INDEX

Explanations

references to favorite or preferred items and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

 Schmidt

-0.44

 ↓,

-0.41

 oral

-0.39

 Cart

-0.39

isNaN

-0.39

tab

-0.38

 Committee

-0.38

 padx

-0.37

cmath

-0.37

bed

-0.37

POSITIVE LOGITS

 favorites

0.84

 favorite

0.81

 favourites

0.78

 faves

0.69

 favourite

0.66

 treasure

0.63

+#+#

0.62

 favori

0.60

 seniors

0.56

 Geſ

0.55

Activations Density 0.523%

references to favorite or preferred items and experiences

No Comments

No Known Activations