INDEX

Explanations

phrases indicating purpose or intention

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

phrases indicating position or time

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 these

-0.59

 этим

-0.54

 diesen

-0.51

 this

-0.48

these

-0.47

 ModelExpression

-0.47

 এই

-0.47

GeneratedValue

-0.46

 diese

-0.46

viding

-0.45

POSITIVE LOGITS

 EconPapers

0.71

 yourselves

0.71

Jereo

0.69

médiaire

0.69

ContentAsync

0.66

paž

0.65

Personendaten

0.64

SharedCtor

0.64

 aDecoder

0.63

Diweddarwch

0.63

Activations Density 0.108%

phrases indicating purpose or intention

phrases indicating position or time

No Comments

No Known Activations