INDEX

Explanations

words denoting profession and origin, especially academic and political

oai_token-act-pair · gemini-2.0-flash

People and professions

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_19/width_16k/average_l0_12

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PMailer

-0.70

ContentAsync

-0.53

because

-0.52

 مشين

-0.52

msgSender

-0.48

rawDesc

-0.48

 because

-0.47

MOD

-0.47

since

-0.46

Mod

-0.44

POSITIVE LOGITS

 himſelf

0.62

 vérit

0.62

 vroeger

0.61

 devenus

0.61

 folosit

0.61

 peindre

0.60

 quæ

0.60

 pintadas

0.59

 potrivit

0.59

 artesanales

0.59

Activations Density 1.930%

words denoting profession and origin, especially academic and political

People and professions

No Comments

No Known Activations