INDEX

Explanations

phrases indicating relationships and connections

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

of word

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.20.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

ơn

-0.59

onHide

-0.50

ctor

-0.49

/**

-0.49

 DialogInterface

-0.48

siyah

-0.47

URIComponent

-0.47

orescence

-0.47

 дни

-0.47

 olmayan

-0.46

POSITIVE LOGITS

 beginnetje

0.65

stdc

0.63

 ivelany

0.60

 Италијани

0.59

 вікі

0.57

Tikang

0.56

#+#

0.56

eBay

0.54

LookAnd

0.54

fromnode

0.53

Activations Density 0.042%

phrases indicating relationships and connections

of word

No Comments

No Known Activations