INDEX

Explanations

references to popular culture, specifically movies and television

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

places and addresses

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 désolés

-0.42

 समीक्षाओं

-0.41

Erstellt

-0.41

šinou

-0.37

Билгалдахарш

-0.35

 CreateTagHelper

-0.34

 للمعارف

-0.31

iramente

-0.31

 Einsch

-0.30

 connaître

-0.30

POSITIVE LOGITS

ViewImports

0.61

mybatisplus

0.60

ChildScrollView

0.59

\{\\

0.55

 SwitchCompat

0.54

windowFixed

0.54

 mijne

0.54

featureID

0.53

estacks

0.50

ⓧ

0.50

Activations Density 0.024%

references to popular culture, specifically movies and television

places and addresses

No Comments

No Known Activations