INDEX

Explanations

code or elements related to programming and libraries

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

define, macro, include

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

aarrggbb

-0.52

RenderAtEndOf

-0.35

clerview

-0.33

viaf

-0.33

 athletes

-0.33

 flat

-0.32

AsUp

-0.31

 athlete

-0.31

 atletas

-0.31

 Durée

-0.31

POSITIVE LOGITS

 Walkover

0.53

 resourceCulture

0.50

بوابة

0.46

Personensuche

0.46

 мәкал

0.45

 intptr

0.44

 CanadaChoose

0.44

 ब्रेकडाउन

0.44

yip

0.43

livery

0.43

Activations Density 0.002%

code or elements related to programming and libraries

define, macro, include

No Comments

No Known Activations