INDEX

Explanations

HTML and UI component structures

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

HTML tags and code snippets

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

 للاسماء

-0.90

 Мексичка

-0.73

 NSCoder

-0.72

 pinulongan

-0.72

 Normdatei

-0.72

 transférez

-0.71

 يتيمه

-0.70

 ujednoznacz

-0.67

NameInMap

-0.66

Diweddarwch

-0.64

POSITIVE LOGITS

dagangan

0.32

dente

0.32

 lisesi

0.31

Cancellation

0.31

CANCE

0.31

 komt

0.31

komt

0.31

 Bidang

0.30

 Utama

0.30

 Masyarakat

0.30

Activations Density 0.006%

HTML and UI component structures

HTML tags and code snippets

No Comments

No Known Activations