INDEX

Explanations

references to popular culture and media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 Publikum

-0.51

 Producción

-0.46

 bolsillo

-0.45

 protectora

-0.43

 hembra

-0.43

 bordada

-0.42

 cortesía

-0.42

 redonda

-0.41

 voluntarios

-0.41

 rodillas

-0.41

POSITIVE LOGITS

 Shrek

0.96

 Avengers

0.95

 Jurassic

0.92

 Seinfeld

0.91

 Batman

0.91

 Transformers

0.91

 Naruto

0.90

 Pokemon

0.88

 SpongeBob

0.88

 Spiderman

0.86

Activations Density 0.483%

references to popular culture and media

No Comments

No Known Activations