INDEX

Explanations

attends to the norm-related mathematical tokens from the corresponding norm notation tokens

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

0:0.11

1:0.14

2:0.11

3:0.11

4:0.10

5:0.08

6:0.13

7:0.19

Negative Logits

herty

-0.27

XmlEnum

-0.26

DockStyle

-0.24

RegressionTest

-0.24

chal

-0.24

BeginInit

-0.24

 säll

-0.23

XmlAccessorType

-0.23

instead

-0.23

열

-0.23

POSITIVE LOGITS

 समीक्षक

0.33

Personensuche

0.29

apunov

0.27

 Autorizaciones

0.27

Baptiste

0.26

guitarra

0.26

pulseira

0.26

 Settembre

0.26

 zvuky

0.25

 quæ

0.25

Activations Density 0.635%

attends to the norm-related mathematical tokens from the corresponding norm notation tokens

No Comments

No Known Activations