INDEX

Explanations

references to humor and comedic elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

神宮

-0.58

}{|

-0.54

deepEqual

-0.50

atsby

-0.49

 kör

-0.48

GMENT

-0.48

த்

-0.47

спол

-0.47

 küpe

-0.46

первых

-0.45

POSITIVE LOGITS

 joke

2.05

 jokes

2.03

 humor

1.99

 humour

1.92

 Humor

1.92

 comedy

1.90

humor

1.85

 comedic

1.83

 funny

1.82

 Joke

1.80

Activations Density 0.131%

references to humor and comedic elements

No Comments

No Known Activations