INDEX

Explanations

attends to various technical or academic terms prefixed by specific keywords from diverse contexts

oai_attention-head · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_0/width_16k/average_l0_104

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

0:0.01

1:0.04

2:0.57

3:0.14

4:0.04

5:0.05

6:0.05

7:0.06

Negative Logits

findpost

-0.31

parseFrom

-0.30

 للمعارف

-0.30

featureID

-0.28

principalColumn

-0.28

 Kind

-0.28

marle

-0.27

ように

-0.26

dox

-0.26

 isInitialized

-0.26

POSITIVE LOGITS

UnknownFields

0.28

 propTypes

0.25

AxisAlignment

0.24

wpdb

0.24

 coreana

0.24

AutoresizingMask

0.23

 projectId

0.23

Referências

0.23

 refirió

0.23

 pymongo

0.22

Activations Density 2.126%

attends to various technical or academic terms prefixed by specific keywords from diverse contexts

No Comments

No Known Activations