INDEX

Explanations

different forms and uses of the word "notice."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 Chad

-0.81

ništvo

-0.76

Mac

-0.76

 Schulte

-0.71

-0.68

 Capo

-0.67

Bag

-0.67

 McFarland

-0.65

 leksikon

-0.64

ividual

-0.64

POSITIVE LOGITS

 NOTICE

1.32

NOTICE

1.26

 Notice

1.23

Notice

1.16

notice

1.13

 Notices

1.13

 notice

1.10

 notices

1.05

Notices

1.04

notices

1.00

Activations Density 0.123%

different forms and uses of the word "notice."

No Comments

No Known Activations