INDEX

Explanations

phrases that imply uncertainty or conditionality in statements

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

followed by a sentence continuation (e.g. and/but/which)

np_acts-logits-general · gemini-2.0-flash

affirmation or declaration stating content

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ngdoc

-0.59

Efq

-0.55

 Reſ

-0.50

 Theſe

-0.50

 Jefus

-0.49

 vulgaires

-0.49

 Anſ

-0.48

 Beſ

-0.47

 Arkivert

-0.46

--)

-0.46

POSITIVE LOGITS

 indeed

0.83

Indeed

0.73

IsContent

0.71

確實

0.70

 inderdaad

0.69

 Indeed

0.68

 effectivement

0.67

确实

0.65

的确

0.64

did

0.62

Activations Density 0.315%

phrases that imply uncertainty or conditionality in statements

followed by a sentence continuation (e.g. and/but/which)

affirmation or declaration stating content

No Comments

No Known Activations