INDEX

Explanations

structured data elements and configuration details in a code or data format

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

Dojo configuration

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

Démographie

-0.57

 continúas

-0.48

 protoimpl

-0.48

DebuggerNonUser

-0.47

BeforeAll

-0.46

 Савезне

-0.46

rawDesc

-0.45

 cliquez

-0.44

الدراسه

-0.44

 Armenia

-0.44

POSITIVE LOGITS

liferay

0.71

 SharePoint

0.60

vaadin

0.45

Bq

0.41

Drupal

0.41

 Drupal

0.40

 cerâmica

0.39

ranet

0.39

Expedia

0.38

Activations Density 0.008%

structured data elements and configuration details in a code or data format

Dojo configuration

No Comments

No Known Activations