INDEX

Explanations

structured data or code syntax elements

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

code identifiers and keywords

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

 again

-0.44

 also

-0.43

又有

-0.34

too

-0.34

 always

-0.33

 tiež

-0.32

又是

-0.32

 LIABLE

-0.30

led

-0.30

 alone

-0.29

POSITIVE LOGITS

 kasarigan

0.88

 autorytatywna

0.84

ImageContext

0.78

sizeCache

0.70

addCriterion

0.68

 المعيارى

0.68

 ویکی‌پدی

0.67

NameInMap

0.66

OGND

0.66

GEBURTSDATUM

0.65

Activations Density 0.083%

structured data or code syntax elements

code identifiers and keywords

No Comments

No Known Activations

structured data or code syntax elements

code identifiers and keywords

No Comments

No Known Activations