatures

0.55

messageInfo

0.54

Activations Density 0.564%

symbols or indicators of formatting or special emphasis in text

No Comments

No Known Activations