INDEX

Explanations

attends to location-related tokens from generic tokens that provide details about those locations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.07

1:0.10

2:0.10

3:0.12

4:0.12

5:0.06

6:0.22

7:0.18

Negative Logits

])));

-0.26

 muualla

-0.25

charlotte

-0.24

 становника

-0.24

 fär

-0.23

enti

-0.23

WebElementEntity

-0.23

 Nearly

-0.23

 \%$\\

-0.22

LOTTE

-0.22

POSITIVE LOGITS

Personendaten

0.38

 Rollen

0.30

reactstrap

0.29

 NUKAT

0.28

ագրություններ

0.28

HtmlAttribute

0.27

JspWriter

0.27

multer

0.27

nextProps

0.27

atm

0.26

Activations Density 0.021%

attends to location-related tokens from generic tokens that provide details about those locations

No Comments

No Known Activations