INDEX

Explanations

phrases related to options and selections

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

options to choose from

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

UserScript

-0.43

 mission

-0.39

principalTable

-0.36

不出

-0.36

homonymie

-0.34

mobileqq

-0.32

kadang

-0.32

 كورة

-0.32

enn

-0.32

 أنها

-0.32

POSITIVE LOGITS

oredCriteria

0.77

AndEndTag

0.67

 GenerationType

0.57

EndContext

0.54

 newOwner

0.54

:✨

0.52

 estekak

0.51

 nahilalakip

0.50

PreferredItem

0.50

 noDo

0.50

Activations Density 0.003%

phrases related to options and selections

options to choose from

No Comments

No Known Activations