INDEX

Explanations

structured information about individuals, particularly in terms of demographic descriptions and details

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ViewFeatures

-0.68

المشاركات

-0.63

 Exactos

-0.59

 natale

-0.59

rrggbb

-0.57

 vulgaires

-0.57

Leaders

-0.54

 tuturor

-0.54

 vacuna

-0.54

rozum

-0.53

POSITIVE LOGITS

 woman

1.79

guy

1.59

man

1.48

 girl

1.47

 lady

1.46

 young

1.37

 female

1.20

woman

1.19

boy

1.18

 gentleman

1.17

Activations Density 0.413%

structured information about individuals, particularly in terms of demographic descriptions and details

No Comments

No Known Activations