INDEX

Explanations

titles and names associated with popular media and entertainment

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

proper nouns and abbreviations

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

RefNanny

-0.49

 employés

-0.41

 specific

-0.38

预防

-0.38

onymie

-0.37

 current

-0.37

 cumplido

-0.37

 estran

-0.36

 профилакти

-0.36

embolso

-0.36

POSITIVE LOGITS

bootstrapcdn

0.54

enumii

0.53

#+#

0.52

OGND

0.48

rrggbb

0.48

تقاوى

0.46

onAttach

0.43

Skocz

0.42

Бахар

0.42

printStackTrace

0.41

Activations Density 0.010%

titles and names associated with popular media and entertainment

proper nouns and abbreviations

No Comments

No Known Activations