INDEX

Explanations

references to the name "Martin" in various contexts, particularly related to significant figures in history

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

"Martin" followed by names or titles

np_acts-logits-general · gemini-2.0-flash

Martin followed by surname

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 autorytatywna

-0.89

 Roskov

-0.78

 Partagez

-0.72

othesis

-0.72

ollectionView

-0.71

ChildScrollView

-0.71

hdysval

-0.69

 Endlich

-0.68

 itſelf

-0.68

futbolista

-0.67

POSITIVE LOGITS

 Martins

0.85

 Martin

0.75

 Marty

0.71

Martin

0.67

 Scorsese

0.64

Marty

0.63

marty

0.62

Martina

0.61

原始内容存档于

0.61

 mart

0.60

Activations Density 0.052%

references to the name "Martin" in various contexts, particularly related to significant figures in history

"Martin" followed by names or titles

Martin followed by surname

No Comments

No Known Activations