INDEX

Explanations

references to media sources and journalists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

RefNanny

-0.52

AntiForgeryToken

-0.48

LookAnd

-0.46

rizona

-0.41

queryInterface

-0.41

 Wikimedijinoj

-0.40

layoutControl

-0.40

AndEndTag

-0.39

fjspx

-0.38

WebServlet

-0.37

POSITIVE LOGITS

Bloomberg

0.65

 Bloomberg

0.62

bloomberg

0.59

wsj

0.56

Finance

0.52

+#+#

0.52

Билгалдахарш

0.51

 CNBC

0.49

Reuters

0.47

 bodyParser

0.47

Activations Density 0.513%

references to media sources and journalists

No Comments

No Known Activations