INDEX

Explanations

attends to tokens that represent changing boolean states from specific syntactic functions including method calls

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

0:0.09

1:0.12

2:0.15

3:0.13

4:0.10

5:0.08

6:0.14

7:0.15

Negative Logits

GeoNames

-0.25

出版年

-0.24

alini

-0.23

 Fairview

-0.23

 moderni

-0.23

 apoder

-0.22

 Monfieur

-0.22

 wrought

-0.22

 sculpt

-0.21

lastic

-0.21

POSITIVE LOGITS

PointerException

0.33

Referencie

0.32

currentColor

0.30

CRIPTION

0.29

BagConstraints

0.29

Erreferentziak

0.28

 onCancelled

0.28

=$?

0.28

sbereich

0.28

+:+

0.27

Activations Density 0.003%

attends to tokens that represent changing boolean states from specific syntactic functions including method calls

No Comments

No Known Activations