INDEX

Explanations

references to programming libraries and frameworks

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

various programming libraries and frameworks

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

 Infór

-0.52

 Ahnung

-0.49

 Autorin

-0.48

 Sünde

-0.41

 hebat

-0.40

الدراسه

-0.39

Välislingid

-0.38

 Bildschirm

-0.38

 Comprometido

-0.38

berdayakan

-0.37

POSITIVE LOGITS

RenderAtEndOf

0.65

 operation

0.52

 appointment

0.50

OND

0.50

Lon

0.49

DeleteBehavior

0.48

壹

0.48

oke

0.48

Mox

0.48

Kariera

0.47

Activations Density 0.188%

references to programming libraries and frameworks

various programming libraries and frameworks

No Comments

No Known Activations