INDEX

Explanations

directives or instructions often related to actions or guidance

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

"to" followed by a verb

np_acts-logits-general · gemini-2.0-flash

go to destinations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 itſelf

-0.76

 שוליים

-0.72

 Florentine

-0.67

 Estonian

-0.65

 ejus

-0.65

 onely

-0.64

 تضيفلها

-0.64

 Transvaal

-0.63

ISupport

-0.63

 Majefty

-0.63

POSITIVE LOGITS

the

0.86

 "/";

0.70

 سوی

0.68

بوابة

0.65

 another

0.62

bed

0.62

EndGlobalSection

0.61

_;

0.61

{};

0.61

'])){

0.61

Activations Density 0.224%

directives or instructions often related to actions or guidance

"to" followed by a verb

go to destinations

No Comments

No Known Activations