INDEX

Explanations

the word "unique" and its variations in different contexts

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

"unique"

np_acts-logits-general · gemini-2.0-flash

unique

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

 AppBar

-0.48

 ioutil

-0.45

 insuffisamment

-0.44

 Fernsehen

-0.42

ConstraintMaker

-0.42

 salaire

-0.42

 hígado

-0.41

Cyfeiriadau

-0.41

 algemeen

-0.41

 Infór

-0.40

POSITIVE LOGITS

 unique

1.12

 UNIQUE

1.06

 Unique

1.04

Unique

1.02

unique

1.00

 uniquely

1.00

 uniqueness

0.99

UNIQUE

0.97

 uniques

0.94

 uniq

0.89

Activations Density 0.075%

the word "unique" and its variations in different contexts

"unique"

unique

No Comments

No Known Activations