INDEX

Explanations

attends to size-related tokens from corresponding tokens that indicate nullability

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

1:0.14

2:0.09

3:0.05

4:0.23

5:0.27

6:0.04

7:0.08

Negative Logits

tagHelperRunner

-0.33

NameInMap

-0.31

 estekak

-0.30

󠁢

-0.30

HasColumnType

-0.28

IsMutable

-0.28

Personendaten

-0.28

ScopeManager

-0.27

 defaultstate

-0.27

AddTagHelper

-0.26

POSITIVE LOGITS

Efq

0.48

 Monfieur

0.47

 ſeveral

0.47

 myſelf

0.46

 itſelf

0.46

 Majefty

0.46

 Reſ

0.42

 Jefus

0.42

 Theſe

0.40

 reaſon

0.40

Activations Density 0.144%

attends to size-related tokens from corresponding tokens that indicate nullability

No Comments

No Known Activations