Article
🔧 L'architecture est un seuil, pas un levier — ce que j'ai appris en optimisant un LLM français de 15M de paramètres 🇫🇷
RDTvlokip
• • 1vloplok PRO
RDTvlokip
AI & ML interests
None yet
Recent Activity
posted an update 13 minutes ago
I spent a week optimizing my 15M French LLM. Not one line of new architecture. And that was the whole point.
After building it from scratch (custom crawler, BPE, LLaMA-style arch, 3-phase trainer), the model wrote perfect French but hallucinated facts and drifted off-topic. So I went hunting for the bottleneck, convinced it was the architecture.
It wasn't. It never is.
The wins came from boring places: a data pipeline that cut documents mid-sentence, two special tokens silently sabotaging generation, and one decoding hyperparameter that doubled coherence (38 → 76 tokens before drift). The flashy research, contrastive decoding, DoLa, gave the smallest gains. One of them was even a false negative caused by my own buggy eval harness.
The real lesson isn't about French LLMs:
Architecture is a threshold, not a lever. Once you clear it, the bottleneck is everywhere except the architecture. Measure first. Read your own data. Verify your code before you trust your conclusion.
The model was never the problem.
Full write-up here 👇
🔗 https://huggingface.co/blog/RDTvlokip/what-i-learned-optimizing-a-15m-frenchOrganizations
upvoted an article 27 minutes ago
upvoted an article 29 minutes ago
Article
🔧 Architecture is a threshold, not a lever — what I learned optimizing a 15M French LLM 🇫🇷
RDTvlokip
• • 1 upvoted 2
articles about 2 months ago
Article
🧠 I trained my own French LLM from scratch — alone, with a 1080 Ti, and the power went out ⚡🇫🇷
RDTvlokip
• • 6 Article
🧠 J'ai entraîné mon propre LLM français from scratch — seul, avec une 1080 Ti, et le courant a coupé ⚡🇫🇷
RDTvlokip
• • 2 upvoted 2
articles 4 months ago
Article
🧲 Embeddings — When AI turns words into GPS coordinates! 📍🧠
RDTvlokip
• • 1 Article
🧲 Embeddings — Quand l'IA transforme les mots en coordonnées GPS ! 📍🧠
RDTvlokip
• • 1 upvoted a changelog 4 months ago
Hugging Face Changelog
Public Storage Add-ons
• 168
upvoted 2
articles 4 months ago
Article
🎯 PCA (Principal Component Analysis) — Compresser les dimensions comme un boss ! 📊🔥
RDTvlokip
• • 1 Article
🎯 PCA (Principal Component Analysis) — Compressing dimensions like a boss! 📊🔥
RDTvlokip
• • 1 upvoted 4
articles 5 months ago
Article
🎯 K-Means — Quand l'IA organise le chaos en boîtes bien rangées ! 📦✨
RDTvlokip
• • 1 Article
🎯 K-Means — When AI organizes chaos into neat boxes! 📦✨
RDTvlokip
• • 1 Article
🎯 F1-Score — Quand l'Accuracy te ment en pleine face ! 📊💥
RDTvlokip
• • 1 Article
🎯 F1-Score — When Accuracy lies to your face! 📊💥
RDTvlokip
• • 1 upvoted 7
articles 6 months ago
Article
🎯 Precision & Recall — Les métriques jumelles qui ne sont jamais d'accord ! ⚖️🔍
RDTvlokip
• • 1 Article
🎯 Precision & Recall — The twin metrics that never agree! ⚖️🔍
RDTvlokip
• • 1 Article
🎆 AI 2026 — The 9 trends that will EXPLODE this year! 🚀💥
RDTvlokip
• • 2 Article
🎆 IA 2026 — Les 9 tendances qui vont exploser cette année ! 🚀💥
RDTvlokip
• • 2 Article
📊 Cross-Entropy — The loss function that KNOWS how to punish! 🎯🔥
RDTvlokip
• • 1 Article
📊 Cross-Entropy — La fonction de perte qui SAIT punir ! 🎯🔥
RDTvlokip
• • 2 Article
🎯 Learning Rate — L'accélérateur de ton réseau de neurones ! 🚗💨
RDTvlokip
• • 1