INDEX

Explanations

proper nouns, particularly names of people

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 autorytatywna

-0.59

Diweddarwch

-0.53

(;;)

-0.50

__':

-0.48

ISupport

-0.48

:]:

-0.47

"}\

-0.47

@"/

-0.46

 चीज़ों

-0.46

__':

-0.45

POSITIVE LOGITS

 producteurs

0.51

 człowiek

0.51

 collaborateurs

0.50

MethodManager

0.48

 Syrie

0.48

 Romains

0.48

 Inscrivez

0.47

parsedMessage

0.47

 kimdir

0.47

 consommateurs

0.46

Activations Density 0.346%

proper nouns, particularly names of people

No Comments

No Known Activations