INDEX

Explanations

sections related to bibliographic references and authorship in texts

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

 AssemblyCompany

-0.55

IsMutable

-0.53

 OMITBAD

-0.50

 intStringLen

-0.47

InputBorder

-0.46

ImageContext

-0.46

صحاب

-0.46

ToBounds

-0.44

 BoxFit

-0.44



-0.43

POSITIVE LOGITS

 disambiguazione

0.41

 ää

0.38

 éditoriale

0.37

 døde

0.37

værende

0.36

extends

0.36

SequentialGroup

0.35

midler

0.35

 réserver

0.35

 kasarigan

0.35

Activations Density 0.934%

sections related to bibliographic references and authorship in texts

No Comments

No Known Activations