INDEX

Explanations

instances of the word "to" and its variations, particularly in contexts indicating intention or direction

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

preceding "to be" or "to [verb]"

np_acts-logits-general · gemini-2.0-flash

to be

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 للمعارف

-0.66

 تعدى

-0.65

ⓧ

-0.65

WebControls

-0.60

าสตร์

-0.54

intios

-0.54

 članak

-0.51

endphp

-0.50

<?

-0.50

évaluateur

-0.50

POSITIVE LOGITS

parsedMessage

0.83

EDEFAULT

0.66

Expedia

0.62

 banget

0.61

///</

0.59

 جدًا

0.59

싶

0.59

gered

0.58

 geworden

0.58

mainen

0.57

Activations Density 0.635%

instances of the word "to" and its variations, particularly in contexts indicating intention or direction

preceding "to be" or "to [verb]"

to be

No Comments

No Known Activations