INDEX

Explanations

the beginning of text marked by the start-of-sequence token

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

Tokens following "Q" or a number

np_acts-logits-general · gemini-2.0-flash

gallery, team, various

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PMailer

-0.63

 محفوظة

-0.51

BoxShadow

-0.51

หมือน

-0.50

 endwhile

-0.47

 typelib

-0.47

 Safe

-0.47

 nélkül

-0.47

ClientSize

-0.46

__((

-0.46

POSITIVE LOGITS

XmlType

0.74

äsenti

0.60

Jeografia

0.57

 Anima

0.57

:✨

0.56

uremberg

0.55

wark

0.55

<?

0.54

Anima

0.54

 ferrite

0.53

Activations Density 0.043%

the beginning of text marked by the start-of-sequence token

Tokens following "Q" or a number

gallery, team, various

No Comments

No Known Activations