INDEX

Explanations

attends to a specific token from flexible context tokens marked in various formats

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

0:0.09

1:0.08

2:0.07

3:0.08

4:0.14

5:0.03

6:0.08

7:0.39

Negative Logits

 distanciation

-0.31

 tema

-0.27

pearing

-0.27

 MessageBoxIcon

-0.26

Wet

-0.24

 desper

-0.24

olverine

-0.23

NOPQRST

-0.23

ieważ

-0.23

óta

-0.23

POSITIVE LOGITS

sizeCache

0.42

Autoritní

0.34

}');

0.33

 viewDidLoad

0.30

OGND

0.29

anking

0.29

 Paglinawan

0.29

 transfieras

0.28

 canadensis

0.28

 referenties

0.28

Activations Density 1.278%

attends to a specific token from flexible context tokens marked in various formats

No Comments

No Known Activations