INDEX

Explanations

specific names and references related to various topics, including blogs, jobs, and food

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

blog posts and news

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 мәкал

-0.81

 Administrativna

-0.76

󠁿

-0.68

 Савезне

-0.67

 مشين

-0.59

rungsseite

-0.59

awtextra

-0.59

endphp

-0.57

 noqa

-0.56

 jScrollPane

-0.56

POSITIVE LOGITS

 nonUne

0.69

ofthe

0.42

fromnode

0.41

mainwindow

0.40

onthe

0.38

 Montal

0.36

helloworld

0.36

CGRectMake

0.36

 Egger

0.35

 motherfucker

0.35

Activations Density 0.228%

specific names and references related to various topics, including blogs, jobs, and food

blog posts and news

No Comments

No Known Activations