INDEX

Explanations

specific names, titles, or identifiers within text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Gegend

-0.55

 propOrder

-0.53

 ProtoMessage

-0.52

 disambiguazione

-0.48

fjspx

-0.48

下载附件

-0.46

 raça

-0.46

BeginInit

-0.46

DispatchToProps

-0.45

 zaken

-0.45

POSITIVE LOGITS

aarrggbb

0.50

 surla

0.44

Subviews

0.39

Wild

0.39

minente

0.39

ACHUSET

0.38

 typelib

0.38

Spelling

0.38

Side

0.38

atsun

0.37

Activations Density 1.328%

specific names, titles, or identifiers within text

No Comments

No Known Activations