INDEX

Explanations

terms related to methods of improvement and applications in medical and industrial contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

improving industrial processes or identifying things

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

Negative Logits

setVerticalGroup

-0.43

PropertyChanging

-0.37

 Olsson

-0.33

Heraus

-0.33

 procès

-0.32

SYSC

-0.32

 circonstances

-0.32

MLLoader

-0.31

 Publikum

-0.30

ఔ

-0.30

POSITIVE LOGITS

 autorytatywna

0.72

OGND

0.64

ſicht

0.58

Personensuche

0.57

ロウィン

0.53

存于互联网档案馆

0.52

 fashiola

0.51

UnusedPrivate

0.50

AndEndTag

0.49

wiliwch

0.49

Activations Density 0.024%

terms related to methods of improvement and applications in medical and industrial contexts

improving industrial processes or identifying things

No Known Activations