INDEX

Explanations

attends to mathematical notation or equality symbols from mathematical expressions or variables

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

0:0.07

1:0.08

2:0.08

3:0.14

4:0.27

5:0.07

6:0.11

7:0.14

Negative Logits

MENAFN

-0.27

olsa

-0.24

 AppComponent

-0.23

 berätt

-0.23

 Utama

-0.23

FetchType

-0.22

ofollow

-0.22

rident

-0.22

cipl

-0.22

 Imre

-0.22

POSITIVE LOGITS

IUrlHelper

0.47

GEBURTSDATUM

0.45

انيف

0.35

 ویکی‌پدی

0.33

TintMode

0.32

AutoScaleMode

0.32

 Numerade

0.32

 betweenstory

0.31

verwijspagina

0.31

 referenties

0.31

Activations Density 0.349%

attends to mathematical notation or equality symbols from mathematical expressions or variables

No Comments

No Known Activations