INDEX

Explanations

references to integrated systems or circuits

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

integrated circuits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 يتيمه

-0.80

OGND

-0.72

PerformLayout

-0.65

Хьажоргаш

-0.63

 estimés

-0.62

 CreateTagHelper

-0.58

RectangleBorder

-0.56

rungsseite

-0.56

CloseOperation

-0.54

AddHtmlAttribute

-0.54

POSITIVE LOGITS

 management

0.63

віт

0.54

Hentet

0.54

management

0.54

diğim

0.52

 multi

0.52

 circuits

0.52

achal

0.50

Management

0.50

 Management

0.50

Activations Density 0.021%

references to integrated systems or circuits

integrated circuits

No Comments

No Known Activations