<p><img alt="Flower Clouds - Odilon Redon" src="https://duarteocarmo.com/images/86/cover.webp" style="border-radius: 2px;" /></p>
<div class="toc">
<ul>
<li><a href="https://duarteocarmo.com/feed.xml#why-do-this">Why do this?</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#a-small-dataset-of-news-headlines">A small dataset of news headlines</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#gpt2-in-pytorch">GPT2 in PyTorch</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#pre-training-headlines-in-portuguese">Pre-training headlines in Portuguese</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#supervised-fine-tuning-sft-on-portuguese-from-portugal">Supervised fine-tuning (SFT) on Portuguese from Portugal</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#reinforcement-learning-grpo-for-sports-news">Reinforcement Learning (GRPO) for sports news</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#final-thoughts-acknowledgements">Final thoughts & Acknowledgements</a></li>
</ul>
</div>
<h2 id="why-do-this">Why do this?</h2>
<p>ML and AI are moving at an incredible pace. The amount of research coming out …</p>

Why do this? A small dataset of news headlines GPT2 in PyTorch Pre-training headlines in Portuguese Supervised fine-tuning (SFT) on Portuguese from Portugal Reinforcement Learning (GRPO) for sports news Final thoughts & Acknowledgements Why do this? ML and AI are moving at an incredible pace. The amount of research coming out …

<p></p>

<ul>
<li><a href="https://duarteocarmo.com/feed.xml#why-do-this" rel="noopener noreferrer">Why do this?</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#a-small-dataset-of-news-headlines" rel="noopener noreferrer">A small dataset of news headlines</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#gpt2-in-pytorch" rel="noopener noreferrer">GPT2 in PyTorch</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#pre-training-headlines-in-portuguese" rel="noopener noreferrer">Pre-training headlines in Portuguese</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#supervised-fine-tuning-sft-on-portuguese-from-portugal" rel="noopener noreferrer">Supervised fine-tuning (SFT) on Portuguese from Portugal</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#reinforcement-learning-grpo-for-sports-news" rel="noopener noreferrer">Reinforcement Learning (GRPO) for sports news</a></li>
<li><a href="https://duarteocarmo.com/feed.xml#final-thoughts-acknowledgements" rel="noopener noreferrer">Final thoughts &amp; Acknowledgements</a></li>
</ul>

Why do this?
<p>ML and AI are moving at an incredible pace. The amount of research coming out …</p>

#57 Faísca: The modern LLM stack in a single script

A podcast version of my personal website. All the episodes are generated from a text to speech model.

Share #57 Faísca: The modern LLM stack in a single script

Sign up to save your podcasts

#57 Faísca: The modern LLM stack in a single script

#57 Faísca: The modern LLM stack in a single script