INDEX

Explanations

programming language syntax elements and structure

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

return*

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 estekak

-0.56

DockStyle

-0.56

/*

-0.52

MLLoader

-0.51

//

-0.50



-0.50

 Administrativna

-0.47

 незавершена

-0.46

CheckBreak

-0.46

POSITIVE LOGITS

ceğine

0.47

transQ

0.46

 declarat

0.40

amodel

0.40

 aVar

0.39

ensos

0.38

 newArray

0.36

 Superan

0.36

 equivale

0.36

那些

0.35

Activations Density 0.002%

programming language syntax elements and structure

return*

No Comments

No Known Activations