INDEX

Explanations

phrases focusing on personal information and self-disclosure

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

about certain topics

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

Jereo

-0.38

辦法

-0.33

Propri

-0.32

 geomé

-0.31

 graphics

-0.30

Lalu

-0.29

ließlich

-0.29

Pes

-0.28

 provid

-0.28

CURS

-0.28

POSITIVE LOGITS



0.71

aktery

0.60

Portail

0.60

GroupLayout

0.60

 AssemblyTitle

0.60

verifyException

0.59

ModelAdmin

0.59

 about

0.59

rungsseite

0.59

styleable

0.55

Activations Density 0.214%

phrases focusing on personal information and self-disclosure

about certain topics

No Comments

No Known Activations