INDEX

Explanations

attends to specific identifiable tokens preceded by tokens that define properties or roles

oai_attention-head · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.attn.hook_z

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.50

1:0.11

2:0.06

3:0.06

4:0.04

5:0.05

6:0.04

7:0.10

Negative Logits

sightly

-0.35

pendente

-0.33

úgó

-0.32

 szív

-0.31

isional

-0.29

 marins

-0.29

PRESA

-0.29

usercontent

-0.28

 groote

-0.28

DotNetBar

-0.27

POSITIVE LOGITS

parsedMessage

0.38

 intptr

0.35

wpi

0.34

Bibliograf

0.33

 useRouter

0.32

Хьажоргаш

0.32

PhysRev

0.32

 noDo

0.31

 تضيفلها

0.31

 Walkover

0.31

Activations Density 0.258%

attends to specific identifiable tokens preceded by tokens that define properties or roles

No Comments

No Known Activations