INDEX

Explanations

legal terminology and references that denote subsequent citations or designations

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

hereinafter definitions

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

orology

-0.55

 Tosh

-0.54

pology

-0.54

 Biôgrafia

-0.52

Msp

-0.50

 AttributeSet

-0.50

ⓧ

-0.50

SBATCH

-0.48

cetamol

-0.48

 etsy

-0.47

POSITIVE LOGITS

Higgins

0.68

 Higgins

0.67

 Heer

0.55

قایناقلار

0.54

 PeEnEo

0.52

!!

0.51

 Sheehan

0.51

 albicans

0.50

 Medan

0.50

 Haller

0.49

Activations Density 0.009%

legal terminology and references that denote subsequent citations or designations

hereinafter definitions

No Comments

No Known Activations