INDEX

Explanations

constructs related to IP addresses and their formatting

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

Negative Logits

Становништво

-0.44

inc

-0.42

 levan

-0.39

Тарихы

-0.39

چ

-0.38

ジャー

-0.38

FileType

-0.38

 iste

-0.37

ционный

-0.37

纂

-0.37

POSITIVE LOGITS

 address

1.50

 addresses

1.38

 Address

1.33

address

1.29

 Addresses

1.27

 ADDRESS

1.26

Address

1.26

 addr

1.26

 getAddress

1.19

 adresse

1.16

Activations Density 0.474%

constructs related to IP addresses and their formatting

No Comments

No Known Activations