INDEX

Explanations

positive depictions of humor in social interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

IVEREF

-0.52

لينكات

-0.51

addContainerGap

-0.43

 propOrder

-0.40

хьтан

-0.38

 pref

-0.37

setVerticalGroup

-0.36

 خارجية

-0.36

netbeans

-0.35

StartTag

-0.35

POSITIVE LOGITS

 humorous

1.34

 humor

1.27

 laugh

1.25

 hilarious

1.24

 laughter

1.20

 comedic

1.20

 humour

1.19

 Humor

1.17

 funny

1.15

 laughs

1.15

Activations Density 0.418%

positive depictions of humor in social interactions

No Comments

No Known Activations