INDEX

Explanations

numeric representations and indicators of change or transition

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

non-English words and phrases

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

 χρή

-0.46

 supérieures

-0.44

vī

-0.43

 cowok

-0.43

 человеком

-0.43

춰

-0.42

 superiori

-0.42

MemoryWarning

-0.41

ดัง

-0.41

 purpoſe

-0.40

POSITIVE LOGITS

:✨

0.84

 ویکی‌پدی

0.74

 Мексичка

0.73

 ProtoMessage

0.70

 мәкал

0.69

 disambiguazione

0.69

setVerticalGroup

0.68

 otomatig

0.66

 Италијани

0.64

RegistryLite

0.64

Activations Density 0.022%

numeric representations and indicators of change or transition

non-English words and phrases

No Comments

No Known Activations