INDEX

Explanations

references to health-related risks and conditions

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

Preceding words of length four or more

np_acts-logits-general · gemini-2.0-flash

academic topics and current events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 ویکی‌پدیا

-0.65

Πηγές

-0.62

 NSCoder

-0.59

اریخ

-0.59

 lenker

-0.54

endpush

-0.53

UpInside

-0.53

</tfoot>

-0.51

MessageOf

-0.50

 محفوظة

-0.50

POSITIVE LOGITS

 reaſon

0.76

 purpoſe

0.74

 miſ

0.74

 myſelf

0.72

 Inſ

0.71

 Conſ

0.70

 ſen

0.70

 ſmall

0.69

 houſe

0.69

 ſol

0.69

Activations Density 0.093%

references to health-related risks and conditions

Preceding words of length four or more

academic topics and current events

No Comments

No Known Activations