INDEX

Explanations

references to news and journalism

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ylde

-0.48

itec

-0.46

bondale

-0.43

Anhalt

-0.41

 Corbu

-0.40

 CGRectMake

-0.40

 Lambeth

-0.40

 scallop

-0.39

 MATE

-0.39

CBE

-0.37

POSITIVE LOGITS

 news

1.39

 NEWS

1.27

 News

1.26

News

1.15

news

1.14

 noticia

1.05

NEWS

1.04

 noticias

0.97

 notícias

0.90

 notícia

0.89

Activations Density 0.055%

references to news and journalism

No Comments

No Known Activations