<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>My New Hugo Project</title>
    <link>https://hugo.ihyamars.my.id/</link>
    <description>Recent content from My New Hugo Project</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    
    <managingEditor>xxx@example.com (D-Sketon)</managingEditor>
    <webMaster>xxx@example.com (D-Sketon)</webMaster>
    
    <copyright>All articles on this blog are licensed under the BY-NC-SA license agreement unless otherwise stated. Please indicate the source when reprinting!</copyright>
    
    <lastBuildDate>Tue, 21 Apr 2026 13:45:00 +0700</lastBuildDate>
    
    
    <atom:link href="https://hugo.ihyamars.my.id/index.xml" rel="self" type="application/rss&#43;xml" />
    

    
    

    <item>
      <title>Detailed Data Flow Schematic: TorCons-MoE System Architecture</title>
      <link>https://hugo.ihyamars.my.id/posts/torcons-data-flow/</link>
      <pubDate>Tue, 21 Apr 2026 13:45:00 &#43;0700</pubDate>
      <author>xxx@example.com (D-Sketon)</author>
      <guid>https://hugo.ihyamars.my.id/posts/torcons-data-flow/</guid>
      <description>
        <![CDATA[<h1>Detailed Data Flow Schematic: TorCons-MoE System Architecture</h1><p>Author: D-Sketon(xxx@example.com)</p>
        
          <h1 id="detailed-data-flow-schematic---step-by-step-description">
<a class="header-anchor" href="#detailed-data-flow-schematic---step-by-step-description"></a>
DETAILED DATA FLOW SCHEMATIC - STEP-BY-STEP DESCRIPTION
</h1><h2 id="figure-2-torcons-moe-system-architecture">
<a class="header-anchor" href="#figure-2-torcons-moe-system-architecture"></a>
(Figure 2: TorCons-MoE System Architecture)
</h2><hr>
<h2 id="stage-1-input-reception-and-preprocessing">
<a class="header-anchor" href="#stage-1-input-reception-and-preprocessing"></a>
<strong>Stage 1: Input Reception and Preprocessing</strong>
</h2><p><strong>Step 1.1: Token Embedding Input</strong></p>
<ul>
<li>The system receives input token embeddings <strong>X ∈ ℝ^(T×d)</strong>, where T is the sequence length and d is the hidden dimension</li>
<li>Position indices <strong>m ∈ [0, T-1]</strong> are generated or retrieved for each token position in the sequence</li>
<li>Input data is loaded into GPU/TPU memory buffers for parallel processing</li>
</ul>
<p><strong>Step 1.2: Input Buffer Allocation</strong></p>
        
        <hr><p>Published on 2026-04-21 at <a href='https://hugo.ihyamars.my.id/'>My New Hugo Project</a>, last modified on 2026-04-21</p>]]>
      </description>
      
    </item>
    
  </channel>
</rss>
