Luca Modica

lucamodica

·

AI & ML interests

None yet

Recent Activity

reacted to SeaWolf-AI's post with 🔥 13 days ago

A small gift for anyone building or studying foundation models. Most "open" models hand you the weights and stop there. With Aether-7B-5Attn we wanted to hand over the whole thing — so you can actually learn from it, reproduce it, and build on it: the data recipe, the training code, every hyperparameter, the complete logs, and the intermediate checkpoints. All Apache-2.0, reproducible byte-for-byte. What you can do with it: 🔁 Rebuild it from scratch, or fork the recipe for your own model 🔬 Study a real heterogeneous-attention MoE — 49 layers place 5 attention mechanisms on a 7×7 Latin square, arranged as a clean, attributable ablation 📈 Trace training dynamics across the released checkpoints (110k / 115k / 162k) It's a modest 6.59B model, and an honest one — the limitations (no KV-cache in this build, small scale) are written right in the card. We're not claiming it's special. If any piece of it saves you time or teaches you something, that's exactly what we hoped for. 🤗 📖 Full write-up → [blog] · https://huggingface.co/blog/FINAL-Bench/opensource-llm 📦 5 Attention Base · https://huggingface.co/FINAL-Bench/Aether-7B-5Attn 🎯 5 Attention Instruct · https://huggingface.co/FINAL-Bench/Aether-7B-5Attn-it 🚀 5 Attention Live demo · https://huggingface.co/spaces/FINAL-Bench/Aether-Sovereign-AI 📦 7 Attention Base · https://huggingface.co/FINAL-Bench/Aether-7B-7Attn-base 📦 11 Attention Base · https://huggingface.co/FINAL-Bench/Aether-6B-11Attn-base 🧬 Collection · https://huggingface.co/collections/FINAL-Bench/aether-foundation-model #opensource #LLM #MoE #reproducibility #Apache2

upvoted a collection 16 days ago

upvoted a paper 23 days ago

Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation

View all activity

Organizations

upvoted a collection 16 days ago

Alpamayo

A collection related to the Alpamayo ecosystem, containing Reasoning VLA models, Physical AI data, simulation frameworks, training utilities, and more • 5 items • Updated 16 days ago • 24

upvoted a paper 23 days ago

Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation

Paper • 2607.07608 • Published 26 days ago • 57

upvoted a paper 2 months ago

Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals

Paper • 2605.26045 • Published May 25 • 12

upvoted 2 papers 3 months ago

Geometric Factual Recall in Transformers

Paper • 2605.12426 • Published May 12 • 1

3D-VCD: Hallucination Mitigation in 3D-LLM Embodied Agents through Visual Contrastive Decoding

Paper • 2604.08645 • Published Apr 9 • 2

upvoted a collection about 1 year ago

Lyra-Model

7 items • Updated Dec 22, 2024 • 3