18 21 96

Stefano Fiorucci

anakin87

https://stefano-fiorucci.netlify.app

AI & ML interests

Contributing to Haystack, the LLM Framework 🏗️. NLP / LLMs.

Articles

Selective fine-tuning of Language Models with Spectrum

17 days ago

• 25

Organizations

Posts 7

Post

999

𝐌𝐲 𝐟𝐢𝐫𝐬𝐭 𝐜𝐨𝐦𝐦𝐮𝐧𝐢𝐭𝐲 𝐚𝐫𝐭𝐢𝐜𝐥𝐞! 𝐒𝐞𝐥𝐞𝐜𝐭𝐢𝐯𝐞 𝐟𝐢𝐧𝐞-𝐭𝐮𝐧𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝐒𝐩𝐞𝐜𝐭𝐫𝐮𝐦 🎯

Full walkthrough on how to get started with Spectrum and TRL for efficient fine-tuning.
📔 👣 https://huggingface.co./blog/anakin87/spectrum

---

Looking to fine-tune Language Models efficiently and save on computational resources?

One popular method is QLoRa, which quantizes the original model and trains low-rank adapters on top.
It's quite effective and uses less GPU than full fine-tuning.

However, QLoRa applies Low-Rank Adaptation uniformly across the entire model.

What if we could identify the most informative layers and only fine-tune those? 🤔

This is exactly what Spectrum does! 👇

🔬 Spectrum analyzes the weight matrices for all layers in a Language Model and calculates a Signal to Noise Ratio (SNR) for each one.
(It uses Random Matrix Theory and Marchenko-Pastur distribution to distinguish signal from noise.)

🎯 Based on a chosen percentage (say, 25%), Spectrum selects the most informative layers of each type (mlp.down_proj, self_attn.o_proj, etc.).

You can then ❄️ freeze the rest of the model and focus your 🏋️‍♂️ training on the chosen layers.

🏆 Results/Evaluation
- Spectrum is competitive with full fine-tuning and beats QLoRA on benchmarks.
- While QLoRA is more memory-efficient on a single GPU, Spectrum shines in distributed training setups.
- Great models trained with Spectrum: Dolphin models, Llama 3.1 Storm, numerous models by VAGO Solutions...

---

For a practical guide, check out the article above.

Post

1575

💬 🇮🇹 Phi 3.5 mini ITA: a Small Language Model for Italian

Lately, I've spent some time fine-tuning language models.

Now I am happy to release Phi 3.5 mini ITA: a fine-tuned version of Phi-3.5-mini-instruct to improve performance on the Italian language

🔹 Small (3.82 B parameters) but capable model
🔹 128k context length

Chat with it on 🤗 Spaces: anakin87/Phi-3.5-mini-ITA
Model card: anakin87/Phi-3.5-mini-ITA

🗃️ Data
Supervised fine-tuning using a good mix of English and Italian data:
- mlabonne/FineTome-100k by @mlabonne
- efederici/capybara-claude-15k-ita by @efederici
🙏 Thanks to the authors for the datasets.

🎯 Targeted training with Spectrum
I used Spectrum, a relatively new technique for parameter-efficient learning.
The idea is to train only the layers of the model with high Signal-to-Noise Ratio (SNR) and ❄️ freeze the rest.
I trained the top 30% of model layers.

📝 Spectrum paper: https://arxiv.org/abs/2406.06623

📊 Vibe check and performance on Italian benchmarks seem encouraging

View all posts