INDEX

Explanations

references to "home" and related terms

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

"Home" (often capitalized)

np_acts-logits-general · gemini-2.0-flash

home followed by another 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-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 considérons

-0.61

 שוליים

-0.57

verständlich

-0.55

Skocz

-0.52

 gezet

-0.50

stretchr

-0.48

 visualiser

-0.47

 Cahill

-0.47

wanne

-0.47

 peristiwa

-0.46

POSITIVE LOGITS

 Home

0.89

Home

0.81

hom

0.78

HOM

0.76

 homer

0.75

 home

0.74

 Heim

0.74

Hom

0.70

 HOME

0.70

Hom

0.70

Activations Density 0.063%

references to "home" and related terms

"Home" (often capitalized)

home followed by another word

No Comments

No Known Activations