INDEX

Explanations

attends to references to information or instructions from tokens that provide additional context or examples

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

0:0.28

1:0.28

2:0.09

3:0.04

4:0.03

5:0.05

6:0.04

7:0.14

Negative Logits

ipts

-0.26

Tembelea

-0.25

AddHtmlAttribute

-0.25

 ddelweddau

-0.24

 propOrder

-0.24

AntiForgeryToken

-0.24

 kasarigan

-0.24

 RouterModule

-0.23

 onOptions

-0.23

êque

-0.23

POSITIVE LOGITS

ंदीखरीदारी

0.33

enumii

0.32

UnusedPrivate

0.32

SourceChecksum

0.31

webElementXpaths

0.31

 reafon

0.29

IonicModule

0.28

 caufe

0.28

}-${

0.28

Personendaten

0.28

Activations Density 0.227%

attends to references to information or instructions from tokens that provide additional context or examples

No Comments

No Known Activations