INDEX

Explanations

numerical values associated with scientific measurements or data

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

references or citations in academic articles

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

journal table citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

FromArgb

-0.66

Demografie

-0.57

/\.

-0.55

EndInit

-0.54

 internetowa

-0.50

 Perseus

-0.50

Personendaten

-0.49

তথ্যসূত্র

-0.48

AMENTE

-0.47

Nuorodos

-0.46

POSITIVE LOGITS

 surla

0.59

setHorizontal

0.56

glGen

0.56

فایل‌لار

0.55

bootstrapcdn

0.54

setVerticalGroup

0.52

ConstraintMaker

0.52

ollectionView

0.52

ViewFeatures

0.51

AccessorTable

0.50

Activations Density 0.014%

numerical values associated with scientific measurements or data

references or citations in academic articles

journal table citations

No Comments

No Known Activations