INDEX

Explanations

terms related to technology and academic contributions

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

scientific and technical terms

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

atalos

-0.38

 though

-0.36

 sick

-0.33

metr

-0.33

 take

-0.33

 comp

-0.33

 UIKit

-0.31

 pescoço

-0.30

chemise

-0.30

Pin

-0.30

POSITIVE LOGITS

 تضيفلها

0.80

ValueStyle

0.74

出版年

0.65

 שוליים

0.60

SBATCH

0.57

CloseOperation

0.57

IVEREF

0.57

RegistryLite

0.57

 disambiguazione

0.56

 ujednoznacz

0.56

Activations Density 0.035%

terms related to technology and academic contributions

scientific and technical terms

No Comments

No Known Activations