INDEX

Explanations

references to software development projects and tools

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

San Jose conference attendance

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

ģ

-0.38

 recognized

-0.37

 successfully

-0.37

 escote

-0.37

 reconnu

-0.36

RefNanny

-0.36

 zapew

-0.35

velocityY

-0.35

 Isaacs

-0.35

 tryck

-0.35

POSITIVE LOGITS

AnchorStyles

0.49

 Spectre

0.48

 okuyayım

0.48

UserScript

0.48

Бележки

0.47

—

0.46

dflare

0.45

 increí

0.43

‽

0.43

agrama

0.42

Activations Density 0.203%

references to software development projects and tools

San Jose conference attendance

No Comments

No Known Activations