INDEX

Explanations

references to existing entities or claims in a context, particularly in relation to databases or applications

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

old claim

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xymatrix

-0.49

 detailsNormal

-0.49

osoba

-0.47

Климат

-0.46

HostException

-0.44

PreferredItem

-0.43

Seeker

-0.43

ValueGenerated

-0.43

 الحره

-0.42

GetEnumerator

-0.41

POSITIVE LOGITS

 AssemblyCompany

0.41

UrlResolution

0.40

 alluminio

0.37

 epidem

0.36

 autorité

0.36

liothèque

0.35

 abre

0.34

 znaj

0.34

 Frust

0.34

 brosse

0.34

Activations Density 0.161%

references to existing entities or claims in a context, particularly in relation to databases or applications

old claim

No Comments

No Known Activations

references to existing entities or claims in a context, particularly in relation to databases or applications

old claim

No Comments

No Known Activations