INDEX

Explanations

phrases that describe names, designations, or terms associated with objects or concepts

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

"Also known as" pattern

np_acts-logits-general · gemini-2.0-flash

known as

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 purpoſe

-0.55

cipline

-0.54

 Commodore

-0.52

MemberId

-0.52

omitempty

-0.52

 ✭✭

-0.52

 EconPapers

-0.52

 ſta

-0.51

 Jefus

-0.51

 ſche

-0.51

POSITIVE LOGITS

name

0.48

 nazy

0.47

 linkovi

0.45

 name

0.44

 nicknames

0.44

 names

0.42

gilan

0.42

Personendaten

0.42

 náz

0.41

 nickname

0.39

Activations Density 0.254%

phrases that describe names, designations, or terms associated with objects or concepts

"Also known as" pattern

known as

No Comments

No Known Activations