INDEX

Explanations

patterns related to warnings or alerts in a procedural context

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

References, URLs, or citations

np_acts-logits-general · gemini-2.0-flash

research papers and citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

UVWXYZ

-0.58

 VIAF

-0.58

DropTable

-0.53

Hauptartikel

-0.53

ernacle

-0.50

 accès

-0.50

ulets

-0.48

Diweddarwch

-0.48

HasForeignKey

-0.48

UTERS

-0.47

POSITIVE LOGITS

Zunanje

0.60

OOTDTY

0.58

 Numerade

0.55

 Waray

0.52

 SEDS

0.51

 SimpleName

0.51

Ӕ

0.51

Дереккөздер

0.49

Välislingid

0.48

yyr

0.48

Activations Density 0.276%

patterns related to warnings or alerts in a procedural context

References, URLs, or citations

research papers and citations

No Comments

No Known Activations