INDEX

Explanations

blocks of code or structured code elements

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

programming keywords: public private static var int

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

ValueStyle

-0.97

 estekak

-0.86

Życiorys

-0.68

 autorytatywna

-0.67

EndInit

-0.63

SourceChecksum

-0.63

исленность

-0.62

parametrize

-0.60

endpush

-0.59

 utafitiHapana

-0.59

POSITIVE LOGITS

abstractmethod

0.39

κε

0.36

 private

0.36

Autowired

0.36

static

0.35

unk

0.35

private

0.35

public

0.34

Activations Density 0.011%

blocks of code or structured code elements

programming keywords: public private static var int

No Comments

No Known Activations