INDEX

Explanations

instances of the word "to" indicating intent or purpose

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 ſind

-1.02

 AssemblyCompany

-0.79

XmlAccessorType

-0.76

wiliwch

-0.75

iſchen

-0.73

iſen

-0.73

ésultats

-0.71

 témoig

-0.69

bcryptjs

-0.69

iſten

-0.68

POSITIVE LOGITS

<bos>

0.41

↵↵

0.39

0.35

0.34

</h2>

0.32

↵

0.31

 kalan

0.30

 Inhalte

0.29

 שלא

0.29

0.28

Activations Density 0.247%

instances of the word "to" indicating intent or purpose

No Comments

No Known Activations