INDEX

Explanations

numbers related to specific years or dates

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

years, dates, and programming classes

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

styleType

-0.67

 ویکی‌پدی

-0.61

 AssemblyCulture

-0.54

httphttps

-0.51

 ujednoznacz

-0.51

 nahilalakip

-0.50

(!__

-0.48

@@@@@

-0.47

IContainer

-0.46

uxxxx

-0.46

POSITIVE LOGITS

ValueGeneration

0.38

createdDate

0.36

ឆ

0.36

 heen

0.35

ArgsConstructor

0.34

AccessorTable

0.34

OrderID

0.34

Herald

0.34

 olsun

0.34

色

0.33

Activations Density 0.060%

numbers related to specific years or dates

years, dates, and programming classes

No Comments

No Known Activations