INDEX

Explanations

words related to stamps and stamping in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

期刊论文

-0.59

Демографія

-0.58

umumkan

-0.54

atzung

-0.53

 autorytatywna

-0.52

лизм

-0.52

 attra

-0.50

Brend

-0.50

mybatisplus

-0.49

 Infórmanos

-0.49

POSITIVE LOGITS

 stamps

0.85

ViewFeatures

0.85

 stamp

0.72

 Stamps

0.67

(!__

0.66

tee

0.65

teenth

0.65

stam

0.65

 STAM

0.63

 /\.(

0.62

Activations Density 0.061%

words related to stamps and stamping in various contexts

No Comments

No Known Activations