INDEX

Explanations

phrases indicating conditions or restrictions in logical or mathematical contexts

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

Tokens preceding mathematical or symbolic expressions

np_acts-logits-general · gemini-2.0-flash

to scrutinize

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

 ModelExpression

-0.54

 intptr

-0.53

ंदीखरीदारी

-0.50

 transférez

-0.48

RegistryLite

-0.48

 صوتيه

-0.47

saraba

-0.47

 informée

-0.47

XmlAccessType

-0.46

 tartalomajánló

-0.46

POSITIVE LOGITS

 Moll

0.45

 répondu

0.43

aat

0.43

화

0.42

tec

0.42

BIN

0.42

ชี

0.41

 scrap

0.40

yat

0.40

avad

0.40

Activations Density 1.302%

phrases indicating conditions or restrictions in logical or mathematical contexts

Tokens preceding mathematical or symbolic expressions

to scrutinize

No Comments

No Known Activations