INDEX

Explanations

references to the concept of "the people" or societal segments like the middle class and wealthy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

WriteTagHelper

-0.68

Tembelea

-0.53

 فريبيس

-0.53

aarrggbb

-0.51

цездатний

-0.49

Personensuche

-0.48

 MainAxisSize

-0.48

ThemeOverlay

-0.47

Viitteet

-0.47

 CallOverrides

-0.47

POSITIVE LOGITS

 masses

0.91

 populace

0.85

 population

0.71

 electorate

0.67

 people

0.66

 citizenry

0.66

 audience

0.65

masses

0.61

 bevol

0.60

 nation

0.59

Activations Density 0.508%

references to the concept of "the people" or societal segments like the middle class and wealthy

No Comments

No Known Activations