INDEX

Explanations

descriptions of locations and access points within buildings

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

Defence

-0.63

 Australia

-0.57

 Australian

-0.56

Australia

-0.55

Australian

-0.53

 Aussie

-0.52

жте

-0.52

defence

-0.51

 defaultstate

-0.51

 customised

-0.50

POSITIVE LOGITS

 تضيفلها

0.76

abestanden

0.68

thoven

0.66

 مواليد

0.62

PhysRevD

0.61

 Freehold

0.59

 createState

0.59

MigrationBuilder

0.59

bandcamp

0.56

transQ

0.55

Activations Density 0.066%

descriptions of locations and access points within buildings

No Comments

No Known Activations